Evidence for Cortical Plasticity with MS

MS may be a particularly appropriate disease for the study of brain plasticity (see Cifelli and Matthews, 2002), as most patients show good recovery from clinical expression of new lesions (relapses) in the earlier stages of the illness, despite evidence for axonal injury with each attack (Fu et al., 1998; Evangelou et al., 2000). Some aspects of axonal injury may be partially reversible (de Stefano et al., 1995), which must contribute to the potential for clinical recovery. However, irreversible axonal injury can be substantial (Evangelou et al., 2000; Trapp et al., 1998), suggesting that other factors, such as adaptive functional reorganization, also must be important (Cifelli et al., 2002).

Yousry et al. (1998) first provided fMRI-based evidence suggesting that patients with MS who have motor weakness show potentially adaptive increased activation of ipsilateral and accessory motor areas, reminiscent of earlier findings with subcortical ischemic stroke (Weiller et al., 1993). Later studies have refined these observations and related patterns of functional change directly to measures of injury or disability (see, for example, Filippi et al., 2003; Rocca et al., 2002; Lee et al., 2000; Reddy et al. 2000c, 2000d; Pantano et al., 2002).

To interpret these changes in brain functional organization as evidence for adaptive plasticity related to disease, at least three conditions need to be fulfilled:

1. The extent of the functional reorganization must be related to the burden of disease (and thus evolve dynamically through the course of the disease).

2. Evidence for functional reorganization must be found even in the absence of disease-associated behavioral impairment (consistent with the notion that adaptive plasticity limits clinical expression of the disease).

3. Altered patterns of functional activation in patients must be independent of conscious patterns of recruitment (as the latter are associated with development of strategies for compensation).

Ideally, it may also be possible to demonstrate that ascribing a relevant, adaptive functional role to the new pattern of brain activation is physiologically reasonable. Not all functional brain changes are either directly relevant to the task or desirable. Brain functional changes simply may be epiphe-nomena (e.g., a result of sensory stimulation by an ambulatory aid while a hemiparetic patient walks) rather than a direct manifestation of altered central programs for motor control. Functional changes can be incidental (e.g., the perceived need to attend more closely to cues if movement is impaired). Brain functional changes also can be maladaptive. For example, the intense training used by professional musicians may lead to focal hand dystonia, in which representations for digit movements become linked, leading to awkward or frankly uncomfortable finger postures (Pujol et al., 2000).

1. Functional Brain Changes in Patients with MS Are Related to Disease Burden

Functional reorganization defined by fMRI in patients with MS is related to the burden of disease. Using a simple hand movement task, Lee et al. (2000) first reported that brain T2 lesion load was correlated with a decreasing relative activation in motor cortex ipsilateral to the hand moved (Fig. 4). For a more specific test, the need to use measures of pathology in the corticospinal tracts (CSTs) is suggested by previous work with focal ischemic brain injury, which established direct correlations between motor functional impairment and damage to the CST (Feydy et al., 2002; Binkofski et al., 1996; Fukui et al., 1994). Reddy et al. (2000c) demonstrated a correlation between NAA decreases (a measure of axonal injury) in the corticospinal tract and changes in patterns of functional activity with hand movement in patients who did not have clinical deficits affecting the limb tested. Pantano et al. (2002) showed a similar correlation between changes in the pattern of brain activation and lesion load pathology in the CST of patients with MS after a first episode of hemiparesis. They confirmed that, with increasing time after clinical presentation (during which disease burden may be expected to rise), there was an increasingly abnormal pattern of brain activity with hand movement. Similar, more intensive correlations with pathology have been reported by the Filippi group in patients with relapsing-remitting (Filippi et al., 2003) and primary-progressive MS (Rocca et al., 2002), although this approach has been more sensitive to changes elsewhere than in the ipsilateral motor cortex.

Aspects of this phenomenon can be generalized to other systems. Staffen et al. (2002) studied patients with MS and healthy controls performing a sustained attention task. Although performance was equivalent between the two

Figure 4 (A) With simple finger flexion, the major regions of brain activation in healthy subjects are the sensorimotor cortex contralateral (C) to the hand moved and the midline supplementary motor area. Patients with multiple sclerosis and corticospinal tract injury show increased relative ipsilateral (I) sensorimotor cortex activation. (B) This difference is highlighted by an activation hemispheric lateralization index (LI), defined as the ratio of sensorimotor cortex activation contralateral to that ipsilateral to the hand moved. The LI decreases (i.e., activation becomes more bihemi-spheric) as disease burden rises, consistent with the notion that ipsilateral sensorimotor cortex recruitment represents cortical adaptive plasticity (see Johansen-Berg et al., 2002b; figure from Lee et al., 2000.)

Figure 4 (A) With simple finger flexion, the major regions of brain activation in healthy subjects are the sensorimotor cortex contralateral (C) to the hand moved and the midline supplementary motor area. Patients with multiple sclerosis and corticospinal tract injury show increased relative ipsilateral (I) sensorimotor cortex activation. (B) This difference is highlighted by an activation hemispheric lateralization index (LI), defined as the ratio of sensorimotor cortex activation contralateral to that ipsilateral to the hand moved. The LI decreases (i.e., activation becomes more bihemi-spheric) as disease burden rises, consistent with the notion that ipsilateral sensorimotor cortex recruitment represents cortical adaptive plasticity (see Johansen-Berg et al., 2002b; figure from Lee et al., 2000.)

groups, patients with MS showed activation in right frontal and left parietal cortex not found in healthy controls, suggesting adaptive recruitment of additional, multimodal control regions. Werring et al. (2000) examined patients with MS who had recovered from optic neuritis using visual stimulation with fMRI. Although the patients showed decreased activation of the visual cortex receiving projections from the affected optic nerve, there were additional extensive task-associated decreases in activation in the claustrum, the lateral temporal and posterior parietal cortices, and the thalamus. All of these brain regions are involved in higher order visual processing or object recognition. The extent to which activity in these additional regions was suppressed was related to the severity of optic nerve pathology as assessed from visual-evoked potentials. A speculative explanation for this observation is that suppression of activity in these areas is a manifestation of adaptively increased atten-tional mechanisms, which may enhance information processing specifically in primary visual cortex immediately after presentation of the visual stimulus.

2. Functional Brain Changes in Patients with MS Are Dynamic

As predicted from evidence that brain activity patterns change with disease burden, functional reorganization observed by fMRI is dynamic, changing as lesions evolve pathologically. In an early report by Reddy et al. (2000d), a patient with a new large demyelinating lesion affecting the corticospinal tract was monitored with serial MRI, MRS, and fMRI studies after the onset of hemiparesis. Clinical recovery preceded normalization of NAA concentrations (an index of structural repair after the lesion) and was associated with relative increases in premotor area and supplementary motor area activation in the hemisphere ipsilateral to the hand moved (an index of functional adaptation). The observation that clinical recovery occurred faster than the structural repair, and was associated with changing patterns of brain activation, is consistent with a functional role for the latter. The general correlation between changes in fMRI responses and magnetic resonance spectroscopy (MRS) measures suggests a progressive normalization of patterns of functional organization in motor cortex in response to the repair of axonal injury associated with the lesion. Qualitatively similar dynamic changes occur after focal ischemic stroke as recovery progresses (Marshall et al., 2000; Ward et al., 2003), Results are consistent with prior cross-sectional studies demonstrating that the magnitude of activation changes was greater with poorer functional recovery (Calautti and Baron, 2003; Johansen-Berg et al., 2002a; Jang et al., 2003).

A key point is that brain functional changes associated with movement are (at least to a large degree) independent of the specific cause of neuroaxonal damage in the CST. The pattern of change discussed previously for MS is similar to that found with serial fMRI studies of motor activation in patients with stroke. These studies have shown a similar initial bilateral activation with lesions of the contralateral motor cortex that reverts toward a more lateralized activation pattern with progression of recovery (Ward et al., 2003; Marshall et al., 2000; Feydy et al., 2002). Altered patterns of recruitment may help to maintain functions (e.g., via adaptive recruitment of additional descending motor pathways) despite persistent injury to the corticospinal tract.

Aspects of these potentially adaptive functional changes may occur rapidly, suggesting that some aspects may rely on unmasking preexisting pathways. Parry and colleagues (2003) studied brain responses to the Stroop paradigm, a test of attention and prepotent inhibition. Patients with MS and healthy controls were studied using a fMRI counting Stroop task. The two subject groups had comparable performances, but patients showed a deficit of activation in a right frontal region (including Brodman's area 45 and the basal ganglia) implicated by modeling studies in key aspects of the task (Cohen et al., 1996). The patients also showed relatively increased activation in a predominantly left medial pre-frontal region previously implicated in supramodal monitoring and related tasks. A direct relationship between these activation measures and relative brain atrophy was consistent with the notion that left prefrontal activation represented a functional adaptation to the deficits. Parry et al.

(2003) tested the effects of acute administration of rivastig-mine, a central cholinesterase inhibitor, on the patterns of brain activation. It was hypothesized that this drug might be able to partially correct functional impairment in the dysfunctional right hemisphere circuit. Within 90 minutes of oral drug administration, there was a relative normalization of the abnormal, Stroop-associated brain activation in all of the patients with MS (with no change in the patterns of brain activation in any of the healthy controls).

The potential time-frame for adaptive functional changes after injury has been reduced further by studies in the healthy brain. For example, Butefisch demonstrated that changes in movement representation for the thumb can occur over training periods as short as minutes (Butefisch et al., 2000). Strens et al. (2002) demonstrated that TMS delivered to the motor cortex to locally disrupt activity is associated with increased excitability in the motor cortex of the opposite hemisphere. Similarly, Lee et al. (2003) used TMS to interfere with activity in the primary motor cortex while monitoring the metabolic response of the entire motor system using PET. Although the TMS parameters used (30 minutes of slow 1-Hz stimulation) should have decreased excitability of the left motor cortex, no behavioral consequences of the disruption were observed with a simple motor task, suggesting that unaffected regions of the motor network were able to compensate for the effects of disruption in the primary motor cortex. A candidate adaptive change was identified in PET scans acquired after the stimulation showing that, after repetitive TMS interference to the left primary motor cortex, increased activation of the opposite premotor cortex was seen with right hand movement. This recalls the interhemispheric functional reorganization that occurs in the premotor cortex after corticospinal tract damage (see previously).

3. Functional Brain Changes with MS Are Found in Patients Without Clinical Deficits Limiting Performance

The second criterion for adaptive functional reorganiza-tion—that it can be found even without behavioral changes—has also been demonstrated. Changes in cortical activation occur early in the course of MS and in patients without symptoms in the affected functional system. Patients with MS without motor or sensory impairment of the upper limbs were investigated with fMRI and MRS using a hand-tapping paradigm (Reddy et al., 2000c). LI was abnormally low in the patients and decreased progressively, with decreases in relative brain NAA concentrations. Because these results were obtained from patients who had normal hand function, the potential confound that arises from performance differences in comparison with healthy controls was absent. Patients with optic neuritis as their only clinical manifestation of MS also show brain functional changes with movement compared to healthy controls, suggesting that even low levels of injury in the motor system are associated with adaptive functional changes (Pantano et al., 2002). Again, emphasizing that the changes are not determined solely by the pathology, qualitatively similar changes were noted in the well-recovered patients with capsular infarcts studied by PET more than a decade ago (Weiller et al., 1993).

4. Functional Brain Changes with MS Occur Independent of Conscious Control

The third criterion for adaptive brain plasticity—that brain functional changes are independent of volitional activity—can also be confirmed in some patients. Because fMRI assesses predominantly presynaptic activity (see previously) and there is strong afferent sensory input into motor cortical regions, parts of the cortical network for limb movement (including primary sensorimotor and pre-motor cortex) should be defined by patterns of brain activity with passive movement of a limb (Sahyoun et al., 2003; Reddy et al., 2001; Weiller et al. 1996). A passive movement task, therefore, can be used to test for abnormal patterns of brain activity in patients relative to healthy controls. Although only some components of the network for motor control can be defined, the use of a passive task has the advantage of being free from potential confounds arising from differences in movement preparation and planning (Reddy et al., 2001; Sahyoun et al., 2003). Reddy et al. (2002) tested a group of patients with MS using both an active and a passive hand movement task, verifying with surface electromyography that the passive task was not associated with motor unit activation. A strong, quantitative correlation was found between indices of relative activation with the active and passive tasks. Separate studies confirmed that the activation was independent of movement imagery, and that it is absent in patients with pure sensory neuropathies who lack hand perception or somatic sensation. The recruitment of ipsilateral motor cortex during movement in MS pathology, therefore, is not dependent on factors related to volition.

5. Demonstrating a Functional Role for the Brain reorganization

In general, behaviorally relevant roles for functional changes such as those found in patients with MS have not been directly established. The observation of clinical recovery in the absence of resolution of structural changes in a functional system provides indirect evidence that functional reorganization is adaptive. More commonly, the rationale is one of physiological reasonableness. For example, increased sensory cortical activity may be expected as an adaptation to the need to alter motor representation with corticospinal tract injury affecting the hand (Rocca et al., 2002). More evidence is available regarding the potential role of increased activity in motor cortex ipsilateral to the hand moved after injury to the CST. Clinical observations have suggested that sensorimotor cortex opposite to a focal lesion affecting motor control may support important new functions. For example, Fisher (1992) described two patients whose functional deficits deteriorated after partial recovery from a second stroke in the other hemisphere and Ago et al. (2003) reported a patient with a right hemisphere stroke and left hemiparesis who sustained deterioration of residual function after a new lacunar infarct in the left corona radi-ata. Transient cortical "lesioning" using TMS also supports this notion. Johansen-Berg et al. (2002b) described measurable functional impairments in cued finger movements after TMS pulse delivery to the dorsal premotor cortex contralateral to a focal ischemic lesion. The extent of this impairment after TMS was correlated with relative activity in the ipsi-lateral motor cortex measured using fMRI.

Your Heart and Nutrition

Your Heart and Nutrition

Prevention is better than a cure. Learn how to cherish your heart by taking the necessary means to keep it pumping healthily and steadily through your life.

Get My Free Ebook

Post a comment