REVIEW ARTICLE
Neural Markers of Neuropathic Pain Associated with Maladaptive Plasticity in Spinal Cord Injury Paula Pascoal-Faria, PhD*,†; Nilufer Yalcin, MD†; Felipe Fregni, MD, PhD, MPH†,‡ *School of Technology and Management and Centre for Rapid and Sustainable Product Development, Polytechnic Institute of Leiria, Leiria, Portugal; †Spaulding Neuromodulation Center, Spaulding Rehabilitation Hospital, Harvard Medical School; ‡Spaulding-Harvard Spinal Cord Injury Model System, Spaulding Rehabilitation Hospital, Harvard Medical School, Charlestown, MA, U.S.A.
& Abstract Objectives: Given the potential use of neural markers for the development of novel treatments in spinal cord pain, we aimed to characterize the most effective neural markers of neuropathic pain following spinal cord injury (SCI). Methods: A systematic PubMed review was conducted, compiling studies that were published prior to April, 2014 that examined neural markers associated with neuropathic pain after SCI using electrophysiological and neuroimaging techniques. Results: We identified 6 studies: Four using electroencephalogram (EEG); 1 using magnetic resonance imaging (MRI) and FDG-PET (positron emission tomography); and 1 using MR spectroscopy. The EEG recordings suggested a reduction in alpha EEG peak frequency activity in the frontal regions of SCI patients with neuropathic pain. The MRI scans showed volume loss, primarily in the gray matter of the left dorsolateral prefrontal cortex, and by FDG-PET, hypometabolism in
the medial prefrontal cortex was observed in SCI patients with neuropathic pain compared with healthy subjects. In the MR spectroscopy findings, the presence of pain was associated with changes in the prefrontal cortex and anterior cingulate cortex. Conclusions: When analyzed together, the results of these studies seem to point out to a common marker of pain in SCI characterized by decreased cortical activity in frontal areas and possibly increased subcortical activity. These results may contribute to planning further mechanistic studies as to better understand the mechanisms by which neuropathic pain is modulated in patients with SCI as well as clinical studies investigating best responders of treatment. & Key Words: pain, spinal cord injury, neuropathic pain, neural markers, neuroplasticity
INTRODUCTION Address correspondence and reprint requests to: Felipe Fregni, MD, PhD, MPH, Spaulding Neuromodulation Center, Spaulding Rehabilitation Hospital, Harvard Medical School, 79/96 13th Street, Charlestown, MA, U.S.A. E-mail: [email protected]. P. Pascoal-Faria and N. Yalcin contributed equally. Submitted: May 9, 2014; Revision accepted: June 27, 2014 DOI. 10.1111/papr.12237
© 2014 World Institute of Pain, 1530-7085/14/$15.00 Pain Practice, Volume , Issue , 2014 –
Chronic pain is a major complication in patients following spinal cord injury (SCI), with a prevalence of 60% to 80%, and it is often refractory to medication.1–3 One-third of such patients describe their pain as severe, decreasing their quality of life.2–6 The mechanisms of pain syndromes after SCI are unknown. Initially, the area of the spinal cord near the site of the injury was considered to be where most
2
PASCOAL-FARIA ET AL.
anatomic and functional changes occurred, due to chemical synaptic alterations.3 However, recent animal studies have demonstrated broader neural areas are responsible for the pain following an SCI.7 Chronic pain in SCI patients increases the synaptic efficacy of somatosensory neurons in the dorsal horn of the spinal cord and effects dysfunction of the cortical circuits that are associated with sensory and pain processing, which is referred to as central sensitization.8–10 This phenomenon leads to an imbalance in the excitatory and inhibitory inputs and reorganizes the cortex, which involves a large neural network, including limbic structures, such as the anterior cingulate cortex, hippocampus, amygdala, and thalamic nuclei.11 Collectively, these changes constitute what is known as maladaptive plasticity. Based on the significance of central changes in the development and sustainment of chronic pain, neural markers that are associated with maladaptive plasticity must be identified to determine the mechanisms of its treatments and prognostic factors. No study has reported consistent and effective pharmacological and nonpharmacological treatments for neuropathic pain following SCI.6,12,13 Thus, the development of novel treatments for this condition remains an unmet clinical need. In this context, neural markers may be useful in clinical trials as surrogate outcomes. Therefore, electrophysiology and brain neuroimaging indexed neural markers could allow us to gain insights into the pathophysiology of neuropathic pain following SCI. In this study, we aimed to identify the most effective neural markers of neuropathic pain in SCI. We report significant regional changes at various electroencephalogram (EEG) frequencies and by magnetic resonance imaging (MRI) and FDG-PET (positron emission tomography).
OR electroencephalography OR electroencephalogram OR neurophysiological OR neurophysiologic OR neurophysiology OR signatures OR markers OR marker). Selection and Exclusion Criteria Studies were included per the following criteria: (1) published in English; (2) human subjects; (3) original research; and (4) articles on neural markers, using electrophysiological recordings or neuroimaging as an outcome measure in SCI patients with neuropathic pain. Additional exclusion criteria were: (1) studies that included surgical procedures, deep brain stimulation, or brain stem cell transplantation; (2) articles on pharmacological, biofeedback, and neuromodulation treatment techniques; and (3) articles that included only results on imaginary movement or other tasks. Data Extraction A data extraction sheet was developed to ensure the systematic collection of relevant data, adopted from the Cochrane Handbook of Systematic Reviews for Intervention Studies. All main information topics that are generally provided regarding neural markers of pain in SCI were covered. The following data were extracted from the included studies (for the full list of topics in the extraction sheet, see Table 1): name of first author; year of publication; guidelines and scales used for inclusion criteria; number of participants in each group (healthy and other cohorts) and their characteristics (gender, age); exclusion criteria; medication; neural markers that have been used and their characteristics; and significant results. Identification of Articles
METHODS Information Sources and Literature Search Studies in English that examined neural markers of neuropathic pain following SCI in human subjects were identified in PubMed; this search was performed on April 1, 2014. The following search string was used to identify these full-text articles: spinal cord injury AND (neuropathic pain OR pain OR neurogenic) AND (functional magnetic resonance imaging OR positron emission tomography OR single positron emission computed tomography OR EEG OR TMS OR tDCS or tACS OR tPCS OR transcranial OR neuronal polarization OR electrophysiology OR electrophysiologic OR electrophysiological
Two investigators (NY and PF) independently evaluated the titles and abstracts of the identified studies to exclude those that did not meet the inclusion criteria. Then, the full text of all eligible studies was retrieved and assessed for inclusion or exclusion by 2 investigators. If there was disagreement about a study’s eligibility, a consensus decision was reached by the investigators.
RESULTS Study Selection The literature search yielded 6 articles: 4 on EEG; 1 on MRI and FDG-PET; and 1 on MR spectroscopy. The
Neural Markers of Pain in Spinal Cord Injury 3
Table 1. Studies on Neural Markers of Neuropathic Pain in Spinal Cord Injury
Studies
Guidelines and scales used for assessment
Study Population
Number and Characteristics of healthy subjects (age/sex/other information)
Number and Characteristics of subjects with Neuropathic pain (NP) after SCI (age/sex/other information)
Number and Characteristics of subjects without NP after SCI (age/sex/other information)
16/Age/gender matched. Mean age 34 (10.7) 26/Age/gender matched
8/7 Male, 1 female. Mean age 35 (11.3). ASIA (4) T4 to T12 levels 14/Mean age of SCI 47 (15). AIS (A to D) C5 to T10 levels
8/8 Male. Mean age 33.5 (10.3). ASIA (4, T4 to T12 levels) 9/Age 47 (15), C5 to T10 levels
28/18 Male, 10 female. Mean age 44.57 (13.96) 10/7 Male, 3 female. Mean age 39.1 (10.1)
38/27 Male, 11 female. Mean age 51.24 (12.04)
16/15 Male, 1 female. Mean age 49 (12.82)
10/7 Male, 3 female. Mean age 45.2 (9.1). Pain level with 5 and more for at least 6 months, ASIA (A to D). VNS (≥ 5) 5/Mean age 36.4 (10.4), SCI more than 12 months. Mean duration of condition 63.6 (49.6) months since injury
10/8 Male, 2 female. Mean age 44.4 (8.1). SCI at least 1 year 5/Mean age 36.4 (10.4), SCI more than 12 months. Mean duration of condition 58.2 (59.7) months since injury
Boord et al. (2008)12
ASIA
32
Wydenkeller et al. (2009)15
AIS and 0 to 10 scale for pain intensity
49
Jensen et al. (2013)14
0 to 10 scale for pain intensity
82
Vuckovic et al. (2014)16
ASIA and VNS
30
Stanwell et al. (2010)17
ASIA
20
10/Mean age 36.4 (5)
Yoon et al. (2013)13
ASIA 11-point NRS system and BDI
20
10/Age/gender matched
Patients exclusion criteria
Patients with NP—Medication
Loss of consciousness or a period of post-traumatic amnesia following injury
Medication was kept (gabapentin, amitriptyline, mexiletine, pregabalin, morphine, temazepam, diazepam). A significant reduction in Peak y–a band frequency, as an effect of medication, was seen only at 3/14 sites but no significant effect on EC/EO reactivity Medication was kept (nonsteroidal antiinflammatory drugs, anticholinergics, antiepileptics, spasmolytics, antidepressants, opioids). No significant effect of medication
Subjects with additional spinal column lesions belowT10. Neurological history except SCI and mild depression. Subjects with NP at the level or below the level of the lesion < 18 years, do not understand English, SCI < 12 months, report pain for < 6 months, pain intensity average < 4 (0 to 10 scale). History of seizure activity, significant head injury, skull defects
Central acting medications and others were kept. No significant effect of medication
SCI < 1 year after injury with a spinal injury above T1. Presence of any chronic or acute pain at the time of the experiment, brain injury, or other know neurology disease
Medication was kept (baclofen, carbamazepine, gabapentin, pregabalin, amitriptyline, diazepam). No reference regarding the medication effect on the results
10/6 Males, 4 females. Mean age 39 (8.6). More than 6 months since SCI, stable chronic pain at least 3 months, only with neuropathic pain. ASIA (A,B) Electrophysiology or NeuroImaging Technique and its Characteristics
Results/Comments
EEG: eyes closed and open/14 channels/bandwidth of DC–500 Hz and sampled at 2,048 Hz. Peak frequency in the theta–alpha band (4 to 13 Hz)
Significantly reduced EEG peak frequency and significantly reduced EEG spectral reactivity were observed in patients with NP
EEG: 10 to 20, 30 scalp electrodes and 2 electrodes below the outer canthus of the eyes, 500 Hz with a 140-Hz low pass filter 3-minutes closed eyes, delta, alpha, beta, and theta. The pain intensity did not correlate with EEG EEG: 10 to 20, 19 electrodes, bandpass 0.3 to 70 Hz, sampled at the rate 250 Hz. Awake state and eyes closed, delta, alpha, beta, and theta. The pain intensity did not correlate with EEG
EEG peak frequency in the 6 to 12 Hz band was significantly slower in patients with bNP compared to patients without bNP. The peak and band topographies did not differ statistically between these groups. The pain intensity did not correlate with EEG
EEG: 10 to 10, 61 electrodes, using and ear linked reference and AFz ground, eyes opened and eyes closed, 2 minutes for each conditions repeated 3 times Channels sampled at 1,000 Hz
Significant effects did emerge in absolute power activity for frontal localizations (FP1, FP2, F3, FZ, and F4), with less alpha in the SCI pain group, relative to the other groups. No significant changes were seen for any electrode site for relative delta or beta. More relative theta was observed only at P3, 01, 02 in patients with pain after SCI. Less relative alpha was seen at 14 of the 19 electrodes (all except F7, T3, T6, = 1, and O2), in the SCI pain group. The pain intensity did not correlate with EEG Paraplegic patients with CNP had increased alpha power in eyes open (EO) state in most of the recording sites than patients with no pain. In EC state, there was significant increase in alpha power at only frontal regions. No difference among groups was found between theta PSDs in the eyes closed (EC) state whereas theta power only increased in frontal and occipital regions in eyes opened (EO) state. No significant difference between groups in the beta range was found
4
PASCOAL-FARIA ET AL.
Table 1. (Continued) Electrophysiology or NeuroImaging Technique and its Characteristics
Patients exclusion criteria
Patients with NP—Medication
Incomplete SCI and/or tetraplegia, history of psychopathology, previous pain syndromes prior to the injury, epilepsy, or the presence of non-MR compatible devices
Medication was kept (pregabalin, morphine, tramadol, amitriptyline). No reference regarding the medication effect on the results
MRS: TR/TE 3,000/20 ms, 2,048 FID points, bandwidth 2 kHZ, 6 cm3 voxel, Location: Left ACC, left PFC, and left thalamus
Patients with neuropsychological disease or any other systemic disease affecting the central or peripheral nervous system
No reference regarding the medication effect on the results
FDG-PET: 10 minutes emission scans. Brain MRI: 3T MRI, TR 8.1 ms, TE 4.6 ms, flip angle 8°, 175 slices, 1 mm thickness and matrix size 256 9 256. DTI: TR 7514.1 ms, TE 65.9 ms, thickness 2 mm, flip angle 100°, 75 slices, matrix size 128 9 128
Results/Comments Spectroscopy from the: (1) thalamus (significant differences for NAA, choline-containing compounds, glutamine, and glutamate), distinguishing SCI patients (with and without pain) from control; (2) prefrontal cortex (NAA, glutamate, glutamine, choline-containing compounds and taurine, GABA, glycine, and others) and anterior cingulate cortex (myoinositol, aspartate and others) distinguishing SCI with pain from patients without pain Hypometabolism in the left and right middle frontal gyrus and decrease gray matter volume was seen in the left middle frontal gyrus, bilateral anterior insula and right suggenual anterior cingulate cortex. Lower mean diffusivity (MD) was identified in deep white matter regions such as in the right internal capsule including the region from the anterior to posterior limb, cerebral peduncle, and anterior corona radiata
ASIA, American Spinal Injury Association; AIS, American Spinal Injury Association Impairment Scale; BDI, Beck Depression Inventory; EEG, electroencephalogram; MRI, magnetic resonance imaging; NRS, Numeric Rating Scale; PET, positron emission tomography; SCI, spinal cord injury; VNS, Visual Numerical Scale.
full text of these articles was reviewed and included in this review.
particularly in the frontal regions, in SCI patients with neuropathic pain.
Study Characteristics
Outcome Measures
The studies included only participants with neuropathic pain after SCI. The severity of the pain was assessed on a pain scale from 0 to 10 in a clinical interview, in which 0 was no pain and 10 was the most intense pain imaginable. In all studies, the clinical features of the neuropathic pain were examined, including the level, severity, and duration of pain after the SCI. The pooled size of the studies was 233 subjects. The sample sizes of each study ranged from 20 to 82. The EEG studies comprised 193 participants, and the neuroimaging studies (MRI, MR spectroscopy, and PET) included 40 participants. All studies allowed participants to continue their current medications. Whereas the results were unaffected by the use of medications in Boord et al., Yoon et al., Jensen et al., and Wydenkeller et al.,12–15 Vuckovic et al. did not analyze the influence of medications, and no references were cited in Stanwell et al.16,17 Table 1 summarizes the results and design of each study. The 6 studies examined neural markers of neuropathic pain after SCI by EEG, MRI, MR spectroscopy, and FDG-PET. Generally, there was a reduction in the anatomic and functional activity of the cortex,
For the EEG studies, peak frequencies and relative and absolute power activities in all frequency bands (alpha, theta, delta, beta, and gamma) were calculated with the eyes open and/or closed. The MRI studies measured volume changes in various brain locations, whereas the MR spectroscopy study focused on the metabolism in the anterior cingulated gyrus, prefrontal cortex, and thalamus. The PET study assessed changes in regional metabolic activity. Electroencephalogram Most EEG studies in patients with neuropathic pain after SCI demonstrated a reduction in EEG peak theta– alpha frequency activity with the eyes closed at most sites compared with healthy controls and SCI patients without pain. Moreover, the alpha power declined significantly, primarily in the frontal areas. However, the pain intensity and EEG activity did not correlate (Table 1). In the larger study, in which EEG was performed with the eyes closed, Jensen et al.14 found that absolute alpha power significantly decreased in frontal locations in SCI
Neural Markers of Pain in Spinal Cord Injury 5
patients with pain vs. SCI patients without pain and healthy controls. Further, less relative alpha activity was noted in the SCI pain group relative to no-pain controls at 14 of the 19 electrode sites (ie, all except F7, T3, T6, O1, and O2). With regard to theta activity, the relative power was significantly larger at P3, O1, and O2 in SCI patients with pain compared with those without pain. Similar findings were reported in Boord et al.12 and Wydenkeller et al.15 Peak theta–alpha frequency decreased at all sites with the eyes closed in SCI patients with pain vs. those without pain. Further, Boord et al.12 generated the same results in comparing SCI patients with pain and healthy controls, whereas Wydenkeller et al.15 did not find any significant differences between these groups. Vuckovic et al.16 was the only study that provided data with the eyes open, noting a significant increase in alpha power at most sites between SCI patients with pain and without pain. Also, theta power was also larger with the eyes open in the frontal and occipital regions in the pain group compared with the no-pain group. Conversely, with the eyes closed, the results were less conclusive. The alpha power increased significantly in the frontal regions in SCI patients with pain vs. those without pain, but there were no significant differences in theta band. Alpha and theta did not differ between SCI patients with pain and healthy controls. For the other EEG bandwidths (delta, beta, and gamma), these studies showed nonsignificant trends, and in 2 of the studies,14,15 mean pain intensity did not correlate with EEG power activity. Magnetic Resonance Imaging Two MRI studies examined 2 techniques: MRI volumetric analysis and MR spectroscopy. In Yoon et al.,13 by MRI, there was significant volume loss in the left middle frontal gyrus, bilateral anterior insulae, and right subgenual ACC using 3 TMRI. There were no increases in gray matter volume in any brain region in patients with neuropathic pain after SCI compared with healthy subjects. Also, there was significantly lower mean diffusivity in several deep and peripheral white matter regions. For more details see Table 1. By MR spectroscopy, the voxels in the anterior cingulated gyrus, prefrontal cortex, and thalamus were examined, demonstrating that changes in the prefrontal cortex and anterior cingulate cortex were related specifically to the presence of pain. The chief metabolites
that contributed to the presence of pain were NAA, glutamate, glutamine, choline-containing compounds, taurine, GABA, and glycine in the prefrontal cortex, and myoinositol and aspartate in the anterior cingulate cortex. In addition, NAA was significantly increased in the thalamus of SCI patients compared with healthy subjects.17 Positron Emission Tomography Metabolism in the left and right medial frontal gyrus declined in SCI patients with neuropathic pain compared with healthy controls. No region was found to have increased metabolism.13
DISCUSSION Six clinical studies that used electrophysiological and neuroimaging techniques met the inclusion criteria. Four studies used EEG, 1 used MRI and FDG-PET, and 1 performed MR spectroscopy. By EEG, patients with neuropathic pain after SCI show a reduction in EEG peak frequency activity with the eyes closed, especially in the alpha and theta bands, compared with healthy controls and SCI patients without pain.12,15 Further, in Jensen et al.,14 alpha power was also reduced in frontal areas in SCI patients with neuropathic pain with the eyes closed. Alpha reflects the level of activity in the cortex. Moreover, alterations in the dynamics of thalamocortical loops affect persistent and synchronous alpha activity.18,19 Reductions in EEG frequency are caused by deafferentation, which can be explained by thalamocortical dysrhythmia that arises from hyperpolarization of thalamic neurons.20 This dysrhythmia mediates the genesis of neuropathic pain through disinhibition and subsequent activation of networks in the cortical pain matrix.12,21 Neuropathic pain has also been linked to thalamic dysrhythmia in in vivo animal studies.22–24 Moreover, these results are consistent with other studies that showed a decrease in dominant EEG frequency toward lower frequencies in patients with various types of chronic pain.25,26 Conversely, Vuckovic et al.16 showed a significant increase in alpha band, primarily in frontal regions, in pain subjects compared with those with no pain with the eyes closed. Notably, despite these changes, there were no differences in alpha or theta band, regardless of the state of the eyes, in the pain group compared with healthy subjects. The variability between studies might
6
PASCOAL-FARIA ET AL.
be attributed to the state of the eyes during the EEG recordings. In fact, alpha band activity is associated with more variability with eyes open. Individual differences, such as the use of medications, between study subjects might also have influenced the results. In Jensen et al.,14 Boord et al.,12 and Wydenkeller et al.,15 there were no significant effects of medication use on the results. However, in Vuckovic et al.,16 although medications were continued during the recordings, their effects were not analyzed. The neuroimaging studies had findings in the same direction as the EEG studies. The MRI study by Yoon et al.13 reported volume loss in the left middle frontal gyrus, bilateral anterior insula, and right subgenual ACC and lower mean diffusivity in white matter regions, whereas FDG-PET revealed hypometabolism in the left and right medial frontal gyrus. Atrophy of the DLPFC has been associated with chronic pain conditions in several studies.27–29 This area of the brain governs the balance between the descending fascilatory systems and descending inhibitory systems, which actively control the perception of pain.11,30 Moreover, the prefrontal area is linked to cognitive and attentional processing of pain stimuli.26,27,31 The subgenual anterial cingulate gyrus (sACC) and insula have been implicated in emotional processes that mediate effective responses to pain stimuli.31,32 Further, although these regions differ in location, they overlap functionally,31 which is consistent with studies that have addressed chronic pain.28,33,34 An MR spectroscopy study by Stanwell et al.17 agrees that critical effective emotional circuits, such as the anterior cingulate cortex and insulae, change significantly between SCI patients with and without pain. Most of our findings were characterized by decreased activity in the frontal areas, as supported by the alpha rhythm dysrhythmia in the EEG studies. These studies did not examine the motor cortex. Other neuropathic syndromes are associated with decreased intracortical inhibition and increased intracortical fascilation, which suggests that M1 modulates thalamic activity directly. These results can be used to develop novel targets for the treatment of neuropathic pain in SCI. Based on the connectivity of these areas, the efficacy of treatments that target the areas that we have discussed could be measured by determining the activity in these targets. Moreno-Duarte et al.35 reported significant pain improvement only when primary motor cortex (using tDCS) and hypothalamic region (using cranial electrical stimulation) were targeted. Yet, other
studies that performed different brain stimulation techniques with disparate targets of the nervous system, such as the brain motor cortex (rTMS), hypothalamus (CES), peripheral nerves (TENS), and the dorsal horn (SCS), found less of an effect on pain or none at all. Recent advances in neuroimaging and neurophysiological methods have provided support for maladaptive plasticity that is associated with chronic pain. Although there have been few such studies exist and despite their small sample size, when analyzed together, they form the basis for a neural signature of maladaptive plasticity in SCI and chronic pain. The main finding appears to be related to distributed cortical-subcortical dysregulation, in which subcortical circuits, such as the thalamus, are overactive and emotional affective circuits, such as the sACC and insulae, are hypoactive. Most studies agree that alpha activity decreases in the frontal areas. Moreover, other cortical circuits, such as M1, are significant areas that are associated with the development and regulation of pain processing in SCI. Further studies that assess other methods, such as TMS indices of cortical excitability, might be useful.
ACKNOWLEDGEMENTS This study was supported by a grant from the Department of Education, NIDRR H133N110010 (SCI— Model Systems). Paula Pascoal-Faria acknowledges the support of the Strategic Project (PEST-OE/EME/ UI4044/2013) funded by the Portuguese Foundation for Science and Technology and also the support of the European Commission through the Marie Curie Project “International Research Exchange for Biomedical Devices Design and Prototyping” “IREBID”.
REFERENCES 1. Bonica JJ. History of pain concepts and pain therapy. Mt Sinai J Med. 1991;58:191–202. 2. Siddall PJ, Taylor DA, McClelland JM, Rutkowski SB, Cousins MJ. Pain report and the relationship of pain to physical factors in the first 6 month following spinal cord injury. Pain. 1999;81:187–197. 3. D’Angello R, Morreale A, Donadio V, et al. Neuropathic pain following spinal cord injury: what we know about mechanisms, assessment and management. Eur Rev Med Pharmacol Sci. 2013;17:3257–3261. 4. St€ ormer S, Gerner HJ, Gr€ uninger W, et al. Chronic pain/dysaesthesiae in spinal cord injury patients: results of a multicentre study. Spinal Cord. 1997;35:446–455. 5. Siddall PJ, McClelland JM, Rutkowski SB, Cousins MJ. A longitudinal study of the prevalence and characteristics
Neural Markers of Pain in Spinal Cord Injury 7
of pain in the first 5 years following spinal cord injury. Pain. 2003;103:249–257. 6. Wasner F. Central pain syndromes. Curr Pain Headache Rep. 2010;14:489–496. 7. Hulsebosch CE, Hains BC, Crown E, Carlton S. Mechanisms of chronic central neuropathic pain after spinal cord injury. Brain Res Rev. 2009;60:202–213. 8. Woolf CJ. Evidence for a central component of postinjury pain hypersensitivity. Nature. 1983;306:686–688. 9. Cook AJ, Woolf CJ, Wall PD, McMahon SB. Dynamic receptive field plasticity in rat spinal cord dorsal horn following C-primary afferent input. Nature. 1987;325:151– 153. 10. Ghosh A, Sydekum E, Haiss F, et al. Functional and anatomical reorganization of the sensory-motor cortex after incomplete spinal cord injury in adult rats. J Neurosci. 2009;29:12210–12219. 11. Kwon M, Altin M, Duenas H, Alev L. The role of descending inhibitory pathways on chronic pain modulation and clinical implications. Pain Pract. 2013; doi:10.1111/papr. 12145. 12. Boord P, Siddall PJ, Tran Y, Herbert D, Middleton J, Craig A. Electroencephalographic slowing and reduced reactivity in neuropathic pain following spinal cord injury. Spinal Cord. 2008;46:118–123. 13. Yoon EJ, Kim YK, Shin HI, Lee Y, Kim SE. Cortical and white matter alterations in patients with neuropathic pain after spinal cord injury. Brain Res. 2013;1540:64–73. 14. Jensen MP, Sherlin LH, Gertz KJ, et al. Brain EEG activity correlates of chronic pain in persons with spinal cord injury: clinical implications. Spinal Cord. 2013;51:55–58. 15. Wydenkeller S, Maurizio S, Dietz V, Halder P. Neuropathic pain in spinal cord injury: significance of clinical and electrophysiological measures. Eur J Neurosci. 2009;30:91– 99. 16. Vuckovic A, Hasan MA, Fraser M, Conway BA, Nasseroleslami B, Allan DB. Dynamic oscillatory signatures of central neuropathic pain in spinal cord injury. J Pain. 2014;15:645–655. 17. Stanwell P, Siddall P, Keshava N, et al. Neuro magnetic resonance spectroscopy using wavelet decomposition and statistical testing identifies biochemical changes in people with spinal cord injury and pain. Neuroimage. 2010;53:544– 552. 18. Llinas RR, Steriade M. Bursting of thalamic neurons and states of vigilance. J Neurophysiol. 2006;95:3297– 3308. 19. Hindriks R, van Putten MJ. Thalamo-cortical mechanisms underlying changes in amplitude and frequency of human alpha oscillations. Neuroimage. 2013;70:150–163. 20. Ching S, Cimenser A, Purdon PL, Brown EN, Kopell NJ. Thalamocortical model for a propofol-induced alpharhythm associated with loss of consciousness. Proc Natl Acad Sci USA. 2010;107:22665–22670.
21. Walton KD, Llin as RR. Central pain as a thalamocortical dysrhythmia. Chapter 13—A thalamic efference disconnection? In: Kruger L, Light AR, eds. Translational Pain Research: From Mouse to Man. Boca Raton, FL: CRC Press; 2010:1–16. 22. Llin as R, Ribary U, Jeanmonod D, et al. Thalamocortical dysrhythmia I. Functional and imaging aspects. Thalamus Relat Syst. 2002;1:237–244. 23. Gerke MB, Duggan AW, Xu L, Siddall PJ. Thalamic neuronal activity in rats with mechanical allodynia following contusive spinal cord injury. Neuroscience. 2003;117:715– 722. 24. Kim D, Park D, Choi S, et al. Thalamic control of visceral nociception mediated by T-type Ca2+ channels. Science. 2003;302:117–119. 25. Stern J, Jeanmonod D, Sarnthein J. Persistent EEG overactivation in the cortical pain matrix of neurogenic pain patients. Neuroimage. 2006;31:721–731. 26. Sarnthein J, Stern J, Aufenberg C, Rousson V, Jeanmonod D. Increased EEG power and slowed dominant frequency in patients with neurogenic pain. Brain. 2006;129:55–64. 27. Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain. 2005;9:463–484. 28. Seminowicz DA, Wideman TH, Naso L, et al. Effective treatment of chronic low back pain in humans reverses abnormal brain anatomy and function. J Neurosci. 2011;31:7540–7550. 29. Obermann M, Rodriguez-Raecke R, Naegel S, et al. Gray matter volume reduction reflects chronic pain in trigeminal neuralgia. Neuroimage. 2013;74:352–358. 30. Lorenz J, Minoshima S, Casey KL. Keeping pain out of mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain. 2003;126:1079–1091. 31. Yoon EJ, Kim YK, Kim HR, Kim SE, Lee Y, Shin HI. Transcranial direct current stimulation to lessen neuropathic pain after spinal cord injury: a mechanistic PET study. Neurorehabil Neural Repair. 2014;28:250–259. 32. Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain. 2011;152(suppl 3):S2– S15. 33. Kuchinad A, Schweinhardt P, Seminowicz DA, et al. Accelerated brain gray matter loss in fibromyalgia patients: premature aging of the brain? J Neurosci. 2007;27:4004– 4007. 34. Egloff N, Sabbioni ME, Salathe C, Wiest R, Juengling FD. Nondermatomal somatosensory deficits in patients with chronic pain disorder: clinical findings and hypometabolic pattern in FDG-PET. Pain. 2009;145:252–258. 35. Moreno-Duarte I, Morse LR, Alam M, Bikson M, Zafonte R, Fregni F. Targeted therapies using electrical and magnetic neural stimulation for the treatment of chronic pain in spinal cord injury. Neuroimage. 2014;85(Pt 3):1003–1013.
Neural Markers of Neuropathic Pain Associated with Maladaptive Plasticity in Spinal Cord Injury Paula Pascoal-Faria, PhD*,†; Nilufer Yalcin, MD†; Felipe Fregni, MD, PhD, MPH†,‡ *School of Technology and Management and Centre for Rapid and Sustainable Product Development, Polytechnic Institute of Leiria, Leiria, Portugal; †Spaulding Neuromodulation Center, Spaulding Rehabilitation Hospital, Harvard Medical School; ‡Spaulding-Harvard Spinal Cord Injury Model System, Spaulding Rehabilitation Hospital, Harvard Medical School, Charlestown, MA, U.S.A.
& Abstract Objectives: Given the potential use of neural markers for the development of novel treatments in spinal cord pain, we aimed to characterize the most effective neural markers of neuropathic pain following spinal cord injury (SCI). Methods: A systematic PubMed review was conducted, compiling studies that were published prior to April, 2014 that examined neural markers associated with neuropathic pain after SCI using electrophysiological and neuroimaging techniques. Results: We identified 6 studies: Four using electroencephalogram (EEG); 1 using magnetic resonance imaging (MRI) and FDG-PET (positron emission tomography); and 1 using MR spectroscopy. The EEG recordings suggested a reduction in alpha EEG peak frequency activity in the frontal regions of SCI patients with neuropathic pain. The MRI scans showed volume loss, primarily in the gray matter of the left dorsolateral prefrontal cortex, and by FDG-PET, hypometabolism in
the medial prefrontal cortex was observed in SCI patients with neuropathic pain compared with healthy subjects. In the MR spectroscopy findings, the presence of pain was associated with changes in the prefrontal cortex and anterior cingulate cortex. Conclusions: When analyzed together, the results of these studies seem to point out to a common marker of pain in SCI characterized by decreased cortical activity in frontal areas and possibly increased subcortical activity. These results may contribute to planning further mechanistic studies as to better understand the mechanisms by which neuropathic pain is modulated in patients with SCI as well as clinical studies investigating best responders of treatment. & Key Words: pain, spinal cord injury, neuropathic pain, neural markers, neuroplasticity
INTRODUCTION Address correspondence and reprint requests to: Felipe Fregni, MD, PhD, MPH, Spaulding Neuromodulation Center, Spaulding Rehabilitation Hospital, Harvard Medical School, 79/96 13th Street, Charlestown, MA, U.S.A. E-mail: [email protected]. P. Pascoal-Faria and N. Yalcin contributed equally. Submitted: May 9, 2014; Revision accepted: June 27, 2014 DOI. 10.1111/papr.12237
© 2014 World Institute of Pain, 1530-7085/14/$15.00 Pain Practice, Volume , Issue , 2014 –
Chronic pain is a major complication in patients following spinal cord injury (SCI), with a prevalence of 60% to 80%, and it is often refractory to medication.1–3 One-third of such patients describe their pain as severe, decreasing their quality of life.2–6 The mechanisms of pain syndromes after SCI are unknown. Initially, the area of the spinal cord near the site of the injury was considered to be where most
2
PASCOAL-FARIA ET AL.
anatomic and functional changes occurred, due to chemical synaptic alterations.3 However, recent animal studies have demonstrated broader neural areas are responsible for the pain following an SCI.7 Chronic pain in SCI patients increases the synaptic efficacy of somatosensory neurons in the dorsal horn of the spinal cord and effects dysfunction of the cortical circuits that are associated with sensory and pain processing, which is referred to as central sensitization.8–10 This phenomenon leads to an imbalance in the excitatory and inhibitory inputs and reorganizes the cortex, which involves a large neural network, including limbic structures, such as the anterior cingulate cortex, hippocampus, amygdala, and thalamic nuclei.11 Collectively, these changes constitute what is known as maladaptive plasticity. Based on the significance of central changes in the development and sustainment of chronic pain, neural markers that are associated with maladaptive plasticity must be identified to determine the mechanisms of its treatments and prognostic factors. No study has reported consistent and effective pharmacological and nonpharmacological treatments for neuropathic pain following SCI.6,12,13 Thus, the development of novel treatments for this condition remains an unmet clinical need. In this context, neural markers may be useful in clinical trials as surrogate outcomes. Therefore, electrophysiology and brain neuroimaging indexed neural markers could allow us to gain insights into the pathophysiology of neuropathic pain following SCI. In this study, we aimed to identify the most effective neural markers of neuropathic pain in SCI. We report significant regional changes at various electroencephalogram (EEG) frequencies and by magnetic resonance imaging (MRI) and FDG-PET (positron emission tomography).
OR electroencephalography OR electroencephalogram OR neurophysiological OR neurophysiologic OR neurophysiology OR signatures OR markers OR marker). Selection and Exclusion Criteria Studies were included per the following criteria: (1) published in English; (2) human subjects; (3) original research; and (4) articles on neural markers, using electrophysiological recordings or neuroimaging as an outcome measure in SCI patients with neuropathic pain. Additional exclusion criteria were: (1) studies that included surgical procedures, deep brain stimulation, or brain stem cell transplantation; (2) articles on pharmacological, biofeedback, and neuromodulation treatment techniques; and (3) articles that included only results on imaginary movement or other tasks. Data Extraction A data extraction sheet was developed to ensure the systematic collection of relevant data, adopted from the Cochrane Handbook of Systematic Reviews for Intervention Studies. All main information topics that are generally provided regarding neural markers of pain in SCI were covered. The following data were extracted from the included studies (for the full list of topics in the extraction sheet, see Table 1): name of first author; year of publication; guidelines and scales used for inclusion criteria; number of participants in each group (healthy and other cohorts) and their characteristics (gender, age); exclusion criteria; medication; neural markers that have been used and their characteristics; and significant results. Identification of Articles
METHODS Information Sources and Literature Search Studies in English that examined neural markers of neuropathic pain following SCI in human subjects were identified in PubMed; this search was performed on April 1, 2014. The following search string was used to identify these full-text articles: spinal cord injury AND (neuropathic pain OR pain OR neurogenic) AND (functional magnetic resonance imaging OR positron emission tomography OR single positron emission computed tomography OR EEG OR TMS OR tDCS or tACS OR tPCS OR transcranial OR neuronal polarization OR electrophysiology OR electrophysiologic OR electrophysiological
Two investigators (NY and PF) independently evaluated the titles and abstracts of the identified studies to exclude those that did not meet the inclusion criteria. Then, the full text of all eligible studies was retrieved and assessed for inclusion or exclusion by 2 investigators. If there was disagreement about a study’s eligibility, a consensus decision was reached by the investigators.
RESULTS Study Selection The literature search yielded 6 articles: 4 on EEG; 1 on MRI and FDG-PET; and 1 on MR spectroscopy. The
Neural Markers of Pain in Spinal Cord Injury 3
Table 1. Studies on Neural Markers of Neuropathic Pain in Spinal Cord Injury
Studies
Guidelines and scales used for assessment
Study Population
Number and Characteristics of healthy subjects (age/sex/other information)
Number and Characteristics of subjects with Neuropathic pain (NP) after SCI (age/sex/other information)
Number and Characteristics of subjects without NP after SCI (age/sex/other information)
16/Age/gender matched. Mean age 34 (10.7) 26/Age/gender matched
8/7 Male, 1 female. Mean age 35 (11.3). ASIA (4) T4 to T12 levels 14/Mean age of SCI 47 (15). AIS (A to D) C5 to T10 levels
8/8 Male. Mean age 33.5 (10.3). ASIA (4, T4 to T12 levels) 9/Age 47 (15), C5 to T10 levels
28/18 Male, 10 female. Mean age 44.57 (13.96) 10/7 Male, 3 female. Mean age 39.1 (10.1)
38/27 Male, 11 female. Mean age 51.24 (12.04)
16/15 Male, 1 female. Mean age 49 (12.82)
10/7 Male, 3 female. Mean age 45.2 (9.1). Pain level with 5 and more for at least 6 months, ASIA (A to D). VNS (≥ 5) 5/Mean age 36.4 (10.4), SCI more than 12 months. Mean duration of condition 63.6 (49.6) months since injury
10/8 Male, 2 female. Mean age 44.4 (8.1). SCI at least 1 year 5/Mean age 36.4 (10.4), SCI more than 12 months. Mean duration of condition 58.2 (59.7) months since injury
Boord et al. (2008)12
ASIA
32
Wydenkeller et al. (2009)15
AIS and 0 to 10 scale for pain intensity
49
Jensen et al. (2013)14
0 to 10 scale for pain intensity
82
Vuckovic et al. (2014)16
ASIA and VNS
30
Stanwell et al. (2010)17
ASIA
20
10/Mean age 36.4 (5)
Yoon et al. (2013)13
ASIA 11-point NRS system and BDI
20
10/Age/gender matched
Patients exclusion criteria
Patients with NP—Medication
Loss of consciousness or a period of post-traumatic amnesia following injury
Medication was kept (gabapentin, amitriptyline, mexiletine, pregabalin, morphine, temazepam, diazepam). A significant reduction in Peak y–a band frequency, as an effect of medication, was seen only at 3/14 sites but no significant effect on EC/EO reactivity Medication was kept (nonsteroidal antiinflammatory drugs, anticholinergics, antiepileptics, spasmolytics, antidepressants, opioids). No significant effect of medication
Subjects with additional spinal column lesions belowT10. Neurological history except SCI and mild depression. Subjects with NP at the level or below the level of the lesion < 18 years, do not understand English, SCI < 12 months, report pain for < 6 months, pain intensity average < 4 (0 to 10 scale). History of seizure activity, significant head injury, skull defects
Central acting medications and others were kept. No significant effect of medication
SCI < 1 year after injury with a spinal injury above T1. Presence of any chronic or acute pain at the time of the experiment, brain injury, or other know neurology disease
Medication was kept (baclofen, carbamazepine, gabapentin, pregabalin, amitriptyline, diazepam). No reference regarding the medication effect on the results
10/6 Males, 4 females. Mean age 39 (8.6). More than 6 months since SCI, stable chronic pain at least 3 months, only with neuropathic pain. ASIA (A,B) Electrophysiology or NeuroImaging Technique and its Characteristics
Results/Comments
EEG: eyes closed and open/14 channels/bandwidth of DC–500 Hz and sampled at 2,048 Hz. Peak frequency in the theta–alpha band (4 to 13 Hz)
Significantly reduced EEG peak frequency and significantly reduced EEG spectral reactivity were observed in patients with NP
EEG: 10 to 20, 30 scalp electrodes and 2 electrodes below the outer canthus of the eyes, 500 Hz with a 140-Hz low pass filter 3-minutes closed eyes, delta, alpha, beta, and theta. The pain intensity did not correlate with EEG EEG: 10 to 20, 19 electrodes, bandpass 0.3 to 70 Hz, sampled at the rate 250 Hz. Awake state and eyes closed, delta, alpha, beta, and theta. The pain intensity did not correlate with EEG
EEG peak frequency in the 6 to 12 Hz band was significantly slower in patients with bNP compared to patients without bNP. The peak and band topographies did not differ statistically between these groups. The pain intensity did not correlate with EEG
EEG: 10 to 10, 61 electrodes, using and ear linked reference and AFz ground, eyes opened and eyes closed, 2 minutes for each conditions repeated 3 times Channels sampled at 1,000 Hz
Significant effects did emerge in absolute power activity for frontal localizations (FP1, FP2, F3, FZ, and F4), with less alpha in the SCI pain group, relative to the other groups. No significant changes were seen for any electrode site for relative delta or beta. More relative theta was observed only at P3, 01, 02 in patients with pain after SCI. Less relative alpha was seen at 14 of the 19 electrodes (all except F7, T3, T6, = 1, and O2), in the SCI pain group. The pain intensity did not correlate with EEG Paraplegic patients with CNP had increased alpha power in eyes open (EO) state in most of the recording sites than patients with no pain. In EC state, there was significant increase in alpha power at only frontal regions. No difference among groups was found between theta PSDs in the eyes closed (EC) state whereas theta power only increased in frontal and occipital regions in eyes opened (EO) state. No significant difference between groups in the beta range was found
4
PASCOAL-FARIA ET AL.
Table 1. (Continued) Electrophysiology or NeuroImaging Technique and its Characteristics
Patients exclusion criteria
Patients with NP—Medication
Incomplete SCI and/or tetraplegia, history of psychopathology, previous pain syndromes prior to the injury, epilepsy, or the presence of non-MR compatible devices
Medication was kept (pregabalin, morphine, tramadol, amitriptyline). No reference regarding the medication effect on the results
MRS: TR/TE 3,000/20 ms, 2,048 FID points, bandwidth 2 kHZ, 6 cm3 voxel, Location: Left ACC, left PFC, and left thalamus
Patients with neuropsychological disease or any other systemic disease affecting the central or peripheral nervous system
No reference regarding the medication effect on the results
FDG-PET: 10 minutes emission scans. Brain MRI: 3T MRI, TR 8.1 ms, TE 4.6 ms, flip angle 8°, 175 slices, 1 mm thickness and matrix size 256 9 256. DTI: TR 7514.1 ms, TE 65.9 ms, thickness 2 mm, flip angle 100°, 75 slices, matrix size 128 9 128
Results/Comments Spectroscopy from the: (1) thalamus (significant differences for NAA, choline-containing compounds, glutamine, and glutamate), distinguishing SCI patients (with and without pain) from control; (2) prefrontal cortex (NAA, glutamate, glutamine, choline-containing compounds and taurine, GABA, glycine, and others) and anterior cingulate cortex (myoinositol, aspartate and others) distinguishing SCI with pain from patients without pain Hypometabolism in the left and right middle frontal gyrus and decrease gray matter volume was seen in the left middle frontal gyrus, bilateral anterior insula and right suggenual anterior cingulate cortex. Lower mean diffusivity (MD) was identified in deep white matter regions such as in the right internal capsule including the region from the anterior to posterior limb, cerebral peduncle, and anterior corona radiata
ASIA, American Spinal Injury Association; AIS, American Spinal Injury Association Impairment Scale; BDI, Beck Depression Inventory; EEG, electroencephalogram; MRI, magnetic resonance imaging; NRS, Numeric Rating Scale; PET, positron emission tomography; SCI, spinal cord injury; VNS, Visual Numerical Scale.
full text of these articles was reviewed and included in this review.
particularly in the frontal regions, in SCI patients with neuropathic pain.
Study Characteristics
Outcome Measures
The studies included only participants with neuropathic pain after SCI. The severity of the pain was assessed on a pain scale from 0 to 10 in a clinical interview, in which 0 was no pain and 10 was the most intense pain imaginable. In all studies, the clinical features of the neuropathic pain were examined, including the level, severity, and duration of pain after the SCI. The pooled size of the studies was 233 subjects. The sample sizes of each study ranged from 20 to 82. The EEG studies comprised 193 participants, and the neuroimaging studies (MRI, MR spectroscopy, and PET) included 40 participants. All studies allowed participants to continue their current medications. Whereas the results were unaffected by the use of medications in Boord et al., Yoon et al., Jensen et al., and Wydenkeller et al.,12–15 Vuckovic et al. did not analyze the influence of medications, and no references were cited in Stanwell et al.16,17 Table 1 summarizes the results and design of each study. The 6 studies examined neural markers of neuropathic pain after SCI by EEG, MRI, MR spectroscopy, and FDG-PET. Generally, there was a reduction in the anatomic and functional activity of the cortex,
For the EEG studies, peak frequencies and relative and absolute power activities in all frequency bands (alpha, theta, delta, beta, and gamma) were calculated with the eyes open and/or closed. The MRI studies measured volume changes in various brain locations, whereas the MR spectroscopy study focused on the metabolism in the anterior cingulated gyrus, prefrontal cortex, and thalamus. The PET study assessed changes in regional metabolic activity. Electroencephalogram Most EEG studies in patients with neuropathic pain after SCI demonstrated a reduction in EEG peak theta– alpha frequency activity with the eyes closed at most sites compared with healthy controls and SCI patients without pain. Moreover, the alpha power declined significantly, primarily in the frontal areas. However, the pain intensity and EEG activity did not correlate (Table 1). In the larger study, in which EEG was performed with the eyes closed, Jensen et al.14 found that absolute alpha power significantly decreased in frontal locations in SCI
Neural Markers of Pain in Spinal Cord Injury 5
patients with pain vs. SCI patients without pain and healthy controls. Further, less relative alpha activity was noted in the SCI pain group relative to no-pain controls at 14 of the 19 electrode sites (ie, all except F7, T3, T6, O1, and O2). With regard to theta activity, the relative power was significantly larger at P3, O1, and O2 in SCI patients with pain compared with those without pain. Similar findings were reported in Boord et al.12 and Wydenkeller et al.15 Peak theta–alpha frequency decreased at all sites with the eyes closed in SCI patients with pain vs. those without pain. Further, Boord et al.12 generated the same results in comparing SCI patients with pain and healthy controls, whereas Wydenkeller et al.15 did not find any significant differences between these groups. Vuckovic et al.16 was the only study that provided data with the eyes open, noting a significant increase in alpha power at most sites between SCI patients with pain and without pain. Also, theta power was also larger with the eyes open in the frontal and occipital regions in the pain group compared with the no-pain group. Conversely, with the eyes closed, the results were less conclusive. The alpha power increased significantly in the frontal regions in SCI patients with pain vs. those without pain, but there were no significant differences in theta band. Alpha and theta did not differ between SCI patients with pain and healthy controls. For the other EEG bandwidths (delta, beta, and gamma), these studies showed nonsignificant trends, and in 2 of the studies,14,15 mean pain intensity did not correlate with EEG power activity. Magnetic Resonance Imaging Two MRI studies examined 2 techniques: MRI volumetric analysis and MR spectroscopy. In Yoon et al.,13 by MRI, there was significant volume loss in the left middle frontal gyrus, bilateral anterior insulae, and right subgenual ACC using 3 TMRI. There were no increases in gray matter volume in any brain region in patients with neuropathic pain after SCI compared with healthy subjects. Also, there was significantly lower mean diffusivity in several deep and peripheral white matter regions. For more details see Table 1. By MR spectroscopy, the voxels in the anterior cingulated gyrus, prefrontal cortex, and thalamus were examined, demonstrating that changes in the prefrontal cortex and anterior cingulate cortex were related specifically to the presence of pain. The chief metabolites
that contributed to the presence of pain were NAA, glutamate, glutamine, choline-containing compounds, taurine, GABA, and glycine in the prefrontal cortex, and myoinositol and aspartate in the anterior cingulate cortex. In addition, NAA was significantly increased in the thalamus of SCI patients compared with healthy subjects.17 Positron Emission Tomography Metabolism in the left and right medial frontal gyrus declined in SCI patients with neuropathic pain compared with healthy controls. No region was found to have increased metabolism.13
DISCUSSION Six clinical studies that used electrophysiological and neuroimaging techniques met the inclusion criteria. Four studies used EEG, 1 used MRI and FDG-PET, and 1 performed MR spectroscopy. By EEG, patients with neuropathic pain after SCI show a reduction in EEG peak frequency activity with the eyes closed, especially in the alpha and theta bands, compared with healthy controls and SCI patients without pain.12,15 Further, in Jensen et al.,14 alpha power was also reduced in frontal areas in SCI patients with neuropathic pain with the eyes closed. Alpha reflects the level of activity in the cortex. Moreover, alterations in the dynamics of thalamocortical loops affect persistent and synchronous alpha activity.18,19 Reductions in EEG frequency are caused by deafferentation, which can be explained by thalamocortical dysrhythmia that arises from hyperpolarization of thalamic neurons.20 This dysrhythmia mediates the genesis of neuropathic pain through disinhibition and subsequent activation of networks in the cortical pain matrix.12,21 Neuropathic pain has also been linked to thalamic dysrhythmia in in vivo animal studies.22–24 Moreover, these results are consistent with other studies that showed a decrease in dominant EEG frequency toward lower frequencies in patients with various types of chronic pain.25,26 Conversely, Vuckovic et al.16 showed a significant increase in alpha band, primarily in frontal regions, in pain subjects compared with those with no pain with the eyes closed. Notably, despite these changes, there were no differences in alpha or theta band, regardless of the state of the eyes, in the pain group compared with healthy subjects. The variability between studies might
6
PASCOAL-FARIA ET AL.
be attributed to the state of the eyes during the EEG recordings. In fact, alpha band activity is associated with more variability with eyes open. Individual differences, such as the use of medications, between study subjects might also have influenced the results. In Jensen et al.,14 Boord et al.,12 and Wydenkeller et al.,15 there were no significant effects of medication use on the results. However, in Vuckovic et al.,16 although medications were continued during the recordings, their effects were not analyzed. The neuroimaging studies had findings in the same direction as the EEG studies. The MRI study by Yoon et al.13 reported volume loss in the left middle frontal gyrus, bilateral anterior insula, and right subgenual ACC and lower mean diffusivity in white matter regions, whereas FDG-PET revealed hypometabolism in the left and right medial frontal gyrus. Atrophy of the DLPFC has been associated with chronic pain conditions in several studies.27–29 This area of the brain governs the balance between the descending fascilatory systems and descending inhibitory systems, which actively control the perception of pain.11,30 Moreover, the prefrontal area is linked to cognitive and attentional processing of pain stimuli.26,27,31 The subgenual anterial cingulate gyrus (sACC) and insula have been implicated in emotional processes that mediate effective responses to pain stimuli.31,32 Further, although these regions differ in location, they overlap functionally,31 which is consistent with studies that have addressed chronic pain.28,33,34 An MR spectroscopy study by Stanwell et al.17 agrees that critical effective emotional circuits, such as the anterior cingulate cortex and insulae, change significantly between SCI patients with and without pain. Most of our findings were characterized by decreased activity in the frontal areas, as supported by the alpha rhythm dysrhythmia in the EEG studies. These studies did not examine the motor cortex. Other neuropathic syndromes are associated with decreased intracortical inhibition and increased intracortical fascilation, which suggests that M1 modulates thalamic activity directly. These results can be used to develop novel targets for the treatment of neuropathic pain in SCI. Based on the connectivity of these areas, the efficacy of treatments that target the areas that we have discussed could be measured by determining the activity in these targets. Moreno-Duarte et al.35 reported significant pain improvement only when primary motor cortex (using tDCS) and hypothalamic region (using cranial electrical stimulation) were targeted. Yet, other
studies that performed different brain stimulation techniques with disparate targets of the nervous system, such as the brain motor cortex (rTMS), hypothalamus (CES), peripheral nerves (TENS), and the dorsal horn (SCS), found less of an effect on pain or none at all. Recent advances in neuroimaging and neurophysiological methods have provided support for maladaptive plasticity that is associated with chronic pain. Although there have been few such studies exist and despite their small sample size, when analyzed together, they form the basis for a neural signature of maladaptive plasticity in SCI and chronic pain. The main finding appears to be related to distributed cortical-subcortical dysregulation, in which subcortical circuits, such as the thalamus, are overactive and emotional affective circuits, such as the sACC and insulae, are hypoactive. Most studies agree that alpha activity decreases in the frontal areas. Moreover, other cortical circuits, such as M1, are significant areas that are associated with the development and regulation of pain processing in SCI. Further studies that assess other methods, such as TMS indices of cortical excitability, might be useful.
ACKNOWLEDGEMENTS This study was supported by a grant from the Department of Education, NIDRR H133N110010 (SCI— Model Systems). Paula Pascoal-Faria acknowledges the support of the Strategic Project (PEST-OE/EME/ UI4044/2013) funded by the Portuguese Foundation for Science and Technology and also the support of the European Commission through the Marie Curie Project “International Research Exchange for Biomedical Devices Design and Prototyping” “IREBID”.
REFERENCES 1. Bonica JJ. History of pain concepts and pain therapy. Mt Sinai J Med. 1991;58:191–202. 2. Siddall PJ, Taylor DA, McClelland JM, Rutkowski SB, Cousins MJ. Pain report and the relationship of pain to physical factors in the first 6 month following spinal cord injury. Pain. 1999;81:187–197. 3. D’Angello R, Morreale A, Donadio V, et al. Neuropathic pain following spinal cord injury: what we know about mechanisms, assessment and management. Eur Rev Med Pharmacol Sci. 2013;17:3257–3261. 4. St€ ormer S, Gerner HJ, Gr€ uninger W, et al. Chronic pain/dysaesthesiae in spinal cord injury patients: results of a multicentre study. Spinal Cord. 1997;35:446–455. 5. Siddall PJ, McClelland JM, Rutkowski SB, Cousins MJ. A longitudinal study of the prevalence and characteristics
Neural Markers of Pain in Spinal Cord Injury 7
of pain in the first 5 years following spinal cord injury. Pain. 2003;103:249–257. 6. Wasner F. Central pain syndromes. Curr Pain Headache Rep. 2010;14:489–496. 7. Hulsebosch CE, Hains BC, Crown E, Carlton S. Mechanisms of chronic central neuropathic pain after spinal cord injury. Brain Res Rev. 2009;60:202–213. 8. Woolf CJ. Evidence for a central component of postinjury pain hypersensitivity. Nature. 1983;306:686–688. 9. Cook AJ, Woolf CJ, Wall PD, McMahon SB. Dynamic receptive field plasticity in rat spinal cord dorsal horn following C-primary afferent input. Nature. 1987;325:151– 153. 10. Ghosh A, Sydekum E, Haiss F, et al. Functional and anatomical reorganization of the sensory-motor cortex after incomplete spinal cord injury in adult rats. J Neurosci. 2009;29:12210–12219. 11. Kwon M, Altin M, Duenas H, Alev L. The role of descending inhibitory pathways on chronic pain modulation and clinical implications. Pain Pract. 2013; doi:10.1111/papr. 12145. 12. Boord P, Siddall PJ, Tran Y, Herbert D, Middleton J, Craig A. Electroencephalographic slowing and reduced reactivity in neuropathic pain following spinal cord injury. Spinal Cord. 2008;46:118–123. 13. Yoon EJ, Kim YK, Shin HI, Lee Y, Kim SE. Cortical and white matter alterations in patients with neuropathic pain after spinal cord injury. Brain Res. 2013;1540:64–73. 14. Jensen MP, Sherlin LH, Gertz KJ, et al. Brain EEG activity correlates of chronic pain in persons with spinal cord injury: clinical implications. Spinal Cord. 2013;51:55–58. 15. Wydenkeller S, Maurizio S, Dietz V, Halder P. Neuropathic pain in spinal cord injury: significance of clinical and electrophysiological measures. Eur J Neurosci. 2009;30:91– 99. 16. Vuckovic A, Hasan MA, Fraser M, Conway BA, Nasseroleslami B, Allan DB. Dynamic oscillatory signatures of central neuropathic pain in spinal cord injury. J Pain. 2014;15:645–655. 17. Stanwell P, Siddall P, Keshava N, et al. Neuro magnetic resonance spectroscopy using wavelet decomposition and statistical testing identifies biochemical changes in people with spinal cord injury and pain. Neuroimage. 2010;53:544– 552. 18. Llinas RR, Steriade M. Bursting of thalamic neurons and states of vigilance. J Neurophysiol. 2006;95:3297– 3308. 19. Hindriks R, van Putten MJ. Thalamo-cortical mechanisms underlying changes in amplitude and frequency of human alpha oscillations. Neuroimage. 2013;70:150–163. 20. Ching S, Cimenser A, Purdon PL, Brown EN, Kopell NJ. Thalamocortical model for a propofol-induced alpharhythm associated with loss of consciousness. Proc Natl Acad Sci USA. 2010;107:22665–22670.
21. Walton KD, Llin as RR. Central pain as a thalamocortical dysrhythmia. Chapter 13—A thalamic efference disconnection? In: Kruger L, Light AR, eds. Translational Pain Research: From Mouse to Man. Boca Raton, FL: CRC Press; 2010:1–16. 22. Llin as R, Ribary U, Jeanmonod D, et al. Thalamocortical dysrhythmia I. Functional and imaging aspects. Thalamus Relat Syst. 2002;1:237–244. 23. Gerke MB, Duggan AW, Xu L, Siddall PJ. Thalamic neuronal activity in rats with mechanical allodynia following contusive spinal cord injury. Neuroscience. 2003;117:715– 722. 24. Kim D, Park D, Choi S, et al. Thalamic control of visceral nociception mediated by T-type Ca2+ channels. Science. 2003;302:117–119. 25. Stern J, Jeanmonod D, Sarnthein J. Persistent EEG overactivation in the cortical pain matrix of neurogenic pain patients. Neuroimage. 2006;31:721–731. 26. Sarnthein J, Stern J, Aufenberg C, Rousson V, Jeanmonod D. Increased EEG power and slowed dominant frequency in patients with neurogenic pain. Brain. 2006;129:55–64. 27. Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain. 2005;9:463–484. 28. Seminowicz DA, Wideman TH, Naso L, et al. Effective treatment of chronic low back pain in humans reverses abnormal brain anatomy and function. J Neurosci. 2011;31:7540–7550. 29. Obermann M, Rodriguez-Raecke R, Naegel S, et al. Gray matter volume reduction reflects chronic pain in trigeminal neuralgia. Neuroimage. 2013;74:352–358. 30. Lorenz J, Minoshima S, Casey KL. Keeping pain out of mind: the role of the dorsolateral prefrontal cortex in pain modulation. Brain. 2003;126:1079–1091. 31. Yoon EJ, Kim YK, Kim HR, Kim SE, Lee Y, Shin HI. Transcranial direct current stimulation to lessen neuropathic pain after spinal cord injury: a mechanistic PET study. Neurorehabil Neural Repair. 2014;28:250–259. 32. Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain. 2011;152(suppl 3):S2– S15. 33. Kuchinad A, Schweinhardt P, Seminowicz DA, et al. Accelerated brain gray matter loss in fibromyalgia patients: premature aging of the brain? J Neurosci. 2007;27:4004– 4007. 34. Egloff N, Sabbioni ME, Salathe C, Wiest R, Juengling FD. Nondermatomal somatosensory deficits in patients with chronic pain disorder: clinical findings and hypometabolic pattern in FDG-PET. Pain. 2009;145:252–258. 35. Moreno-Duarte I, Morse LR, Alam M, Bikson M, Zafonte R, Fregni F. Targeted therapies using electrical and magnetic neural stimulation for the treatment of chronic pain in spinal cord injury. Neuroimage. 2014;85(Pt 3):1003–1013.
