Published online before print April 17, 2003, doi: 10.1161/01.STR.0000071140.00153.05
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2003;34:e26.)
© 2003 American Heart Association, Inc.
Research Reports |
Editorial Comment—Functional MRI: A Potential Physiologic Indicator for Stroke Rehabilitation Interventions
Department of Neurology, David Geffen School of Medicine of the University of California Los Angeles, Neurologic Rehabilitation and Research Program, Reed Neurologic Research Center, Los Angeles, California
 |
Introduction |
|---|
 Top Introduction References |
|---|
A patient recovering from a modest hemiparesis from stroke alternates between rest and tapping the affected index finger1 or gripping a transducer with visual feedback about the force exerted2 during functional MRI (fMRI). Do the evoked patterns of cerebral activity reveal the reorganization and rehabilitation of the motor mind? They may, if considered within the context of potentially confounding technical, statistical, anatomical, experiential, and task-dependent factors.3,4
What Are the General Determinants of fMRI Patterns of Activation Induced By a Movement?
Maps of functional anatomy obtained using the blood-oxygen–level-dependent (BOLD) technique depend on the spatial extent of metabolic and hemodynamic changes induced by local synaptic activity and local field potentials, but do not precisely correlate with such activity.5 Localization and spatial resolution of neuronal activity may be confounded by a range of signal-dependent factors. These include BOLD-dependent capillary density and draining veins, perfusion of hemodynamically compromised tissue, links between one active population of neurons to others,6 possible differences in the BOLD signal caused by presynaptic inhibition compared with excitation, and fine differences between subjects in the location of regions of interest.7 fMRI methods are not a done deal. Some controversy and room for error accompany every aspect of data acquisition and analysis. Choices are made about the MR sequences for scans, data smoothing and correction schemes, registration of activations onto anatomical space especially when an infarct distorts morphology, modeling choices for the statistics represented by colored voxels of activity, statistical inferences, and approaches to individual subject versus group analyses. In addition, physiological factors such as sleep deprivation and estrogen levels and drugs such as caffeine may alter the BOLD signal from one day to another in the same subject. Any agent or injury that affects neurotransmitters, especially widely projecting neuromodulators such as dopamine,8,9 acetylcholine,10 norepinephrine,11 and serotonin12 may alter large-scale synaptic excitability and, in turn, the BOLD response.
The primary sensory (S1) and motor cortex (M1) is the most common region of interest studied. Other sensorimotor areas, some of which project bilaterally, also contribute to the corticospinal control of movement (Table). Some or all regions are activated by a motor task, more so as task demands increase. A hemiparesis will add to task difficulty. Performance, then, depends in part on the relative sparing of the nodes included in the Table in relation to task requirements. Greater knowledge about spared sensorimotor projections would help put activations into the context of structure and function. The integrity of descending and ascending projections of parallel fiber arrays may come to be better appreciated using diffusion tensor imaging algorithms that estimate the anatomical connectivity between the cortex and subcortical white matter tracts.13 The colorful maps of activity, however, may not offer predictable patterns, even if the volume of spared neurons can be estimated. Gains in motor skills and fMRI activity also depend on the subject’s history of practice, as well as handedness, sex, education, motivation, age-related prior brain and biomechanical adaptations, and the age-related capacity for plasticity.2,14 These experiential and genetic differences, along with various adaptive and feedback strategies for motor control,15,16 create a plethora of ways for cortical assemblies to compute a movement. The motor system is also part of a network activated by anticipation, imagination, observation, and execution of an action.17,18 During an imaging study, the subject’s state of mind is unknown to the investigator, which may compromise the resting or active state.
|
The typical fMRI study after stroke has another interpretive confound, especially for rehabilitation. The investigator arbitrarily chooses a doable movement at some point in time. The movement may not reflect the skill needed for tasks that are important to patients. Also, the properties of cortical sensory and motoneurons adapt quickly to new learning experiences.19 Improving task-specific skills by a hemiparetic subject may be associated with a progressive or fluctuating change in the cortical representations for the task. One study, or several serial scans, then, may only capture a work in progress.
What Conclusions Can Be Drawn From Movement-Related Activation Studies After Stroke?
In studies of patients who regain most of their ability to tap fingers or open the hand, fMRI reveals a ventral expansion of neuronal assemblies in S1M1 representing the hand toward those for the face. For ankle dorsiflexion, the representation expands toward the proximal leg and back muscles.20 These activations may later decrease with behavioral gains. Activation is often prominent along the rim of a cortical infarct, which may be a bastion for long-term potentiation soon after stroke.21
Animal22 and human studies of cerebral reorganization with fMRI also reveal a shift of activation to the contralesional cortex soon after a stroke, somewhat in parallel to the extent of tissue loss. As functional use of the hand improves, activity shifts more to ipsilesional S1M1, if it is relatively spared.23,24 At least for subcortical stroke, worse behavioral outcomes for the hand correlate with a shift in the balance of activation toward the contralesional S1M1 with greater ipsilesional M1 injury.24,25 In this issue of Stroke, Zemke and colleagues find that reorganization within S1M1 and premotor areas differs, depending on the side of the stroke and whether minimal or mild impairments persist.26 The investigators controlled their motor task for force, range of joint motion, speed, muscles activated, and attention, so the data are especially likely to reflect a change in cortical function, rather than reflect a different movement strategy. Other studies reveal that less premotor activity and higher contralesional activity in the cerebellar hemisphere correlate with greater motor control, consistent with sparing of corticopontocerebellar projections.27 A posterior extension of peak activation in S1M1 of uncertain relevance to behavioral gains has been noted in some subjects28 but may reflect a compensatory sensory drive for movement. The limitations of the BOLD signal and analytic techniques mean that signals from reticulospinal inputs to propriospinal premotoneurons for control of a synergistic grasp29 and rubrospinal contributions to individuated finger and feeding movements30 will not be detected. Patterns of activation, then, seem to ride on the individual subject’s ability to recruit residual portions of a bilateral motor network, driven in part by residual sensory feedback related to task performance.
Can We Improve on Our Understanding of Activation Patterns?
Theory-driven, experimental evidence is rapidly improving fMRI acquisition and analytical methodologies.31–34 Correlations between representational or network changes and behavior require continuing careful study.
Learning-dependent plasticity is a function of the novelty, intensity, duration, and specificity of the movement skill that is practiced. As rehabilitation practice proceeds after stroke, fMRI could be used to reveal both the capacity and progress of experience-dependent changes in the interactive nodes of the motor network. That would increase our understanding of the meanings of patterns of activity over time. Indeed, a great value of fMRI for rehabilitation could be to serially monitor functional anatomical changes over the course of a physical, cognitive, pharmacologic, or, one day, biologic repair intervention.3 Pushing the potential for gains after stroke by continuing a specific therapy, until no further behavioral improvements and no further evolution in the activity of regions of interest result, may also ensure that the intervention has been maximized and that the reorganizational potential of the regions studied has peaked, at least as far as fMRI can discern.35 Used this way, fMRI patterns may provide a physiological indicator to monitor and eventually to predict the utility of a rehabilitative intervention for patients.
|
Acknowledgments |
|---|
Support was provided by NIH grants HD39629 and HD07479. I thank my colleagues at the Brain Mapping Center at UCLA.
|
References |
|---|
|
TopIntroductionReferences |
|---|
1. Cramer S, Nelles G, Schaechter J, Kaplan J, Finklestein S, Rosen B. A functional MRI study of three motor tasks in the evaluation of stroke recovery. Neurorehabil Neural Repair. 2001; 15: 1–8.
2. Ward N, Frackowiak R. Age-related changes in the neural correlates of motor performance. Brain. 2003; 126: 873–888.
3. Dobkin B. The Clinical Science of Neurologic Rehabilitation. New York, NY: Oxford University Press; 2003.
4. Toga A, Mazziotta J, eds. Brain Mapping: The Systems. San Diego, Calif: Academic Press; 2000.
5. Logothetis N, Pauls J, Augath M, Trinath T, Oeltermann A. Neurophysiological investigation of the basis of the fMRI signal. Nature. 2001; 412: 150–157.[CrossRef][Medline] [Order article via Infotrieve]
6. Ugurbil K, Toth L, Kim DS. How accurate is magnetic resonance imaging of brain function? Trends Neurosci. 2003; 26: 108–14.[CrossRef][Medline] [Order article via Infotrieve]
7. Rademacher J, Burgel U, Geyer S, Schormann T, Freund H-J, Zilles K. Variability and asymmetry in the human precentral motor system: a cytoarchitectonic and myeloarchitectonic brain mapping study. Brain. 2001; 124: 2232–2258.
8. Ziemann U, Bruns D, Baudewig J, Paulus W. Changes in human motor cortex excitability induced by dopaminergic and anti-dopaminergic drugs. EEG Clin Neurophysiol. 1997; 105: 430–437.[CrossRef][Medline] [Order article via Infotrieve]
9. Mattay V, Tessitore A, Callicott J, et al. Dopaminergic modulation of cortical function in patients with Parkinson’s disease. Ann Neurol. 2002; 51: 156–164.[CrossRef][Medline] [Order article via Infotrieve]
10. Juliano S. Mapping the sensory mosaic. Science. 1998; 279: 1653–1654.
11. Plewnia C, Hoppe J, Hiemke C, Bartels M, Cohen LG, Gerloff C. Enhancement of human cortico-motoneuronal excitability by the selective norepinephrine reuptake inhibitor reboxetine. Neurosci Lett. 2002; 330: 231–234.[CrossRef][Medline] [Order article via Infotrieve]
12. Loubinoux I, Pariente J, Boulanouar K, et al. A single dose of the serotonin neurotransmission agonist paroxetine enhances motor output: double-blind, placebo-controlled, fMRI study in healthy subjects. Neuroimage. 2002; 15: 26–36.[CrossRef][Medline] [Order article via Infotrieve]
13. Ciccarelli O, Parker G, Toosy A, et al. From diffusion tractography to quantitative white matter tract measures: a reproducibility study. Neuroimage. 2003; 18: 348–359.[CrossRef][Medline] [Order article via Infotrieve]
14. Hutchinson S, Kobayashi M, Horkan CM, Pascual-Leone A, Alexander MP, Schlaug G. Age-related differences in movement representation. Neuroimage. 2002; 17: 1720–1728.[CrossRef][Medline] [Order article via Infotrieve]
15. Todorov E, Jordan M. Optimal feedback control as a theory of motor coordination. Nat Neurosci. 2002; 5: 1226–1235.[CrossRef][Medline] [Order article via Infotrieve]
16. Gribble P, Scott S. Overlap of internal models in motor cortex for mechanical loads during reaching. Nature. 2002; 417: 938–941.[CrossRef][Medline] [Order article via Infotrieve]
17. Jeannerod M. Neural simulation of action: a unifying mechanism for motor cognition. Neuroimage. 2001; 14: S103–S109.[CrossRef][Medline] [Order article via Infotrieve]
18. Koski L, Wohlschlager A, Bekkering H, et al. Modulation of motor and premotor activity during imitation of target-directed actions. Cereb Cortex. 2002; 12: 847–855.
19. Taylor D, Helms Tillery S, Schwartz A. Direct cortical control of 3D neuroprosthetic devices. Science. 2002; 296: 1829–1831.
20. Dobkin B. Spinal and supraspinal plasticity after incomplete spinal cord injury: correlations between functional magnetic resonance imaging and engaged locomotor networks. In: Seil F, ed. Progress in Brain Research.Vol. 128. Amsterdam: Elsevier; 2000: 99–111.
21. Dobkin B. Activity-dependent learning contributes to motor recovery. Ann Neurol. 1998; 44: 158–160.[CrossRef][Medline] [Order article via Infotrieve]
22. Dijkhuizen RM, Singhal AB, Mandeville JB, et al. Correlation between brain reorganization, ischemic damage, and neurologic status after transient focal cerebral ischemia in rats: a functional magnetic resonance imaging study. J Neurosci. 2003; 23: 510–517.
23. Marshall R, Perera G, Lazar R, Krakauer J, Constantine R, DeLaPaz R. Evolution of cortical activation during recovery from corticospinal tract infarction. Stroke. 2000; 31: 656–661.
24. Feydy A, Carlier R, Roby-Brami A, et al. Longitudinal study of motor recovery after stroke: recruitment and focusing of brain activation. Stroke. 2002; 33: 1610–1617.
25. Calautti C, Leroy F, Guincestre J-Y, Marie R-M, Baron J-C. Sequential activation brain mapping after subcortical stroke: changes in hemispheric balance and recovery. Neuro Report. 2002; 12: 3883–3886.
26. Zemke A, Heagerty P, Lee C, Cramer S. Motor reorganization after stroke is related to side of stroke and level of recovery. Stroke. 2003; 34: e23–e26.[CrossRef]
27. Small S, Hlustik P, Noll D, Genovese C, Solodkin A. Cerebellar hemispheric activation ipsilateral to the paretic hand correlates with functional recovery after stroke. Brain. 2002; 125: 1544–1557.
28. Pineiro R, Pendlebury S, Johansen-Berg H, Matthews P. Functional MRI detects posterior shifts in primary sensorimotor cortex activation after stroke: evidence of local adaptive reorganization? Stroke. 2001; 32: 1134–1139.
29. Mazevet D, Meunier S, Pradat-Diehl P, Marchand-Pauvert V, Pierrot-Deseilligny E. Changes in propriospinally mediated excitation of upper limb motoneurons in stroke patients. Brain. 2003; 126: 988–1000.
30. Iwaniuk A, Whishaw I. On the origin of skilled forelimb movements. Trends Neurosci. 2000; 23: 372–376.[CrossRef][Medline] [Order article via Infotrieve]
31. Friston K, Glaser D, Henson R, Kiebel S, Phillips C, Ashburner J. Classical and Bayesian inference in neuroimaging: applications. Neuroimage. 2002; 16: 484–512.[CrossRef][Medline] [Order article via Infotrieve]
32. Svensen M, Kruggel F, Benali H. Independent components analysis of fMRI group study data. Neuro Image. 2002; 16: 551–563.[Medline] [Order article via Infotrieve]
33. Lazar N, Luna B, Sweeney J, Eddy W. Combining brains: a survey of methods for statistical pooling of information. Neuroimage. 2002; 16: 538–550.[CrossRef][Medline] [Order article via Infotrieve]
34. Andersen A, Gash D, Avison M. Principal component analysis of the dynamic response measured by fMRI: a generalized linear systems framework. Magn Reson Imaging. 1999; 17: 795–815.[CrossRef][Medline] [Order article via Infotrieve]
35. Dobkin B, Sullivan K. Sensorimotor cortex plasticity and locomotor and motor control gains induced by body weight-supported treadmill training after stroke. Neurorehabil Neural Repair. 2001; 15: 258.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||