From the Department of Neurology, Castellón General Hospital
(J.V.E.); Department of Neurology, Valencia University School of Medicine,
Valencia University General Hospital (J.S.); Department of Public Health,
Research Service, Valencia University Clinic Hospital (D.B.); Department of
Neurophysiology, University Peset Hospital, Valencia (M.E.); and Laboratory of
Neurophysiology, Department of Neurology, Valencia University General Hospital
(J.L.-T.) (Spain).
Methods—Fifty patients with different degrees of hemiparesis were
studied in the first week after ischemic stroke and evaluated
by clinical scales (Medical Research Council Scale, Canadian
Neurological Scale, Barthel Index), with clinical follow-up over 6
months. TMS (Magstim 200) was performed at the same time,
recording the motor evoked potential (MEP) in the thenar
eminence muscles, with facilitation by voluntary contraction.
Results—Of the total group of 50 patients, MEP was absent in 20
and present in 30 (17 with normal and 13 with delayed central
conduction time [CCT]). The patients with MEP showed better motor and
functional recovery than those without. The MEP provided information on
patient recovery, regardless of the initial strength and/or Barthel
values. The degree of recovery was better in those patients with normal
CCT than in those with delayed CCT.
Conclusions—MEP obtained by TMS in patients with hemiparesis
after acute ischemic stroke is useful as an early prognostic
indicator of motor and functional recovery. This technique would allow
the early identification of those patients who will have a good
recovery, particularly among those with severe initial paresis.
The introduction of neurophysiological techniques
for studying motor function (transcranial electric and
magnetic brain stimulation techniques)7 8 has
allowed their application to the study of patients with motor
disabilities as a result of cerebrovascular
disorders.9 Despite methodological differences,
the results of the literature point to the possible prognostic utility
of these techniques in predicting motor function
recovery.10 11 12 13 14 More recent studies nevertheless
report discordant results.15 16 17
We present a prospective study using transcranial
magnetic brain stimulation (TMS) in a group of patients with impaired
motor function caused by acute ischemic stroke. A prolonged
clinical follow-up period was involved to evaluate the prognostic
utility of the technique in assessing motor and functional recovery in
these patients.
Clinical Study
A clinical follow-up study was performed in all patients, again with
application of the MRC and Barthel Index at 2 and 6 months. The
patients who had totally recovered at the second month (MRC=5 and/or
Barthel=100) were not studied at the sixth month. The patients who died
were evaluated by the Barthel Index and MRC at the last visit. Of the
total group, 11 patients died (8 before 2 months and 3 after). The last
measure of the Barthel Index and/or strength index for each patient was
used for the evolution study (second clinical examination). We
considered clinical improvement or good recovery when the patient
reached a Barthel Index of
Neurophysiological Study
The TMS technique and the measurement of latencies and amplitudes were
performed according to the usual protocol,18 with
20% above-threshold and maximal stimulation output. The MEPs were
obtained in the 4 extremities, healthy and affected sides, of each
patient. For left hemisphere stimulation, we positioned coil face B
(current counterclockwise) with the maximum stimulation band fitted
tangential 3 to 4 cm lateral and posterior to the vertex to study the
arm, and in the vertex midline for the leg. In the case of the right
hemisphere, the coil position was reversed in position A (current
clockwise). We used facilitation techniques by voluntary contraction of
the homolateral target muscle or the heterolateral muscle in the case
of paralyzed muscles. The time taken by the response in traveling from
the brain hemisphere to the muscle was termed total conduction time. To
calculate peripheral conduction time, we used magnetic
radicular stimulation and cervical and lumbar sites. We obtained the
central conduction time (CCT) by subtracting peripheral
conduction time from total conduction time.
Absent MEP or absent CCT was defined when it failed to appear
after 3 successive discharges with maximum output. Although we studied
the arm and leg MEP, only the presence or absence of the arm MEP was
analyzed in the final study. The MEP with the shortest latency
and greatest amplitude was taken for analysis. Regarding CCT,
the patients were divided into 2 groups: those with a normal CCT and
those with delayed CCT. Delayed CCT was defined when it was >±2.5 SD
of the CCT of the arm MEP in the control group of healthy subjects in
our laboratory (>8.13 milliseconds); in this group of patients we also
analyzed the difference between the healthy side and the
affected side to confirm CCT prolongation. To estimate the normal
values of this technique, we studied a group of 30 subjects without
signs of neurological disease.
Statistical Analysis
MEP sensitivity and specificity in predicting motor and
functional outcome were calculated, and the corresponding 95%
confidence intervals were also estimated. Sensitivity refers to the
proportion of patients whose recovery is graded as good for the Barthel
and strength scores at the second exploration and who have MEP presence
in the acute ischemic stroke phase. Specificity in turn refers
to the proportion of patients whose outcome is defined as poor and
whose MEP is initially absent. All statistical analyses were
performed with the use of the SPSS statistical package except the
sensitivities and specificities, which were obtained with the Epiinfo
program. Values of P<0.05 were considered statistically
significant.
Most patients showed motor function recovery. In this sense, recovery
was reflected by the increase in Barthel Index and strength in both
groups (Table 2
As shown in Table 4
The high sensitivity (77.1 to 87.1) and specificity (81 to 85.7)
values for the total group indicate that patients with good recovery
have a high probability of MEP presence in the acute stroke phase,
while those with poor recovery have a high probability of absent MEP
(Table 5
On the basis of the multiple linear regression analysis
(Table 6
Of those patients with MEP absence in the first examination (20
patients), 11 were again studied by TMS during recovery. Of these
individuals, 4 continued to show no potential, while 7 exhibited MEP.
The clinical recovery of those patients who evoked MEP was
significantly better (5 good and 2 poor recoveries) than in the case of
those who continued to show no MEP (all with poor recoveries)
(P=0.04; Fisher's exact test).
Of the total group of patients, 11 died: 8 before 2 months (6 without
and 2 with MEP) and 3 later (with the presence of MEP). On
analysis of the relationship between MEP presence and death, a
statistically significant association was only found between absent MEP
and early death (P=0.04), with the use of Fisher's exact
test. None of the patients suffered complications derived from
application of the technique.
TMS allows the simple and painless study of the human motor
pathway in conscious subjects.7 8 It has been
applied to patients with cerebrovascular disorders, and
analyses have been made of its possible motor prognostic
value.9 The first studies were undertaken by
electric transcranial brain stimulation, involving a
follow-up period of 2 months. They concluded that the MEP possesses
prognostic value and that CCT might correlate with clinical
recovery.10 11 25 Posteriorly, magnetic
brain stimulation was applied with variable follow-up periods (1 to
3 months), the results again suggesting a possible predictive value for
this technique.12 13 14 26 27 28 29 30 31 32 33 In this context, MEP
amplitude was considered the most sensitive
parameter.30
On the basis of methodology very similar to our own, Heald et
al12 13 (1993) published the largest TMS study
(118 patients with ischemic stroke) involving the longest
follow-up period (12 months) to date. They concluded that the presence
of MEP with normal or delayed CCT is able to differentiate a group of
patients with a high probability of survival and good functional
recovery. In contrast, its absence would indicate poor recovery or an
increased risk of death. In their series, only 6% of patients
exhibited delayed CCT, compared with 26% in our study. This may be
explained by the fact that the upper limit of CCT normality used by
these authors was higher than our own (>9.2 versus >8.13
milliseconds, respectively). On the other hand, the difference could
also be justified by the promptness with which these authors applied
stimulation (between the first and third days after stroke). In such
early stages, factors such as edema or ischemic penumbra may
intervene at the cerebral level, reversibly affecting the absence of
potential, so that a few days later MEP appears with delayed CCT. These
authors found differences in functional scores between patients with
delayed CCT and normal CCT throughout the recovery period, although
after 1 year the differences were not statistically significant.
According to our data, statistically significant differences existed
between the 2 groups of patients, with those with normal CCT
presenting a better recovery.
Rapisarda et al33 (1996) performed a TMS
study of 26 acute ischemic stroke patients, with a follow-up of
14 days. These patients had a brain infarct in the sylvian region, with
total hand paralysis. MEPs were present in 42.6% of patients, and
their presence with an amplitude >5% that of the amplitude of the M
wave implied a favorable prognosis for motor function recovery.
According to these authors, MEP amplitude was of greater value than MEP
latency. However, patient follow-up was very short, and although a
large number of stroke patients recovered motor function mainly during
the first month, some patients had slower or later
recoveries.26
In contrast to these studies in support of the prognostic value of the
technique, other authors failed to corroborate its utility.
Rijckevorsel-Hartman et al15 (1993), in a study
of 27 patients followed up for 6 months, and Zgur et
al16 (1993), in a study involving 18 patients
followed up for 3 months, reported a lack of prognostic value for this
technique. Arac et al17 (1994) studied 26
patients with acute ischemic (19 patients) and hemorrhagic (8
patients) stroke using TMS with a follow-up period of 6 months and
methodology similar to our own. On analyzing the recovery of their
patients, they found that patients both with and without MEP recovered
and that the differences between the groups were not statistically
significant. Of note, however, is the fact that their percentage of
patients without MEP was high (63%) in comparison with our own
results, and in the patients with greatest weakness (13 patients), MEP
was only present in 1 patient, compared with 20% in our series. In
addition, these authors lost 22% patients to follow-up, and their
sample was more heterogeneous, including ischemic
and hemorrhagic strokes.
On analyzing clinical recovery by multiple linear regression
analysis, we found MEP to provide information on patient
recovery, regardless of the initial strength and/or Barthel values. The
presence of MEP, with either normal or delayed CCT, predicted better
recovery than in its absence. However, the prognostic value was more
notable in the group with the most affected motor function, in which
case recovery is more difficult to predict. Application of the
technique would allow us to identify the small group of paralytic
patients with good recovery, which cannot be identified in any other
way. On the other hand, the absence of MEP does not exclude good
recovery, which effectively occurs in a small percentage of these
patients. Those patients without MEP who recovered well also exhibited
MEP reappearance over time, ie, such reappearance during recovery
constitutes a good prognostic sign. It is thus seen that the multiple
regression models account for improvement of patient strength better
than the improvement in the Barthel Index. This may be due to the fact
that the TMS technique in particular evaluates the corticospinal motor
pathway, and thus motor function,18 while the
Barthel Index depends on other factors in addition to
strength.20
The utilization by some authors30 33 of MEP
amplitude as the most sensitive parameter involves the
problem of its important variability.26 34 In
this sense, we are of the opinion that MEP amplitude is of use when
applied in combination with changes in latency and threshold (which
usually occur together), while its isolated application does not
contribute more information.9 According to some
authors, the study of lower limb MEP is essential to secure precision
in motor prognostic value.12 17 In our series, we
found that most patients with MEP in the arm also recorded MEP in
the leg, perhaps because of the prevalence of faciobrachial paralysis
distribution in our series. The absence of both forms of MEP indicated
a worse recovery than when MEP was only absent in the arm.
Nevertheless, interpretation of lower limb MEP is more difficult, given
the inconsistency in recording these potentials,
than in the case of upper limb MEP.34 In our
series, no complications and no side effects were derived from the
technique. Similar observations have been reported by most authors,
with some exceptions.35 36
In physiopathological terms, the prognostic value of the technique
could be explained by considering that ischemia of the
corticomotoneurons and/or corticospinal tract (responsible for motor
impulse genesis and conduction) can cause partial or total, reversible
or irreversible damage of these structures.26 37
Recovery may be due to the brain plasticity
phenomenon,38 which through cortical
reorganization,39 40 collateral sprouting,
unmasking, or other mechanisms41 may succeed in
reestablishing the damaged connections or may generate new ones with
the spinal motoneuron. The application of this technique may contribute
to identify such connections at an early stage, even when they are not
functional for the patient. In an early stage after acute
ischemic stroke, the procedure would allow us to identify those
patients with less serious lesions, or with greater possibilities for
recovery, who cannot be detected by traditional clinical
methods.4
In conclusion, our results support the idea that MEP obtained by TMS
represents a useful early prognostic marker of motor function
recovery in ischemic stroke patients. In addition, the
technique could be used for monitoring and quantifying motor function,
in parallel with corticospinal tract permeability, in the course of
patient recovery.
Received March 9, 1998;
revision received June 8, 1998;
accepted June 8, 1998.
© 1998 American Heart Association, Inc.
Original Contributions
Prognostic Value of Motor Evoked Potential Obtained by Transcranial Magnetic Brain Stimulation in Motor Function Recovery in Patients With Acute Ischemic Stroke
Abstract
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Background and Purpose—The early
prognostic application of transcranial magnetic brain
stimulation (TMS) for assessing motor and functional recovery in
ischemic stroke patients has yielded contradictory results. We
performed a prospective study of patients with acute ischemic
stroke and motor deficit to evaluate the early prognostic value of TMS
in motor and functional recovery.
Key Words: evoked potentials, motor • prognosis • stimulation, transcranial magnetic • stroke, acute • stroke, ischemic
Introduction
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Although the establishment of an early prognosis of motor
function recovery and daily life capabilities in ischemic
stroke patients is difficult,1 the World Health
Organization2 nevertheless recommends the
definition of such prognostic indicators. These patients tend to
spontaneously recover part of their motor
function,3 although it is difficult in the early
stages of recovery to know the precise course, especially in severely
affected patients. The clinical indicators used are based on the
quantification of neurological scales4 or
residual strength of the paretic muscles5 and on
the observation of recovery over the first 4
weeks.6
Subjects and Methods
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Patients
Fifty-four consecutive patients with acute ischemic
stroke were recruited for 1 year from the Service of Neurology of a
university hospital. Patients were included in the study when the first
acute ischemic stroke had been stabilized, with motor function
having been affected for >24 hours. Patients were excluded if their
clinical situation did not allow them to collaborate in the study or if
they presented the usual contraindications for this
technique.18 Of the total group, 4 patients were
excluded from the study because their ulterior course could not be
established. Thus, 50 patients were finally evaluated; their clinical
characteristics are shown in Table 1
. The protocol was
approved by the Hospital Research and Ethics Committee, and all
patients gave informed consent to participate in the study.
View this table:
[in a new window]
Table 1. Clinical and MEP Characteristics of Stroke Patients
and Control Group
Stroke was diagnosed by medical history and clinical
investigation, with all patients subjected to a CT scan and/or brain
MRI study to evaluate infarct size and location. The patients were
studied by the same researcher in the first week (first clinical
examination). The clinical data on neurological status were gathered by
the Canadian Neurological Scale (CNS).19
Functional capacity was determined by the Barthel
Index,20 and control muscle strength was assessed
by the Medical Research Council scale (MRC)21 of
the abductor pollicis brevis in the upper limb and of the abductor
hallucis in the lower limb. 
60 and/or strength index in the tested
hand muscle of 4. We analyzed only the values of hand muscle
strength and the Barthel Index corresponding to the second examination,
as well as the increase in strength and Barthel Index for each patient
in the period between the first and second examinations. All patients
were first analyzed combined (total group), followed by
evaluation of patients with severely affected motor function (hand
palsy, including strength in the thenar eminence muscles of degrees 0
to 1 according to the MRC).
The patients were studied by TMS between the third and
seventh days after stroke onset by the same researcher (first TMS
examination) in the Neurophysiological Laboratory.
A MAGSTIM Novametrix 200 magnetic stimulator was used with a 9-cm
diameter coil, capable of generating a 2-T maximum field intensity
(NOVAMETRIX Inc). The motor evoked potentials (MEPs) were
studied with conventional cup-shaped surface electrodes in the abductor
pollicis brevis muscle for the arm and on the abductor hallucis
for the leg, with classic bipolar mounting (tendon-muscle). The
responses were recorded by a conventional electromyograph (Mistro
Medelec); the usual filters were used for motor conduction.
The Mann-Whitney U test was used to compare the
distributions of 2 independent samples derived from quantitative
variables, while the 
2 or Fisher's exact
test was performed to compare proportions. The independent predictive
value of the variables initial Barthel Index, initial strength,
CNS, and CCT (normal, delayed, or absent) was studied by multiple
linear backward regression analysis, using as dependent
variable either the improvement in Barthel score (final minus
initial) or the improvement in strength score. The basic requirements
of the regression models were evaluated by the regression
diagnostic methods described by Kleinbaum et
al.22
Results
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Regarding the CCT values obtained in the first TMS examination,
the patients were divided into 3 groups. The first comprised 20
individuals with absent CCT (or absent MEP), the second 13 with delayed
CCT, and the third 17 with normal CCT. In the group with severe
hemiparesis (24 patients), there were 19 patients with absent CCT, 3
with delayed CCT, and 2 with normal CCT. The latter 2 patients
presented prolonged CCT (although the values did not exceed the
limits regarded as delayed) and combined a reduced amplitude (compared
with the healthy side) of 
50% to 70% with increased thresholds
(Table 2).
View this table:
[in a new window]
Table 2. Evaluation of Patients by Muscle Strength (MRC),
Barthel Index, and
CNS 
) and by the values of these indices in the second
investigation. According to the bivariate analysis (Table 3) regarding the total
group of patients, the increase in values for strength between the
first and the second examinations differed in each group according to
its CCT (P=0.042). Regarding the group with severe
hemiparesis, the presence of CCT was associated with a significantly
greater improvement in both indices.
View this table:
[in a new window]
Table 3. Increases in Strength and Barthel Index
Scores , the
proportions of patients with MEP who exhibited good recovery were
significantly higher than for patients without MEP, both in the total
group of patients and in those with low levels of initial strength. On
analyzing complete recovery of those patients with delayed or normal
CCT (Table 4), we found that the proportion of patients who recovered
completely was greater for those with normal CCT than for those with
delayed CCT, since a great majority of those patients reached the
highest values for both indices.
View this table:
[in a new window]
Table 4. MEP and Motor Recovery of Stroke
Patients ). Such a high
positive predictive value of the test indicates that patients with MEP
presence in the acute phase of stroke have a high probability of good
recovery. Nevertheless, the negative predictive value is lower (in the
range of 60 to 80), since the patients with absent MEP include a group
who will have a good recovery. In the case of patients with greater
initial weakness the sensitivity is low (36.4 to 62.5), although in
contrast the specificity is very high; thus, the MEP presence has a
high predictive value in this particular group of patients.
View this table:
[in a new window]
Table 5. Sensitivity and Specificity (with 95% Confidence
Intervals) ), the presence of
MEP (with normal or delayed CCT) and CNS was positively associated with
Barthel score improvement; in contrast, the arm strength score and
initial Barthel score were negatively associated with Barthel score
improvement. Twenty-nine percent of the variation in Barthel score
improvement was accounted for by this model. Likewise, in the other
model, MEP presence (with normal or delayed CCT) and the initial
Barthel score were positively associated with strength score
improvement, whereas the initial arm strength score was negatively
associated with strength score improvement. Fifty-four percent of the
variation in strength score improvement was accounted for by these 3
variables.
View this table:
[in a new window]
Table 6. Factors Associated With Improvement in Barthel and
Strength Indices, using Multiple Linear
Regression
Discussion
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
The capacity to recover motor function is one of the most
important considerations for the subsequent adaptation and recovery of
ischemic stroke patients. In this context, it is essential to
secure early prognostic indicators of this
capacity.2 Many patients recover variable
degrees of motor function over the first 3 months, particularly in the
first 4 weeks after stroke.3 The severity of
paralysis at the time of stroke is less consistently related to
poor prognosis than to other indicators5 6 such
as the presence of a previous stroke, sphincter incontinence,
perceptive and cognitive losses, or low functional scores on
admission.1 Of those patients with most severe
arm or hand palsy, only a limited number experience good recovery, with
percentages of 15% to 27% according to different
authors.5 23 24 At present, no precise
indicators are available to differentiate from the severely affected
patients those small numbers of cases with better recovery.
Acknowledgments
This study was supported by Hidroelectrica Española,
SA (Spain).
Footnotes
Reprint requests to Joaquín V. Escudero, MD, PhD, Servicio de Neurología, Hospital General de Castellón, Avda Benicassin s/n, 12004-Castellón, Spain.
References
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  R. E. Frye, A. Rotenberg, M. Ousley, and A. Pascual-Leone Transcranial Magnetic Stimulation in Child Neurology: Current and Future Directions J Child Neurol, January 1, 2008; 23(1): 79 - 96. [Abstract] [PDF]
|
|
|
|
 |
|
 M. Hatakenaka, I. Miyai, S. Sakoda, and T. Yanagihara Proximal paresis of the upper extremity in patients with stroke Neurology, July 24, 2007; 69(4): 348 - 355. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J Butler and S. L Wolf Putting the Brain on the Map: Use of Transcranial Magnetic Stimulation to Assess and Induce Cortical Plasticity of Upper-Extremity Movement Physical Therapy, June 1, 2007; 87(6): 719 - 736. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Desmurget, F. Bonnetblanc, and H. Duffau Contrasting acute and slow-growing lesions: a new door to brain plasticity Brain, April 1, 2007; 130(4): 898 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Stinear, P. A. Barber, P. R. Smale, J. P. Coxon, M. K. Fleming, and W. D. Byblow Functional potential in chronic stroke patients depends on corticospinal tract integrity Brain, January 1, 2007; 130(1): 170 - 180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Binkofski and R. J. Seitz Modulation of the BOLD-response in early recovery from sensorimotor stroke Neurology, October 12, 2004; 63(7): 1223 - 1229. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Pennisi, G. Rapisarda, R. Bella, V. Calabrese, A. Maertens de Noordhout, and P. J. Delwaide Absence of Response to Early Transcranial Magnetic Stimulation in Ischemic Stroke Patients : Prognostic Value for Hand Motor Recovery Stroke, December 1, 1999; 30(12): 2666 - 2670. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||