neuro(ipeport ‘Tue present study aims to provide neurophysiological evidence of ipsilateral activation daring moter recovery in patients alter stroke. The effects of cortical reorgani- zation were investigated using magnetic brain stimula- tion in order to record motor evoked potentials (MEPs) both from contralateral and ipsilateral hands. Ten healthy subjects and 13 patients were examined. The patients had suffered thet frst hemispheric stroke and consequent motor deficit. While ipsilateral responses (MEPs) were absent in normal subjects, they were obtained from both ipsilateral and contralateral hands in patients, The ipsilateral MEP differed from contralat- tral MEP in the fllowing respects (1) elicitation during contraction; (2) a shorter latency; (3) a lower amplitude, ‘The presence of optimal iMEPs (lower excitability threshold, larger amplitude) in recovered hands points to's role for the undamaged hemisphere, and in parti ular to the involvement of secondary motor areas. We suggest that IMEPs in general may bea marker of brain lasticty, and that the specific type described here appear in the context of fast autochthonous motor recovery. Key Words: Motor recovery; Stroke; Magnete stimulation; Plasticity; Ipsilateral MEPs Neurophysiology, Basic and Clinical NeuroReport 7, 1756-1760 (1996) Cerebral plasticity after stroke as revealed by ipsilateral responses to magnetic stimulation 1,24 Cesare lani' and ‘Clinica Neurologica, Universita di Roma Tor Vergata, Ospedale S. Eugenio, Pile dell’Umanesimo 10, 00144, Roma; IRCCS S. Lucia, Rome. Italy ‘ACorresponding Author Introduction Varying degrees of motor function improvement are frequently observed in patients affected by stroke, however the mechanisms underlying the recovery process are not yet fully identified. Changes in the functional anatomy underlying motor recovery have been investigated by clinical, clectrophysiological, metabolic (PET) and Doppler methods, and by experiments on animals." There is general agreement regarding one key aspect, namely the role played by the undamaged hemisphere, with reorganization of motor areas ipsilateral to the lesion side. In partic- ular, PET studies on patients who have recovered from stroke have revealed bilateral activation of cortical motor areas during voluntary contraction of the affected hand, as well as the recruitment of secondary motorcortical areas 2 In the present study, magnetic brain stimulation was used in order to obtain a deeper understanding of the nature of the changes in cortical reorganiza- tion which might be the neurophysiological coun- terpart of both PET and recent Doppler‘ findings concerning ipsilateral activation. The value of ic stimulation lies in its sensitivity and in the possibility of applying a wide range of parameters for the evaluation of motor evoked responses (excitability threshold, silent period, amplitude, latency onset). For this purpose, stroke patients with hand motor deficits who were in the course of a rapid recovery were selected. 1756 Vol 7 No 11 29 July 1996 Subjects and Methods Thirteen fully informed patients (eight men and five women; all right handed, age range 41-81 years) ‘were investigated during clinical recovery of motor function, 3-15 days after the first occurrence of hemi- spheric stroke producing a predominantly facio- brachial hemiparesis. Symptoms persisted for at least 72h. All patients underwent considerable sponta- neous motor recovery without physiotherapeutic support. The clinical picture was assessed and scored ian Scale (Table 1). nts were also examined in the acute stage, within 24 hours of the stroke. In each case brain CT scan or MRI revealed a single hemispheric lesion located in the territory of the middle cerebral artery (Table 1). Nine healthy right handed volunteers and one left corrected handed, aged 25-60 years, represented the control population. The procedures utilized in the study were approved by the Local Ethical Committee. Magnetic stimulation: Stimulation of the motor cortex was performed by a Cadwell MES 10 connected to a regular round coil. Stimuli were administered alternately over both normal and damaged hemispheres, with the coil tangentially placed over the hand motor area.* First recordings ‘were obtained using surface electrodes applied to intrinsic hand muscles (APB or FDI). The cortical © Ropid Science PublishersCerebral plasticity after stroke neurofpeport threshold intensity for liciting contralateral responses was determined during relaxation’ in which state ipsilateral MEPs were absent. In order to obtain these, recordings were acquired during contraction at increasing stimulation intensities, until reproducible responses could be collected. Traces, averaged and replicated at least three times with suitable calibration, were acquired with a Multibasis Esaote with a filter bandwidth and sampling rate of 20-2000 Hz and 10 KHz, respec- tively, and stored on floppy disks for further analysis. Data analysis: Both ipsilateral and contralateral recordings were evaluated according to the following parameters: (j) excitability threshold measured in percentage of the stimulator output, during muscle relaxation on the contralateral side and during contraction ipsilaterally. (ii) MEP latency, measured. at the beginning of the first reproducible negative deflection and at threshold intensity during contrac- tion. (iii) MEP amplitude, measured peak to peak during contraction. (jv) Silent period duration, considered as the EMG silence interposed between the end of the MEP and the resumption of the back- ground muscular activity, measured at 80% of the stimulator output. Statistical analysis was undertaken using arithmetic means with standard deviations (SDs), paired comparisons with Student's t-test for interhemi- spheric differences recorded in patients and Student's t-test between patients and control subjects. Results Patients’ clinical scores, means and statistical eval- uation of data are given in Tables 1 and 2, respec- tively. Table 2 also includes normative values and measurements of MEPs recorded in the traditional way, contralateral to the side of cortical stimulation. In none of the right handed control subjects could ipsilateral MEPs be recorded during rest as well as during contraction, even at a stimulation intensity of 100%. In accordance with previous reports,! ipsilat- eral TCS (delivered at the optimal position for elic- iting contralateral MEPs) was able to produce only a brief silent period of about 20-30 ms, without a preceding MEP. As might be anticipated, however, “corrected” left handed subject displayed ipsilateral MEPs obtained during bilateral contraction (22 ms, 2.8 mV, followed by a SP of 45 ms; Fig. 1). By contrast, in the patients with functional recovery, transcranial stimulation of both hemi- spheres was always capable of eliciting ipsilateral MEPs, followed by a silent period, during voluntary contraction, regardless of the lesion side (Fig. 2). iMEPs were elicited with a TCS intensity higher than that necessary for producing contralateral responses (Table 2). Since ipsilateral MEPs could not be elicited without voluntary activation, the threshold value to which we refer represents the level of TCS used during voluntary activation. An interhemi- spheric. difference was also. revealed: the mean threshold for the elicitation of contracted ipsilateral MEPs was lower in the unaffected hemisphere than in the damaged hemisphere (p < 0.01). The latency of iMEPs recorded from the recovering hand during stimulation of the unaffected hemisphere was markedly shorter than that obtained, in the same target muscle, by contralateral hemispheric stimula- tion (mean = 16.7 ms vs 19.1 ms, p<0.001). This latency was also shorter than that of ipsilateral MEPs in the normal hand produced by stimulation of the damaged hemisphere (mean = 16,7 ms vs 20.06 ms, p < 0.001). The amplitude of iMEPs recorded in the recov- Table 1. CANADIAN Pr. AGE/Sex CLINICAL SIGNS SCALE crt 1 3a LT Hemiparesis 5/8 Capsular infarct CT 2 asm RT Hemiparesis 7/98 3 com 719 4 ear 65/9 5 65M 75195 8 45 85/95 7 aa 65/75 8 em 5/8 8 aM 7195 10 52M 7/85 poral infarct CT " 52M RT Hemiparesis 75195 Capsular infarct CT 2 SOF LT Homiparesis 518 Fronto-temporal infarct CT|AM 3 ave RT Hemiparesis 7195 Parietal infarct CT deficit; the Vol 7 No 11 29 July 1986 1757M. D. Caramia et al THR THR MEP LATENCY MEP AMPLITUDE SILENT PERIOD % % (msec) (mv) Aims CONTRA IPSI_ CONTRA IPSI_CONTRA _IPSI_ CONTRA _IPSI MEAN 55,30 78,18 1828 ~=20,08 «= 4,200.67 140,30 23,90 Affected hemisphere SDs 13.00 10.15 1.35.55 0,80—40,90 14.41 MEAN 51,00 66,14" 19,10 16,70"** 5,59" 1,51*** 149,00 32,48 Unaffected hemisphere. ‘SDs 990 1300 090 140213 ,00— 380 11.50 MEAN 43,68" 940 42, Control values ‘80s 531 190 220 + p<.05; ** p<01; ** p<.001 THR ering hand was significantly greater than that obtained on the other side (p <0 001). The ipsilat- eral silent period (SP) obtained by stimulating the unaffected hemisphere was longer than that obtained in response to the stimulation of the affected hemi- sphere (32.4 ms vs 23.9 ms, p < 0.05). In the three patients examined before the onset of motor recovery, within 24h of the stroke, no ipsi- lateral MEP could be obtained on either side. This observation was not the consequence of a severe motor function impairment, since patients were still able to perform the mild contraction usually needed to evoke these category of MEPs. Ipsilateral MEPs, as described above, only became manifest with the onset of motor function recovery. The threshold intensity for the clicitation of contralateral MEPs was significantly higher than normal in patients (p< 0.001), independent of the anatomical lesion side (Table 2). Statistical compar- ison of thresholds between the two hemispheres showed no significant difference (55.3% vs 51%). ‘An interside significant difference was recorded Lett Hemisphere Stimulation (TCS 70%) Right side Contralateral MEP FIG. 1. Left handed ‘corrected’ control subject who wa se the Fight hand. The ower trace Note the prolonged onset (marked by the ti compared with that for The contralateral recordings (upper traces) Y75B_ Vol 7 No 11 29 July 1996 xcitablity threshold; IPSI = ipsilateral; CONTRA = contra I; Means of SPs are evaluated at 80% of TCS. between the amplitude of contralateral MEPs, with greater values measured on the unaffected hand (5.59 mV os 4.2 mV, p <0.05; Fig. 3). Indeed, amplitudes recorded from the unaffected hand were not only greater than those of the recovered side, but also significantly higher than mean amplitudes recorded in control subjects (5.59 mV vs 4.12 mV, p< 0.05) No appreciable difference between latencies of contralateral MEPs was detected between the two sides (Table 2). Unattected Hemisphere Stimulation (TCS 70%) 1 mv| : Right side FIG. 2. Comparison between cont Fecarded in espore to unaffected (lef) hemisphere st two patents. Neto the shorter latency af the IMEPs (in both lower traces.Cerebral plasticity after stroke AMPLITUDES OF CONTRALATERAL MEPs FIG. 9. Histograms showing mean amplitudes of contralateral control and unaffected hand MEPs, the latter being significantly higher Discussion Our results provide neurophysiological evidence of ipsilateral corticospinal activation during motor recovery after stroke. This suggests that the unaf- fected hemisphere, ipsilateral to the symptomatic hand, may partake in the functional recovery, a finding which is compatible with recent PET and Doppler studies? This phenomenon, representing. an example of plastic reorganization in the adult human brain, may open up horizons of a certain interest in the field of rehabilitation. However, the problem of defining both the origin and the fune- tional value of such ipsilateral activation needs to be investigated first, The mechanisms involved in ipsilateral activation may follow two types of hypothesis, one envisaging the ereation of a new pathway (sprouting), the other the pressing into service of existing ones. The latter could be achieved either by synaptic reorganization or by the ‘unmasking’ of pre-existing routes which lay either dormant or undetected in the normal brain. The latency of ipsilateral MEPs in recovering hand muscles is very short, with a noteworthy difference of about 3 ms with respect to contralateral MEPs in both patients and controls. This short latency is clearly the expression of a fast conducting and oligosynaptie pathway, which may find a ewo-fold explanation. Either the neurones involved belong to a rank of large cortical cells or they are activated deeply in the brain; both possibilities are consistent with the fact that iMEPs are obtained with relatively high intensities during voluntary facilitation. ‘On the basis of experiments on animals, there is as yet no anatomical evidence of neogenesis which matches short term functional recovery.” The hypothesis of sprouting was explored and then neuro(geport discarded in a study on monkeys with unilateral pyra- ‘midal lesions, it being concluded that there was no consistent evidence of regeneration to explain the fast recovery. Anew ipsilateral corticomotoneuronal transmission line was presumed to appear in only one animal after one year of survival.” The conjectural activation of a new pathway cannot be applied to either the short MEP latency or to their relatively rapid appearance, parallel to the clinical recovery, within 3 weeks of the stroke. Another possibility is a local sprouting from ipsi- lateral corticospinal fibres innervating proximal muscles towards distal motoneurones." Again, this cannot account for the short latency of iMEPs, since this too implies additional synapses and, therefore, an increase in the propagation time. Moreover, clin- ical recovery in these patients occurred within 3 weeks from the ictal episode, which represents too brief an interval for the manifestation of a sprouting process PET studies in stroke patients indicate that the movement of recovered fingers resulted in a signifi- cant rCBF increase in both contralateral and ipsilat- eral sensorimotor cortex. The ipsilateral increase was seen in a large area including the premotor cortex (PM) and the supplementary motor area (SMA). The in the secondary motor areas, descending outflow of ipsilateral volleys!" can be reasonably invoked for the genesis of ipsilateral MEPs after brain damage. Experimental data collected from monkeys show that there is a ‘normal’ small percentage of ipsilateral neuronal activity related to secondary motor areas.'* We might assume that in man this small percentage becomes more significant after stroke, by virtue of compensating, for the effects of vascular damage. Evidence for such ipsilateral reorganization was obtained by single neurone recordings in the supplementary motor area after experimental lesions of the primary motor cortex. According to this last hypothesis, if secondary motor areas are magnetically activated deeply in their ipsilateral projections, they could account for both the relatively high threshold and short latency of ipsilateral MEPs obtained in recov- ered patients. Finally, in support of this hypothesis SPECT evidence documents a reduced perfusion of the ipsilateral secondary motor areas in patients lacking motor recovery.’ The increase in contralateral MEP amplitude of the unaffected hand, with respect to both the affected hand and those of control subjects, presumably repre- sents extra facilitation of some kind. This extra quota might reflect an ipsilateral component from the damaged hemisphere which is triggered by the volun- tary command. Vol 7 No 11.28 July 1996 1759neurof geport Functional significance: In 1936, Foester speculated that the functional restitution in a hemiplegic patient with a motor cortical lesion may have been caused sphere.'® Are these ipsilateral MEPs the expression of a such functioning of the undamaged hemisphere, or simply the resule of a pathological rearrangement without a final functional correlate? Both interpreta- tions may be simultaneously valid. ents, optimal iMEPs were obtained in 1g hand, and this may suport a role for the undamaged hemisphere. At the same time, however, ipsilateral MEPs ~ albeit to a lesser extent ~were also recorded in the normal hand, where a compensatory function should not be required. Is this latter finding the effect of a reduced inhibi control on pre-existing ipsilateral pathways, which normally are suppressed? cis interesting to note that a recent mapping study in healthy subjects revealed occasional ipsilateral responses.” These authors speculated that ipsi- lateral responses may become functional and perhaps more accessible in the process of recovery from lesions impairing the normal route to the hand from the contralateral hemisphere. We may postulate that some sort of ipsilateral pathways always existed, but substantially suppressed, and therefore not easily detected in healthy subjects. It is as yet unclear exactly what function this has, apart from acting as a reserve to be taken up once contralateral routes are damaged. One has the impression that when things ‘go wrong this ‘reserve’ system is activated in a gener- alized manner, because, of course, it could not be tailored in advance for an unknown eventuality. Conclusion Ieshould be emphasized that our study has focused on a relatively specific and narrow clinical picture. The patients were all affected by stroke producing a mild impairment of motor function, followed by a fast and almost complete recovery, one which was M. D. Caramia et al not aided by physiotherapy, but which took place naturally. There is nothing which tells us that ipsi- lateral MEPs obtained in other circumstances will behave in a similar way.!™" It is worth recalling here that the mapping study mentioned earlier!” recorded hand iMEPs with latencies between 28 and 32 ms, that is to say almost double the latencies encountered here. The possibility of additional variety is further suggested by the nature of iMEPs which we recorded from a single control subject, the one who was left handed, but who had been taught to use the right hand from infancy. In her case the iMEPs had a latency of 22ms, and might be interpreted as the collateral of ‘early’ plasticity, taking place during maturation and doubtless the result of a different modality of reorganization with respect to the adult pattern. With time we will surely come to deal with just as wide a range of types of ipsilateral MEPs as we have now become accustomed to sceing in the context, of normal contralateral responses. References 1. Fes W, bara A. Sehictmann Kota, Bran 148, 368-389 198 imma Meee Mc Mcahsto Ho J Bey Brat Rar a, 669-79 (90 5 Garamla MD ~Clinal'P Porasco Cat a Eocvooncepolog” Cin {ah ae 0b 12, Bam AP and Stk PL J Nowronel 91, 66789 (108 13: Re och A and Si Pe Reon 18, Sen 96 a. i KE J Nero 0, 225943 (108 ted Fon, adh Handbuch der Newslog, Band rl Laone A td Halls M. Exp Brain Res 100, Wa TH. Ann Newt 29, 8-860 (98. nd Freund NG. Exp Sai os 032), 418-428 (900, rare Aca Brn 490, 12124 (1983, tar ose. 18, Fee W, Danek Cart Hanson Received 18 March 1996; accepted 1 May 1996 ‘otontials (MEPs) in hand muscl hemisphere, Under certain conditions, however, the activation. Tha present study describes pacity phenomenon by 1760. Vol 7 No 11.29 July 1996