Is Transcranial Magnetic Stimulation of the Motor Cortex a Prognostic Tool for Motor Recovery After Stroke?
To the Editor:
The recent and important study from Pennisi et al1 has focused attention on the ability of a noninvasive and relatively simple technique, which can be carried out at the patient’s bedside and is potentially useful in gathering prognostic information on the future recovery of motor function, especially of the hand. Due to the growing amount of reports produced in this field,2 3 4 I believe that there should be some consensus on a few prerequisites for future studies in this area. This proposal is an attempt to limit the (only apparent) controversies of the related literature, which in my opinion are due mainly to selection and methodological bias.
It is a common experience that two groups of stroke patients, although similarly affected in hand function in the early days after the stroke, can recover in a quite different way. This might be ascribed to various reasons, including short-term phenomena that are known to disappear in a few days, such as the perilesional edema, and the functional block of still-living neurons and fibers (others, on the contrary, are going to die in the same period because of a cascade of mechanisms that lead to enlargement of the original volume of lesion); the prevalently cortical or subcortical site of lesion; the total volume of lesioned area; the interindividual variability in middle cerebral artery perfusion territory and/or topography of motor output to the hand muscles (including the number of multiple representations of the same muscles in separate clusters of motoneurons within the primary motor cortex); the presence and amount of ipsilateral corticospinal fibers;5 and the role of the unaffected (it should not be called “healthy”) hemisphere. By taking as an example the Pennisi et al study,1 most of the relevant information, as defined above, is missing on both clinical and neurophysiological grounds. We know only that the examined patients had complete hand palsy at the onset; nothing is written about the arm, shoulder, face, and leg muscles. More importantly, the transcranial magnetic stimulation (TMS) examination was not repeated immediately before discharge from the hospital, a period during which most of the short-term phenomena would have been resolved.
The motor evoked potential (MEP) was defined “normal” on the healthy—or rather the “unaffected”—side at days 1 and 365, but is this information really relevant? It is indeed known that MEPs which are exceptionally and progressively larger than normal can be elicited during follow-up from the unaffected hemisphere of stroke patients, when the affected hemisphere remains inexcitable.4 This hyperexcitability has been ascribed to the experience-dependent increase of synaptic efficiency in the unaffected motor cortex, which on its own might exert a depressive modulation on the affected motor areas via cortico-cortical, transcallosal connections;5 therefore, the hyperexcitability of the unaffected hemisphere might represent a bad prognostic indicator for recovery.
By positioning the stimulating coil (a large, nonfocal type) at the vertex, one misses the definition of the hot spot site, that is, the scalp position from which TMS triggers MEPs of largest amplitudes and minimal latencies with the lowest amount of stimulus intensity. There are international standards now available (to which one of the coauthors of the study by Pennisi et al has contributed) that properly define all these items for clinically oriented studies.6
It is also important to consider that most of the population in the study of Pennisi et al7 recovered very little at follow-up. Does this sample reflect the wide variety of recovery levels seen in the real life? In the literature as well as in the clinical experience of every neurologist there is a significant fringe of patients who recover within a few days or weeks from even a complete palsy of the hand. This type of patient, probably because of good recovery, does not come back as an outpatient. It would have been extremely interesting to have recruited a few cases with a better recovery as the final outcome, in order to test properly the prognostic validity of the method and to avoid the “floor effect” caused by the comparison of TMS with a undirectional phenomenon (no recovery or poor recovery), which is not representative of the variety of final outcomes following a stroke.
Very little is written about the drugs used in the tested patients, despite the fact that some drugs are known to interfere with the excitability of motor cortex.
In conclusion, I feel that the TMS has accumulated sufficient scientific credence to be proposed as a relevant tool in predicting the functional outcome from stroke in the very early stages. However, to avoid an accumulation of literature with little or no clinical repercussion (as has been the case until now), there should be an effort to recruit a larger number of patients (multicenter studies?) in a series that is designed well in statistical terms, is truly representative of the general population affected by stroke, and is tested with the appropriate methodology.
- Copyright © 2000 by American Heart Association
Pennisi G, Rapisarda G, Bella R, Calabrese V, Maertens de Noordhout A, Delwaide PJ. Absence of response to early transcranial magnetic stimulation in ischemic stroke patients. Prognostic value for hand recovery. Stroke.. 1999;30:2666–2670.
Heald A, Bates D, Cartlidge NEF, French JM, Miller S. Longitudinal study of central motor conduction following stroke, II: central motor conduction measured within 72 hours after stroke as a predictor of functional outcome at 12 months. Brain.. 1993;116:1371–1385.
Catano A, Houa M, Caroyer JM, Ducarne H, Noel P. Magnetic transcranial stimulation in non-haemorragic sylvian strokes: interest of facilitation for early functional prognosis. Electroencephalogr Clin Neurophysiol.. 1995;97:349–354.
Cicinelli P, Traversa R, Rossini PM. Post-stroke reorganization of brain motor output to the hand: 2–4 months follow-up with focal magnetic transcranial stimulation. Electroencephalogr Clin Neurophysiol.. 1997;105:438–450.
Andrews RJ. Transhemispheric diaschisis. Stroke.. 1991;22:943–949.
Rossini PM, Barker AT, Berardelli A, Caramia MD, Caruso G, Cracco RQ, Dimitrijevic MR, Hallett M, Katayama Y, Lucking CH, Maertens de Noordhout AL, Marsden CD, Murray NMF, Rothwell JC, Swash M, Tomberg C. Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application: report of an IFCN committee. Electroencephalogr Clin Neurophysiol.. 1994;91:476–482.
The comment by Dr Rossini contains some very helpful suggestions for future studies of the prognostic value of TMS in stroke but also questions the methodology used in our study. Indeed, the electrophysiological study was not repeated after a few weeks, but this was not crucial here, because the primary aim was to determine whether early TMS (within days from stroke) was useful in predicting long-term motor outcome. Of course, if mechanisms involved in the clinical recovery are to be investigated in detail, TMS sessions and clinical assessments need to be repeated over time. Also, use of the focal stimulating coil is crucial in this respect, but this was not the aim of this work. Large circular coils are powerful, widely available, and convenient for most clinical applications of TMS, and they are certainly sufficient to determine whether early motor responses are present or not. If a method of investigation becomes complicated and time consuming, it is unlikely to be adopted by clinicians, except in a few highly specialized centers. Our study focused on hand strength and dexterity, because of its functional importance in humans and also because hand muscles are particularly easy to activate with TMS, even under difficult conditions. For the sake of conciseness, motor function of other areas was not detailed in the paper, but from global clinical scores (National Institutes of Health Stroke Scale) it is clear that the population was rather homogeneous at stroke onset and that global deficits were moderate to severe. Indeed, the population sample was limited, so that ultimate recovery of our patients may not reflect the outcome of the general population. However, one of the selection criteria was absence of MEP in the affected hand within 48 hours from stroke. We know from previous studiesR1 R2 that patients with persistent responses early after stroke recover better, so that simply this selection criterion could have influenced the final outcome. As for drugs used, we generally avoid in stroke patients the administration of central nervous system depressants, which can alter the level of consciousness and hamper recovery and rehabilitation. Moreover, to rule out effects of drugs, associated diseases, or level of arousal on corticospinal excitability; only patients with normal responses on the “unaffected” side were included. Finally, we believe that the role of ipsilateral corticospinal fibers in the process of recovery remains purely speculative, because they represent <10% of motor cortex output, the majority of which ultimately crosses in the segmental cord,R3 and that many other mechanisms must be considered.
Catano A, Houa M, Caroyer JM, Ducarne H, Noel P. Magnetic transcranial stimulation in nonhaemorrhagic sylvian strokes: interest of facilitation for early functional prognosis. Electroencephalogr. Clin. Neurophysiol.. 1995;97:349–354.
Rapisarda G, Bastings E, Maertens de Noordhout A, Pennisi G, Delwaide PJ. Can motor recovery in stroke patients be predicted by early transcranial magnetic stimulation? Stroke.. 1996;27:2191–2196.
Porter R, Lemon RN. Corticospinal Function and Voluntary Movement. Oxford, UK: Clarendon Press; 1993.