вход по аккаунту


Auditory evoked magnetic fields after ischemic brain lesions.

код для вставкиСкачать
Auditory Evoked Magnetic Fields
after Ischemic Brain Lesions
J. P. Mgkeh, MScD,” R. Hari, MScD,? L. Valanne, MD,+ and A. Ahonen, DrTechnf
Auditory evoked magnetic fields to noiselsquare-wave sequences, presented once every 2.2 seconds, were recorded in
8 patients who had ischemic lesions in the auditory cortex or in its vicinity. In 2 patients with large temporoparietal
lesions, the magnetic 100-msec deflection (N100m) was absent over the damaged side. In 1 patient with a large but
less deep frontotemporal lesion, a small NlOOm could be discerned on the defective side. Frontal lesions or small
lesions in the vicinity of the supratemporal plane had no effect on N100m. Auditory evoked magnetic field recordings
may be useful in clinical studies of auditory cortical functions.
Makela JP, Hari R, Valanne L, Ahonen A. Auditory evoked magnetic fields after
ischemic brain lesions. Ann Neurol 1991;30:76-82
Ischemic lesions in temporal brain regions can cause
pronounced functional defects, including disturbances
in speech perception and production. Although the anatomical extent and site of these lesions are routinely
determined with various neuroradiological imaging
methods, reliable studies of the electric function are
sparse. Furthermore, results of auditory evoked electric potential (AEP) recordings have been controversial. Bilateral temporal lobe lesions, causing cortical
deafness, have been reported to lead either to the disappearance of NlOO (the vertex-negative 100-ms response) or to have no effect on it (for a review, see
Auditory evoked magnetic fields (AEFs) (for a review, see [2)) allow more precise determination of the
active brain sites than the evoked potentials recorded
on the scalp. The magnetic fields remain relatively undistorted as they emerge through tissue layers of inhomogeneous electric conductivities, whereas electric potentials are smeared [ 3 , 41. Magnetoencephalography
(MEG) allows selective studies of the auditory areas in
each hemisphere, so that activity on the opposite side
does not interfere with the interpretation of the data.
The source areas are generally modeled with current
dipoles; an “equivalent dipole” is the source that best
explains the measured field pattern. Although AEPs at
about 100 msec reflect activity of several brain regions,
magnetic NIOOm seems to be generated in the auditory areas of the supratemporal cortex 12, 5-71, and it
may be affected by vascular lesions of the temporal
lobe [S].
In the present study, we recorded AEFs from pa-
From the +Low Temperature Laboratory, Helsinki University of
Technology, Espoo, and Departments of *Neurology and $Radio]ogy, Helsinki University Central Hospital, Helsinki, Finland.
tients with accurately defined vascular lesions in or near
the auditory cortex to find out by which type of lesions
the responses are altered. We also wanted to investigate the clinical applicability of MEG in characterizing
the functional defects caused by ischemic brain lesions.
Materials and Methods
Eight patients ( 5 women, 3 men; age range, 23-64 years;
mean age, 38 years) with a unilateral ischemic infarction were
studied. Five patients had a lesion in the left hemisphere and
3 patients in the right. The experiment protocol was accepted
by the Ethical Committee of the Helsinki University Central
Hospital. The patients gave their informed consent for the
measurements. Routine blood tests, thorax radiograph, and
the electrocardiogram (EKG) were studied in all patients;
Patient 1 had atrial fibrillation in his EKG at the time of
admission to hospital. Aortocervical angiography was made
in Patients 2 to 8; Patient 2 had a thin right vertebral artery,
Patient 4 had left internal carotid artery dissection, and Patient 6 a small atherosclerotic plaque in the right internal
carotid artery. In carotid angiography, vasospasm was seen in
the arteries ascending from the left medial cerebral artery of
Patient 2. Cognitive functions were studied in all patients,
either by a neuropsychologistor by a speech pathologist. All
5 patients with left hemisphere lesions were aphasic. From
these, Patients 1, 2, and 6 were studied with the Finnish
version of the Western Aphasia Battery 191 and Patients 3
and 4 with the Finnish nonstandard aphasia examination battery [lo}. The clinical data are presented in the Table. The
lesions, determined by computed tomography (CT) or magnetic resonance imaging (MRI), are depicted in Figure 1.
Patients 1,5,and 6 had the most extensive lesions; in Patients
4 and 8, the lesions were limited to three slices.
Received Sep 4, 1990, and in revised form Nov 21, 1990, and Jan
11, 1991. Accepted for publication Jan 17, 1991.
Address correspondence to Dr Mikela, Low Temperature Laboratory, Helsinki University of Technology, 02150 Espoo, Finland.
76 Copyright 0 1991 by the American Neurological Association
Clinical Featarres of Eight Patients with Prior Stroke in or Near the Aaditov Cortex
Age (yr)/
Previous Diseases
Cause of Stroke
Main Symptoms
Cardiac embolizacion
Cilrotid dissection
Depression, transient
ischemic attack
Conduction aphasia,
right hemianopia
Transcortical motor
aphasia, weakness in
right hand
Transcortical motor
Mild anomic aphasia
Left hemifield inattention, mild left facial
Conduction aphasia
AEF = auditory evoked magnetic field; CT
Left hemiparesthesia
and clumsiness
computed tomography; MRI
The auditory stimuli were noise/square-wave sequences; a
300-msec white-noise burst was followed by a 250-Hz square
wave, 200 msec in duration. The stimulus was repeated every
2.2 seconds. Responses to this type of stimuli resemble those
evoked by syllables consisting of a fricative consonant followed by a vowel [ll, 12). Similar stimuli were used earlier
in our laboratory for the study of aphasic subjects 18).The
noise bursts were gated from a signal stored on magnetic
tape. To prevent artifacts, the stimuli were converted to
sounds outside the shielded room with a TDH-39 earphone
(Madsen Electronics, Denmark), and led to the patient by a
3-m long, plastic tube, 18 mm in diameter. The frequency
response of the sound transmission system was flat up to 5
kHz. The intensities of the stimuli were 70 to 80 dB sound
pressure level, and we ensured that the patients clearly heard
the sounds.
The recordings were made in our magnetically shielded
room. During the measurement, patients were lying on a
nonmagnetic bed with the head fixed by a vacuum cast. Patients were instructed to keep the eyes open during the measurement. One of the authors stayed with the patients inside
the shielded room.
The magnetic responses over each hemisphere were measured either with a seven-channel first-order axial SQUID
gradiometer (Patients 1-4) or with our new 24-channel planar SQUIDgradiometer (Patients 5-8). In the seven-channel
device [13], the hexagonally mounted pickup coils (diameter,
20 mrn) are 36.5 mm apart on a spherical surface with a
radius of 125 mm, 18 to 20 mm above the scalp. The base
length of the axial gradiometers is 60 mm. In the 24-channel
gradiometer [14], a planar flux transformer configuration is
used; instead of recording the field component B, normal to
the head, the two orthogonal figure eight-shaped pickup
Stroke and
Lesion Determined by
2 mo
CT x 2
1 mo
1 mo
a x 3
3 mo
2 wk
CT x 2
2 Yr
CT, MRI (0.02 T)
5 Yr
3 mo
CT 1985, MRI (1 T) 1990
CT, MRI (1 T)
2, MRI (0.02 T)
magnetic resonance imaging.
coils measure the tangential derivatives dBr/i3x and dB,/ay
simultaneously at 12 locations, 3 cm apart. Whereas the axial
magnetometer detects the largest responses at the field extrema on both sides of a current dipole, the 24-SQUID planar gradiometer records the largest signal just above the dipole where the field gradient is at its maximum. The wide
coverage and the large number of channels in this device
often allow the source to be determined with a single measurement, without moving the instrument, which is very useful for patient studies.
The exact locations and orientations of the pickup coils
with respect to the head were determined by measuring the
magnetic field produced by a set of small coils fixed on the
scalp; the method results in an accuracy of about + 2 mm
with respect to external landmarks on the skull [131. MEG
signals were measured over each hemisphere. To cover both
of the field extrema in the measurements performed with the
seven-channel magnetometer, responses were recorded both
from the anterior and posterior ends of the sylvian fissure.
With the 24-channel instrument, one recording just above
the auditory cortex was usually sufficient. Vertical electrooculogram was monitored during the magnetic measurements to reject data contaminated by artifacts caused by vertical eye movements and blinks.
Signal A nulysis
The band-pass filtered signals (3 dB points at 0.05 and 100
Hz, roll-off of the high-pass filter 35 dB/decade and over 80
dB/decade for the low-pass filter) were digitized at 500 Hz,
averaged on line, and stored in a computer. About 100 responses were averaged at each location. If the state of the
patient permitted, the measurement was repeated to demonstrate the replicability of the responses. The analysis time
included a 150-msec prestimulus baseline, used as the zerolevel for amplitude measurements. The first two responses of
each series and responses coinciding with eye-channel signals
M&ela et
AEFs after Stroke 77
Patient 1
Patient 2
Patient 3
Patient 4
Patient 5
Patient 6
Patient 7
Patient 8
78 Annals of Neurology Vol 30 No 1 July 1991
m m
exceeding 150 pV were rejected. The averaged responses
were digitally filtered with a passband of 0.05 to 40 Hz.
Source locations were determined from data obtained with
the 24-channel instrument. The equivalent current dipoles
were found with a least-squaresfit by assuming that the head
is a spherical volume conductor with a curvature of 120 mm
in the measurement area; the sphere model is a reasonable
approximation of the posterotemporal head region 16, 151.
The CT and MFU data were extrapolated to a standard series
of anatomical pictures in the axial plane, parallel to the cantomeatal line, slice thickness being 10 mm.
Figure 2 illustrates responses of Patient 5, recorded
with the 24-channel device. Over the healthy side, a
typical sequence of two prominent deflections is evident; the first, NlOOm, peaks at 92 msec. The transition from noise to square-wave tone elicited another
large deflection, NlOOm’, with a latency of 430 msec,
that is, 130 msec after the change within the stimulus.
This patient had a large temporoparietal lesion, and the
responses over the lesioned side were strongly attenuated. The transient NlOOm and N100m’ were totally
missing, but a slow shift was seen at about 400 msec.
In Figure 2, the black arrow on the measurement
grid indicates the site of the equivalent dipole of
NlOOm in the healthy left hemisphere. This location,
including its depth of 33 mm, agrees with activation of
the supratemporal auditory cortex (compare with {21).
The dipole moments, indicating the strength of the
response, are 24 and 35 nA-m during NlOOm and
NlOOm’, respectively. The estimated equivalent dipoles account for 93 and 96% of the measured field
pattern. In measurements over the infarcted right side,
the equivalent dipole accounts for 88% of the measured field pattern at 403 msec. This dipole has an
opposite polarity to that observed in the healthy side,
however, and is about 3 cm more posterior; the depth
is 28 mm. The finding suggests that some activity, although totally abnormal, has remained on the lesioned
auditory cortex.
Figure 3 illustrates the data from all patients in the
channel with the most prominent responses. The mean
latency of NlOOm was, on the healthy side, 91
msec (mean
SEM; 7 patients) and 101 2 9 msec
(6 patients) on the lesioned side; the difference was
statistically nonsignificant. The corresponding latencies
9 msec. In
for N100m’ were 112 2 5 and 116
Patients 1 and 5 , the typical response pattern was totally absent over the lesioned side. In Patients 2 and
Fig 1 . Extent of the lesions shown schematically in computed to-
4 mographic
(magnetic resonance imaging, slices fir all patients
(Patients 1-8). The locations of the slicer
shmun in the inare
sert. The scans are numbered and viewed from b e l w .
3 , the waveform was preserved, but the latencies of
the responses were increased by 61 and 23 msec for
NlOOm and by 37 and 41 msec for NlOOm’, respectively. In Patient 6, artifacts due to dental screws prevented measurements over the healthy hemisphere.
Over the diseased hemisphere, s m a l l NlOOm and
N100m’ with normal latencies were present. In Patients 4, 7, and 8, the responses were normal over both
In two of our patients, Patients 1 and 5, the magnetic
NlOOm was absent over the lesioned side. In both
patients, the lesions extended into the middle region
of the supratemporal plane and to the inferior parietal
area. It has been observed that similar bilateral lesions
abolish electric long-latency responses (e.g., Cll),
whereas unilateral lesions in the inferior parietal area
alone did not have any significant effect [16}. Our results are in good agreement with the suggestion of
Scherg and von Cramon (171 that the medial parts of
the supratemporal plane are necessary for generation
of the electric N100. This interpretation is also supported by the remaining responses in our Patient 6
who has severe aphasia and an extensive lesion sparing
the most medial regions of the temporal auditory cortex. The data do not allow us to conclude, however,
that destruction of medial parts of the supratemporal
plane, only, is necessary and sufficient for abolition of
N 1OOm.
In Patient 4 , with an infarction in the supratemporal
plane, in close vicinity of the Heschl’s gyri but again
saving the most medial parts, NlOOrn seemed intact.
This was also so in Patient 7, with lesions in the anterior temporal lobe and in the inferior parietal area,
posterior to the supratemporal plane. These findings
suggest that small temporal infarctions may not alter
NlOOm significantly. The remaining electric N 100 that
has been observed in patients with infarctions in the
temporal area has been attributed to sources outside
the temporal lobes [lS]; our data show that this is not
necessarily so in small infarctions.
Patients 2 and 3, with frontal lesions anterior to the
auditory cortex, showed no clear alterations in amplitudes of the auditory 100-msec responses, in contrast
to earlier findings {S, 19}. The amplitudes were smaller
over the left, diseased hemisphere; this is, however,
often so in normal subjects as well. Nevertheless, there
was a clear delay of NlOOm and N100m‘; the reason
for this remains unclear. Lesions deep in the posterior
temporal lobe and lateral thalamus in Patient 8 also left
NlOOm intact, implying that these areas do not take
part in the generation of N100m.
Disappearance of AEPs has been attributed to retrograde degeneration of thalamic cells after a lesion on
the supratemporal plane (11. It seems, however, imMakela et al: AEFs after Stroke 79
0 300
870 ms
[I00 f T / c m
I \ * -
80 Annals of Neurology Vol 30 No 1 July 1991
0 300
870 m s
probable that retrograde degeneration would be fast
enough to cause total depletion of responses as soon
as 2 weeks after the infarction, as seen in our Patient
5 ; this further implies the critical role of the auditory
cortex in the generation of N100m. In our small group
of patients, the amplitude and latency values of N lOOm
and N100m’ over the intact hemispheres corresponded well with those obtained previously from
healthy subjects [S, 12). For example, we did not observe any clear decrease of responses over the intact
hemisphere, as described for the electric NlOO in patients with similar lesions { 16, 193. This difference may
be due to the selectivity of MEG to nearby sources.
Our noise/square-wave stimulus, although resembling
speech sounds acoustically, did not add to the diagnostic value of MEG. Absence of N 1OOm or N 100m’ over
the right hemisphere was not correlated with speech
There are several caveats that should be kept in
mind when evaluating effects of ischemic lesions on
the generators of auditory evoked responses. Preserved and metabolically active cortical tissue, not evident in CT scans but seen by positron emission tomography, can produce visual evoked potentials [207; a
similar situation may exist in the auditory system. Furthermore, enhancement with the contrast medium implies metabolical alteration, not complete destruction.
It is possible that a reversal of hypoxia and return of
normal function in a part of the area with contrast medium enhancement has occurred before the MEG measurements. Nevertheless, the functional recovery does
not necessarily imply that total recovery of the ischemic damage has occurred. This is evident from studies
of monkeys in which small surgical lesions in the auditory cortex may cause only transient functional defects
[2 1, 221, possibly because of reorganization of function
after focal lesions [23, 24).
[ 200
Patient 3
each pair re$& t o jidd gradients measured in the y- and x-directions, respectively. The inserts depict the arrangement of sensors.
The shaded and white bar drawn an the time scale indicate
noise and square-wave stimuli, respectively. Two independent
measurements are superinposed to illustrate the replicability of
the responses. The black arrow in the upper insert head shows
the location of the equivalent dipolefor NlOOm at 92 msec i n
the healthy hemisphere. The equivalent dipole o f N100m’at
431 msec (white arrow) was 7 mm anterior and 6 mm below
this location; the 95% conjdence limits (compare with {225)) f i r
the location were 2 t o 3 mm transversal t o dipole orientation.
The white arrow in the lmer imert head correspands to the abnormally oriented equivalent dipole at 403 msec in the lesioned
hemisphere (time indicated by the vertical dashed line on upper
trace of location 4); at 80 to 120 msec, no dipolarfieldpattern
was detected over this side.
[SO f T / c m
Patient 6
Patient 7
Patient 8
2. Auditory evoked magnetic responses of Patient 5, recorded
4 Fig
with our 24-channel gradiometer. The upper and lower traces o f
870 ms
Fig 3. Illustrative responm of all patients on the healthy and
infarcted sides.,The most prominent signals are shown in each
case. To facilitate comparisons, all responses are depicted with
the NlOOm dejection directed upwards. Note that the sevenchannel instrument, used for Patients 1 to 4, measures the mugnetic field (in fmtoteslaj, whereas the 24-channel &ice (Patients 5-8) records the jield gradient (in femtoteslalcm).
Makek e t
AEFs after Stroke 81
The side of altered evoked responses is evident directly from the magnetic recording, without any extra
analysis, which is needed to interpret the corresponding electric responses [17J MEG, which monitors neuronal activity with millisecond time resolution, may
offer complementary information to MRI and positron emission tomographic studies. Multichannel MEG
measurements can be performed quickly because fixing of electrodes is not necessary. This may prove useful in monitoring patients with acute diseases. Some
strokes in the region of medial cerebral artery clearly
alter N100m; follow-up of this change from the onset
of the symptoms to the stable phase thus might give
new information of cortical processes involved in the
infarction and may even allow evaluation of therapeutic procedures. For this purpose, multichannel instruments are needed in clinical surroundings.
This study was financially supported by the Academy of Finland and
by the Korber Stiftung (Hamburg, Germany). We thank Dr M.
Hiimmsl;iinen for computer software and Mr M. Lehtihalmes for suggestions in patient selection and in classifying the speech symptoms
of the patients. Dr M. Kaste, Mr M. Lehrihalmes, and Prof 0. V.
Lounasmaa made valuable comments on the manuscript.
1. Woods DL, Clayworth CC, Knight RT, et al. Generators of
middle and long-latency auditory evoked potentials: implications
from srudies of patients with bitemporal lesions. Electroencephalogr Clin Neurophysiol 1987;68:132-148
2. Hari R. The neuromagnetic method in the study of the human
auditory cortex. In: Grandori F, Hoke M, Romani GL, eds.
Auditory evoked magnetic fields and potentials. Advances in
audiology, vol 6. Basel: Karger, 1990:222-282
3. Wihamson SJ, Kaufman L. Biomagnetism. Journal of Magnetism and Magnetic Materials 1981;22:129-202
4. Hari R, Ilmoniemi RJ. Cerebral magnetic fields. CRC Crit Rev
Biomed Eng 1986;14:93-126
5. Hari R, Aittoniemi K, J h i n e n ML, et al. Auditory evoked
transient and sustained magnetic fields of the human brain. Localization of neural generators. Exp Brain Res 1980;40:237-240
6. Yamamoto TY, Williamson SJ, Kaufman L, et al. Magnetic localization of neuronal activity in the human brain. Proc Nacl
Acad Sci USA 1988;85:8732-8736
7. Pantev C, Hoke M, Lehnem K, et al. Identification of sources
of brain neuronal activity with high spatiotemporal resolution
through combination of neuromagnetic source localization
(NMSL) and magnetic resonance imaging (MRI). Electroencephalogr Clin Neurophysiol 1990;75:173-184
8. Leinonen L, Joutsiniemi SL. Auditory evoked potentials and
magnetic field in patients with lesions of the auditory cortex.
Acta Neurol Scand 1989:79:316-325
82 Annals of Neurology Vol 30 No 1 July 1991
9. Kertesz A. The Western Aphasia Battery. Test manual. New
York: Grune and Stratton, 1982
10. Lempinen M, Soderholm S. Lyhyt Afasiatutkimus. Helsinki: Suomen Puheterapeuttiliitto r.y., 1986
11. Kaukoranta E, Hari R, Lounasmaa OV. Responses of the human
auditory cortex to vowel onset after fricative consonants. Exp
Brain Res 1987;69:19-23
12. Miikela J, Hari R, Leinonen L. Magnetic responses of the human
auditory cortex to noiseitone-transitions. Electroencephalogr
Clin Neurophysiol 1988;69:423-430
13. Knuutila J, Ahlfors S, Ahonen A, et al. A large-area low-noise
seven-channel DC SQUID magnetometer for brain research.
Rev Sci Instr 1987;58:2145-2156
14. Kajola M, Ahlfors S, Enholm GJ, er al. A 24channel magnetometer for brain research. In: Williamson SJ, Hoke M, Stroink G,
Kotani M, eds. Advances in biomagnetism. New York: Plenum
Press, 1989673-676
15. Himdiiinen MS, S m a s J. Realistic conductivity geometry
model of the human head for interpretation of neuromagnetic
data. IEEE Trans Biomed Eng 1989;36:165-171
16. Knight RT, Scabini D, Woods DL, Clayworth C. The effects of
lesions of superior temporal gyms and inferior parietal lobe on
temporal and vertex components of the human AEP. Electroencephalogr Clin Neurophysiol 1988;70:499-509
17. Scherg M, von Cramon D. Dipole source potentials of the auditory cortex in normal subjects and in patients with temporal
lobe lesions. In: Grandori F, Hoke M, Romani GL, eds. Auditory evoked magnetic fields and potentials. Advances in audiology, vol 6. Basel: Karger, 1990:165-193
18. Pool KD, Finitzo T, Hong C-T, et al. Infarction of the superior
temporal gyms: a description of auditory evoked potential latency and amplitude topology. Ear Hear 1989;10:144-152
19. Knight RT, Hillyard SA, Woods DL, Neville HJ. The effects
of frontal and temporal-parietal lesions on the auditory evoked
potential in man. Electroencephalogr Clin Neurophysiol 1980;
20. Celesia GC, Polcyn RD, Holden JR, et al. Visual evoked potentials and positron emission tomographic mapping of regional
cerebral blood flow and cerebral metabolism: can the neuronal
generators be visualized ? Electroencephalogr Clin Neurophysiol
21. Dewson JH, Pribram KH, Lynch JC. Effects of ablations of
temporal cortex upon speech sound discrimination in the monkey. Exp Neurol 1969;24:579-591
22. Heffner HE, Heffner RS. Effect of unilateral and bilateral auditory cortex lesions on the discrimination of vocalizationsby Japanese macaques. J Neurophysiol 1986;56:683-701
23. Kaas JH, Krubiaer LA, Chino YM, et al. Reorganization of
retinotopic cortical maps in adult mammals after lesions of the
retina. Science 1990;248:229-23 1
24. Jenkins WM, Merzenich MM. Role of cat primary auditory cortex for sound-localization behavior. J Neurophysiol 1984;52:
25. Kaukoranta E, HimZliinen M, Sarvas J, Hari R. Mixed sensory
nerve stimulations activate different cytoarchitectonic areas at
the human primacy somatosensory cortex SI. Neuromagnetic
recordings and statistical considerations. Exp Brain Res 1986;
Без категории
Размер файла
605 Кб
ischemia, lesions, magnetic, auditors, field, brain, evoked
Пожаловаться на содержимое документа