Changes In Brain

The Peripheral Neuropathy Solution

The Peripheral Neuropathy Program

Get Instant Access

Great advances have been made during last couple of decades in our understanding of the pathophysiology of neuropathic pain, as a result of basic scientific findings of the pathological and biochemical changes in the peripheral and central nervous system. The neuronal changes that take place at the level of human brain as the consequence of peripheral neuronal injury, such as in case of PDN, lead to the processes that maintain neuropathic pain known as central sensitization. Examination of the cerebral correlates of central processes related to pain and central sensitization has only been possible during last decade and a half with introduction of neuroimaging. Noninvasive brain imaging technologies provide the opportunity to directly study human clinical conditions. Initially, blood flow based positron emission tomography (PET) scan was utilized (9), but advantages of functional magnetic resonance imaging (fMRI), such as improved temporal and spatial resolution have made it a method of choice for pain studies (10,11). Electrophysiological studies using evoked potentials and magnetoencephalography have also been used to elucidate cerebral mechanisms related to pain (12). This chapter will review what has been learned about changes in brain as the result of neuropathic pain on the basis of fMRI and electrophysiological studies.

Certainly, the role of brain in pain mechanisms is very complex ranging from the perception and experience of pain, to pain modulation, and to formation of behavioral response to pain. Complexity increases in the case of neurological disorders, such as neuropathic pain, when the brain itself undergoes significant changes. This degree of complexity has been appreciated for a long time, but the means to study underlying mechanisms have not been available until recently. At the present time neuroscience is able to address most general aspects of brain mechanisms related to pain, including neuropathic pain, which will be discussed here.

In spite of rapidly increasing number of pain imaging studies, the number of studies specifically focussed on neuropathic pain is a very limited. On the extensive search of published literature no study of painful diabetic neuropathic has been identified. Consequently, discussion in this chapter is presented with the commonly made assumption that findings from other neuropathic pain disorders, which were studied with brain functional imaging, are applicable to PDN. These assumptions are based on the essential clinical similarities of PDN and other painful neuropathic disorders. Certainly, information about the brain-specific changes that occur as a consequence of PDN will come from future functional imaging studies of pain in patients with diabetic neuropathy.

First functional laboratory studies that helped to identify cerebral structures that participate in pain perception will be briefly reviewed, and this will be followed with the review of studies with chronic pain, including neuropathic pain.

Functional Brain Responses to Acute Laboratory Pain

Current functional brain studies utilize stimulation paradigms to identify brain structure involved in physiological processes, such as pain. Numerous pain studies have been conducted using wide variety of stimulation protocols including thermal pulses from heat and cold thermodes, laser heat pulses, limb immersions into cold or hot temperature baths, thermal grill illusion, capsaicin intradermal injections, and distention of internal organs, such as rectum and esophagus (11,13-18). These studies revealed a number of brain structures that are involved in the experience of pain, and a complex (matrix/network instead?) of brain structures that are involved depends on many factors, including the type and length of stimulus used and experimental conditions. Regional cerebral blood flow (rCBF) increases, which reflects increase in neuronal activity of corresponding brain structures, noxious stimuli are most consistently observed in secondary somatic (SII) and insular cortex (IC) region, and in the anterior cingulate cortex (ACC), and less consistently in the contralateral thalamus and the primary somatic cortical area (SI) (11,17,19-23). Activity of the lateral thalamus, SI, SII, and posterior IC are believed to be related to the sensory-discriminative aspects of pain experience, which provides the subject with the ability to localize the site of stimulation and intensity of stimulus. Activity in SI is observed in less than half of the studies and the probability that SI activation is observed appears to be because of influences such as the stimulated body surface, which would represent spatial summation and the attention to or away form the stimulus (11,17,19-21,24,25). A number of studies reported the thalamic responses as bilateral and this observation would probably reflect generalized arousal in reaction to pain (17,19,26). The ACC appears to participate in the affective and attentional components of pain sensation and in selection of the response to the stimulus (20,21,25,27,28). Functional properties of the ACC from imaging studies would suggest that ACC does not code for stimulus intensity or for location of the stimulus. Increased rCBF in the posterior parietal and prefrontal cortices is probably because of activation of attention and memory-related brain structures in response to noxious stimuli. Activation of motor-system related brain structures, such as the striatum, cerebellum, and supplementary motor area are frequently observable, although not frequently commented on, and they are probably involved in motor planning in response to pain and generating pain-related behaviors. Physiologically, a few brain regions are involved in descending pain control and modulation, such as the periaqueductal gray and brainstem nuclei that at times were imaged, but this area is difficult to image, resulting in inconsistent findings. A significant intersubject variability in the activation of any one of the pain-related cerebral regions, particularly during heat- and cold-evoked pain have been noted (20,29,30). In contrast to nonpainful stimuli which showed only transient responses to the onset or offset, painful stimulation was observed to result in a sustained response throughout its duration in the temporoparietal, inferior frontal cortex (IFC), and ACC. These regions therefore show tonic responses to stimuli with ongoing salience (31). The thalamus and putamen also responded throughout tonically painful, but not nonpainful stimulation. These observations would then implicate the basal ganglia in supporting voluntary sustained attention and would suggest that the basal ganglia may play a more general role in supporting sustained attention.

When visceral stimuli are applied and produce an amount of pain that is reported as equivalent to the intensity of pain applied to skin in the same subjects, the similar overall pattern of activation at the SII and parietal cortices, thalamus, basal ganglia, and cerebellum are observed. However, at the insular, primary somatosensory, motor, and prefrontal cortices there are different patterns of activation, suggesting somewhat different brain correlation for cutaneous vs visceral pain experiences (32-34).

Electrophysiological studies, such as evoked potentials have advantage over fMRI because of superior time resolution. Several dipole source analysis as well as subdural recordings have confirmed that the earliest evoked potential following painful laser stimulation of the skin derives from sources in the parietal operculum. Based on imaging and electrophysiological studies in humans it has been concluded that parasylvian cortex is activated by painful stimuli, and this would be one of the first cortical relay stations in the central processing of these stimuli (35). There is evidence for close location but separate representation in parasylvian cortex of pain in a few areas, such as deep parietal operculum and anterior insula and these are distinct from representations for innocuous touch, such as SII and posterior insula. It is likely that some of these areas are involved in sensory-limbic projection pathways that may subserve the recognition of potentially tissue damaging stimuli as well as pain memory (35). With these types of studies it is possible to initiate further analysis of the functional anatomy specific to sensory-discriminative, affective-motivational, and cognitive-evaluative components of pain.

In summary, there is now ample evidence that a well-defined cerebral network participates in human perception of acute pain paradigms. The areas most consistently observed include: SI, SII, ACC, IC, prefrontal cortex (PFC), thalamus, and cerebellum (Fig. 1.).

Changes With Chronic Pain, Including Neuropathic Pain

Manifestations of chronic pain are fundamentally different from acute pain and attempts have been made to identify those differences in functional imaging and electro-physiological studies. There are many conceptual and technical challenges in studying chronic pain, especially, neuropathic pain because of its complexity. Components of neuropathic pain symptoms include spontaneous ongoing pain, spontaneous paroxysms, and stimulus evoked pain (36,37). It is difficult to gain an accurate picture of this range of symptoms with neuroimaging techniques. Spontaneous paroxysms are unpredictable and random by nature and consequently, almost impossible to study. Imaging of spontaneous ongoing pain is difficult because it is necessary to utilize subtraction from nonpain control state for imaging pain studies, so only patients who can achieve a substantial degree of pain relief can be studied with current imaging technology. Even if these conditions are met, it is difficult to know how this report of relief compares to the pain-free state in nonpatient populations and how neuroplastic changes resulting from chronic pain affect either state. Most of the imaging and electrophysiological studies have therefore concentrated on evoked components of pain. Understanding how these results represent the neural changes that underlie the range of symptoms experienced in syndromes like PDN is an ongoing challenge for chronic pain research.

In patients with chronic spontaneous pain, a few pain-imaging studies revealed relative decreases of resting rCBF in contralateral thalamus, when compared with the ipsi-lateral side (38,39). These findings suggest that ongoing neuropathic pain because of either central or peripheral etiology is linked to thalamic hypoperfusion and that analgesic treatments are mediated through an increase in thalamic blood flow (38,40,41).

One of the aspects of pain most relevant to the study of chronic pain such as PDN are phenomena of hypersensitivity, such as allodynia or hyperalgesia, depending on whether stimulus is innocuous or noxious, respectively. In patients suffering from

Fig. 1. Regions of brain activated with painful stimuli. Cerebral regions activated by pain are primary somatosensory cortex, secondary sensory cortex/insular cortex, and thalamus seen on horizontal planes, anterior cingulate cortex seen in medial sagittal view and dorsolateral prefrontal cortex on coronal view (Courtesy of Tim Salomons and Regina Lapate, The Waisman laboratory for brain imaging and behavior, University of Wisconsin -Madison).

Fig. 1. Regions of brain activated with painful stimuli. Cerebral regions activated by pain are primary somatosensory cortex, secondary sensory cortex/insular cortex, and thalamus seen on horizontal planes, anterior cingulate cortex seen in medial sagittal view and dorsolateral prefrontal cortex on coronal view (Courtesy of Tim Salomons and Regina Lapate, The Waisman laboratory for brain imaging and behavior, University of Wisconsin -Madison).

allodynia, this phenomenon is evoked during PET or fMRI sessions by applying repeated innocuous stimuli. The images from scans obtained during allodynia stimulation are compared either with resting no-stimulation scans, or with an identical stimulation of a nonaffected nonpainful body area.

Brushing the nonpainful limb leads to activations of the contralateral SI, contralateral parietal association cortex (PA), bilateral SII, and contralateral IC, and brushing the skin where patients experience allodynia pain, than cortical responses are partially overlapping with those induced by the nonpainful brush stimulation (42). In addition, the contralateral IFC and the ipsilateral IC are activated. Direct comparison between nonpainful brushing and brush-evoked allodynia revealed significant increases in blood-oxygen-level dependent (BOLD) signals in contralateral SI, PA, IFC, and bilateral SII/IC during allodynia. This study highlights the importance of a cortical network consisting SI, PA, SII/IC, and IFC in the processing of dynamic-mechanical allodynia in the human brain (42) and possibly relating to sensory and attentional abnormalities underlying allodynia.

Temporal characteristics of brain responses have also been investigated. The cortical processing of allodynia resulting from neuralgia of the lateral cutaneous femoral nerve was investigated by means of magnetoencephalography (43,45). Brushing the unaffected thigh produced subsequent activation of the contralateral SI with peak latencies of 37 and 56 ms. Allodynic stimulation with brushing of the affected side leads to comparable activation of the contralateral SI cortex but with stronger magnetic fields, and the corresponding equivalent current dipoles located more laterally, suggestive of cortical reorganization and hyper-responsivity. Allodynia is also accompanied by an activation of the cingulate cortex, occurring only 92 ms. after stimulus onset and this observation would suggests that A-P-fiber-mediated neuronal pathways are involved in this type of allodynia. The results cited here support the possibility that cortical reorganization is one of the underlying processes that characterizes allodynia and possibly chronic pain and that early activation of the cingulate cortex may be involved in the cortical processing of allodynia.

When brain activity to acute thermal painful stimuli are examined in chronic pain patients, the resulting pattern is very similar to that seen for acute pain in normal subjects, independent of the type of chronic pain as seen in chronic back pain patients (46) and in complex regional pain syndrome patients (41,47). Although chronic pain patients may have various cutaneous sensory abnormalities (48), mapping brain responses to acute pain does not distinguish them from normal subjects. On the other hand, when brain activity specifically related to the chronic pain is isolated then activity seems to be preferentially involving prefrontal cortical regions (41).

Injury of peripheral or central neural tissues leads to long lasting spinal and supraspinal reorganization that includes the forebrain. These forebrain changes may be adaptive and facilitate functional recovery, or they may be maladaptive preventing or prolonging the painful condition and interfering with treatment (42). In an experimental model of heat allodynia, functional brain imaging was used to show that the forebrain activity during heat allodynia is different from that during normal heat pain by increase of dorsolateral prefrontal cortex (DLPFC). Consequently, as it could be argued that during heat allodynia, cortical areas, specifically the DLPFC, can attenuate specific components of the pain experience such as, by reducing the functional connectivity of subcortical pathways. The forebrain of patients with chronic neuropathic pain may undergo pathologically induced changes that can impair the clinical response to treatment (42). Stimuli that are severely painful when applied to the allodynic side, activated regions in the contralateral hemisphere that mirrored the "control" network, with however, lesser activation of the SII and insular cortices in patients with neuropathic pain (49). Increased activation volumes were found in contralateral SI and primary motor cortex, whereas ipsilateral responses appeared very small and restricted after control stimuli, they represented the most salient effect of allodynia and were observed mainly in the ipsilateral parietal operculum of the SI, SII, and insula. Allodynia producing stimuli also recruited additional responses in motor and premotor areas (supplementary motor area), in regions involved in spatial attention (posterior parietal cortices), and in regions linking attention and motor control (mid-ACC) (49).

Series of studies that were conducted not only with fMRI but also using magnetic resonance spectroscopy (MRS) indicate that cortical circuitry underlying chronic pain is distinct from that observed in acute pain, and preferentially involves orbital prefrontal cortex. More specifically regional brain chemistry changes in patients with chronic low back pain were examined and when compared with age- and gendermatched healthy control subjects, these subjects were found to have decreased brain chemical concentrations for multiple chemicals in both DLPFC and orbital frontal cortex and no detectable changes in primary sensory-motor cortex, ACC, insular cortex, or thalamus. Relationships between brain chemicals disrupted in patients with chronic low back pain, in a unique pattern the relation to pain as compared with anxiety were demonstrated in the analysis across number of brain regions. In addition to abnormalities of brain chemistry there was a decreased cortical gray matter size, as well as decreased prefrontal cortical gray matter density. Moreover, chronic pain patients showed a specific cognitive deficit, which is consistent with the brain activity observed in such patients along with the observed chemical and morphological abnormalities as well. Therefore, it could be concluded that chronic pain is reflected at the cortical level, and is associated with cortical reorganization and perhaps even neurodegeneration.

In summary, cerebral responses to pain are complex and dynamic in nature and in the case of chronic pain, especially neuropathic pain, changes involve more profound characterized by the involvement of entire pain-related network and some of the elements undergo significant reorganization and possibly neurodegeneration. All of these observations have significant implication for development of treatment strategies. As stated in the introduction of this section, the above discussed findings were result of studies with various neuropathic pain disorders which share fundamental characteristics with PDN, and it would be expected that many of the observations above would apply to patients with PDN.

Pain Models and Translational Neuropathic Pain Research

Models of diabetic neuropathy and translational research related to diabetic neuropathy, including painful neuropathies, are also discussed in previous chapters of this book, so to avoid repetition this chapter will concentrate on neuropathic pain in more general terms. Certainly, the appropriate reference to diabetic neuropathy will be made.

Major advances in understanding of neuropathy have been achieved during last couple of decades primarily because of development and from results of studies of animal models that mimic human neuropathic pain (50,51). First animal models utilized nerve trauma such as constriction (52) or partial nerve cut (53) or nerve root cut (50), as the mechanisms of nerve insult, however, other disease processes were utilized to induce nerve injury, such as diabetes (1,55-57) and chemotherapy (58). On the basis of these studies, a wealth of information about underlying mechanisms was obtained and one of the main conclusions is that neuropathic pain is a complex and dynamic disease process. Although injury can be limited to the peripheral nerves, the pain system along the entire neuroaxis is affected, from the receptors, to primary afferents, dorsal horn, spinothalamic tract, brain-stem nuclei with discending modulatory projection pathways, to thalamus and brain. Pathophysiological processes that characterize neuropathic pain include peripheral and central sensitization, which are manifestations of a complex interaction between increased excitability, decreased inhibition and activation of immune responses. The degree of complexity of neuropathic pain indicates that neuropathic pain is a disordered system, rather than a disfunction of particular pain system component. This view could than explain the complexity of PDN, and also the possibility that peripheral nerve injury because of diabetes could than be maintained by central processes as suggested by Calcutt (1,54).

In addition to animal models, advances in neuropathic pain were possible because of the increased sophistication of human laboratory studies in patients with neuropathic pain. Although basic concept of quantitative sensory testing (QST) have been known for more than a century and applied to various neuropathic pain disorders such as posther-petic neuralgia and PDN for decades, expanding the number of testing methods during QST and combining QST with questionnaires provides opportunity for further investigation of neuropathic pain disorders.

Neuropathic pain research is of recent inception and as a field is still evolving. This is reflected also in rudimentary efforts in translational pain research, which is an important engine for advancements in more mature areas of medical research, such as oncology and cardiology. Development of new tools, specifically biochemistry and genetics (59), as well as imaging as discussed earlier, provide new opportunities for the further advancements in pain research but complexity of neuropathic pain and especially its dynamic nature poses significant obstacle. Translational neuropathic pain research requires conceptual models that would respect complex and dynamic nature of neuropathic pain and guide progress, but those models are currently lacking. Attempts have been made to chart the course of translational pain research (60) but at this point in time there is no specific example of successful model of translational neuropathic pain research and specifically there is no model of translational pain research related to PDN.

There are a few obstacles that have been mentioned earlier, which pain research has to overcome and they include lack of clinical and pain translational research standards, lack of clear communication among basic scientists and among clinicians as well as between scientists and clinicians, and lack of standard in measurement tools that cross from bench to bedside and vice versa. However, there are strong efforts to advance communication and to develop methods relevant to translational neuropathic pain research.

Was this article helpful?

0 0
Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


  • Anke
    What behavior changes take place in a diabetic after a few beers?
    9 years ago

Post a comment