Available online at www.sciencedirect.com

Magnetic Resonance Imaging 27 (2009) 1360 – 1369

Neurophysiological and functional MRI evidence of reorganization of
cortical motor areas in cerebral arteriovenous malformation
Francesca Caramia a,⁎, Ada Francia a , Caterina Mainero b , Emanuele Tinelli a ,
Maria Giuseppina Palmieri c , Claudio Colonnese a , Luigi Bozzao a , Maria Donatella Caramia c
b

a
Department of Neurological Sciences, University of Rome “La Sapienza,” 00185 Rome, Italy
Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA 02129, USA
c
Department of Public Health, University of Rome “Tor Vergata”, Via Montpellier 1, 00133 Rome, Italy
Received 16 October 2008; revised 3 April 2009; accepted 7 May 2009

Abstract
Functional magnetic resonance imaging (f MRI) research has shown that brain arteriovenous malformations (AVMs) lead to
reorganization of cortical motor areas. Since it is known that blood oxygenation level-dependent signal in f MRI may be influenced by the
hemodynamic perturbation associated with the presence of the AVM, in the present study, a combined exploration with f MRI and
transcranial magnetic stimulation was performed in a patient with a right rolandic AVM in order to explore the relationship between neuronal
and hemodynamic activity. The combined protocol of investigation adopted in this study was able to provide significant information
regarding neuronal activity of the different cortical areas that partake to post-lesional reorganization.
© 2009 Elsevier Inc. All rights reserved.
Keywords: Cortical evoked potential; f MRI; Arteriovenous malformation

1. Introduction
Arteriovenous malformations (AVMs) are generally
considered to be congenital vascular lesions even though
there is an increasing evidence that the majority of cerebral
AVMs develop postnatally and represent various types of
endothelial cell dysfunction [1].
By means of functional magnetic resonance imaging
(f MRI), it is possible to explore the functional reorganization
which occurs in patients with AVMs. f MRI investigates
brain function by mapping regional changes in blood
oxygenation levels, in blood flow and/or volume in areas
proximal to activated neurons. Therefore, any significant
perturbation in brain hemodynamics would impact on f MRI
results. AVMs dramatically alter normal cerebral circulatory
dynamics even at sites remote from the AVM as well as in the
contralateral hemisphere [2–9].

⁎ Corresponding author. Tel.: +39 06 49914719; fax: +39 06 49914903.
E-mail address: francesca.caramia@uniroma1.it (F. Caramia).
0730–725X/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.mri.2009.05.029

Transcranial magnetic stimulation (TMS) is able to
explore directly the electrical activity of cortical neurons
independently from cerebral hemodynamics, thus providing
information exclusively on stimulated corticospinal
projections.
We studied the effects of reorganization in a patient
suffering from a right rolandic AVM, by using a combined
exploration which included both TMS and f MRI, in order to
achieve a comprehensive pattern of electrical and morphological changes.

2. Case report
A 33-year-old right-handed woman was admitted at our
hospital for a sudden onset of paraesthesias of her left hand
followed by loss of consciousness.
On admission, her neurological exam showed only mild
clumsiness at left finger movements, but normal strength and
sensitivity. She underwent a brain MRI that revealed the
presence of an AVM extending from the right frontoparietal

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

to the temporal lobe. Cerebral angiography confirmed the
diagnosis of arteriovenous malformation nourished by the
middle cerebral artery.
She was then submitted to both f MRI and TMS to record
motor evoked potentials (MEPs).
Five right-handed healthy age-matched volunteers, who
underwent the same exploration procedures with f MRI and
TMS, represented the control population.
All subjects were informed about the nature of the study
and gave their written consent. This study was conducted in
accordance with the ethical standards of the Committee on
Clinical Investigation on Human Experimentation at our
institution and with the Helsinki Declaration of 1964.
3. Methods
3.1. Functional MRI
f MRI data were acquired using a 1.5-T magnet (Gyroscan
NT Intera; Philips Medical Systems, Best, The Netherlands)
with echoplanar capabilities and a head volume radiofrequency coil. T2*-weighted echo planar images (64×64
matrix over a 24-cm field of view) consisting of 25
consecutive, 4-mm-thick axial sections, with TR/TE=3000/
50 ms, a 90°flip angle and one excitation were then acquired.
During the f MRI acquisition, the subjects were asked to
perform a self-paced sequential finger-to-thumb opposition
task. Two f MRI trials (one for each hand) were obtained
during the motor task.
After the f MRI study, proton density and T2-weighted
spin-echo images (TR=2500 ms; TE=20/90 ms; two
excitations) were acquired with the same geometry of the
f MRI study (same slice orientation, contiguous 4-mm
slices). f MRI data were analysed using SPM99 software
(Wellcome Dept. of Cognitive Neurology, Institute of
Neurology, London) according to the following procedure.
In a preliminary analysis, patient's data were realigned and
smoothed but not normalized and activation foci were
superimposed on patients original T2-weighted images, so as
to evaluate the exact spatial relationship between the AVM
and the patients cortical areas. This preliminary procedure
was aimed at investigating possible distortion caused by the
malformation. After confirmation that no anatomical distortion was caused by the AVM for the final analysis images
were realigned, normalized and spatially smoothed using a
Gaussian kernel of 12 mm.
Statistical analyses were then performed using the
principles of the general linear model extended to allow
the analysis of f MRI data as a time series [10,11].
Statistical tests were then performed to determine signal
changes that were significantly related to hand movement.
Significance was determined on a voxel-by-voxel basis
using a t statistic, which was then transformed into a
normal distribution. Within each region of statistical
significance, local maxima of signal increase were determined (the voxels of maximum significance), and their

1361

location was expressed in coordinates (x, y and z) in the
space of Tailarach and Tournoux [12]. Results were
displayed using a statistical threshold of Pb.05 corrected
for multiple measures across the whole brain.
3.2. Transcranial magnetic stimulation
Stimulation of the cortex was performed by a Magstim 200
stimulator, connected to a figure-of-eight coil applied over the
hand motor area (M1). The coil was held in a lateral orientation
(45° from the midline) and moved on each hemiscalp over the
central sulcus until the optimal site (hot spot) for eliciting
motor responses in the contralateral hand target muscles,
during relaxation, was localized at about 5 to 6 cm lateral to the
vertex (indicated as Cz according to the 10–20 International
system).
Ipsilateral MEPs were sought first by using the same spot as
for contralateral MEPs (cMEPs) then moving the coil to a
more medial region, with the center of the coil positioned at 3
cm lateral and 3 cm anterior to Cz, according to the
methodology established in a study of Caramia et al. [13].
All MEP recordings were acquired by using 80% intensity of
stimulator's output. This procedure was chosen in order to
avoid excessive stimulation. Our purpose was then focused on
picking up a large motor activation, including ipsilateral MEPs
(iMEPs), attesting to the process of reorganization expected in
the patient with respect to control subjects. Contralateral MEPs
were simultaneously obtained. Therefore, measurements
concerning cMEPs were also performed on recordings
obtained (outside M1) from the stimulation of the “ipsilateral
area” at 3 cm anterior and 3 cm lateral to Cz, with the same
TMS intensity used for producing iMEPs [14].
3.3. MEP recordings
MEPs were recorded from thenar muscles (Opponens
Pollicis) by using surface disk electrodes taped in a bellytendon montage. The contraction of the target muscles
consisted in a transient opposition of the thumbs, performed,
bilaterally, at about 40% of maximal force (as measured via a
force transducer). An acoustic feedback was also provided
from the recording electrodes in order to monitor the
electromyography (EMG) background. Recordings, averaged
and replicated at least three times using a suitable amplitude
calibration, were obtained with a Multibasis Esaote at a filter
bandwidth and sampling rate of, respectively, 20–2000 Hz and
10 KHz and stored on floppy disks.
3.4. Transcranial magnetic stimulation data analysis
With TMS delivered at 3 cm lateral and 3 cm anterior to
Cz, the following measurements were performed:
1. MEP latency onset measured at the beginning of the
first reproducible negative deflection.
2. MEP amplitude measured from peak to peak.
3. MEP duration considered from the latency onset of the
potential to its return to the baseline.

1362

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

4. Silent period (SP) duration (ms) from the end of the
MEP to the onset of EMG ongoing activity.
4. Results
Both f MRI and TMS yielded congruous results with
respect to the position of local maxima of f MRI activation
and hot spots for MEPs, involvement of ipsilateral activation
and changes in both left (unaffected) and right (affected)
hemispheres. These are principally characterized by: (1)
presence of activation within the unaffected sensorimotor
area ipsilateral to the moving hand and presence of iMEPs by
stimulating the corresponding scalp sites; (2) functional
displacement within the affected motor areas corresponding
to polyphasic MEPs, attesting to reorganization and impulse
scattering; (3) significant activation in secondary motor areas
in both hemispheres.
f MRI has shown a significant alteration and enlargement
of cortical motor areas, which was predominant in the
unaffected hemisphere with a much greater ipsilateral
activity than normal. TMS performed in positions corresponding to local maxima of f MRI analysis gave rise to a
pattern of motor activation which matched the blood
oxygenation level-dependent (BOLD) results, thus documenting that ipsilateral motor pathways extended from the
unaffected left hemisphere to the left hand.
4.1. Functional MRI
Conventional T2-weighted MR images of the patient
confirmed the presence of a large arteriovenous malformation located in the rolandic (precentral) region in the right
hemisphere extending inferiorly to the right temporal lobe
(Fig. 1). The conventional proton density and T2 weighted

images also showed small foci of hyperintensity within the
AVM, interpreted as perinidal gliosis; no signs of hemorrhage were observed.
4.1.1. Controls hand task
The f MRI study of control subjects revealed activation,
which was predominantly located in the contralateral
sensory-motor and pre-motor areas and in the ipsilateral
cerebellum during either the movement of the right or of the
left hand. Small foci of activation were also observed in the
ispilateral motor areas, during the left hand movement in
three of five subjects. A small cluster of activation was also
observed in the ipsilateral premotor area during left hand
movement in one control subject (#3) (Tables 1 and 2, Fig. 2
and Fig. 3).
4.1.2. Patient left hand task
In the patient, f MRI analysis demonstrated a large
activation area which was mostly located in the ispilateral
hemisphere during the left hand task, involving all motor
areas (sensory-motor, pre-motor, supplementary motor,
inferior parietal lobule and insula) and extended anteriorly
and posteriorly with respect to the central sulcus (3 cm
anteriorly and 2 cm posteriorly). The contralateral (right)
hemisphere showed smaller and more scattered areas of
significant activation, involving all motor areas and located
above, superiorly, and posteriorly with respect to the AVM,
both posteriorly and anteriorly with respect to the central
sulcus (about 1 cm anteriorly and posteriorly). No BOLD
signal could be detected within the nidus of the AVM (Fig. 1
and Fig. 2, Table 1).
4.1.3. Patient right hand task
In the patient, movement of the right hand determined a
prevalent large contralateral activation, located in all motor

Fig. 1. Activation areas obtained during either the left-hand (A) or the right-hand (B) movement superimposed on T2-weighted images in the patient harboring a
right rolandic arterovenous malformation. Activation was predominantly in the left (unaffected) hemisphere during either hands movement. Smaller and more
scattered foci of activation were observed around the AVM prevalently during the left hand movement.

Table 1
Locations of significant activation in controls and patient during left-hand movement
Region
(BA)

R premotor (6)
R parietal (40)

Patient

#1

#2

#3

Talairach
coordinates
(x, y, z)

Z

k

Talairach
coordinates
(x, y, z)

Z

48, −21, 45
44, −12, 59
40,−36,53
30, −14, 69

6.05
5.43

100
53

50, −19, 53

5.41

7.06

137

6.05
5.72

50, −40, 43
40, −36, 53

22, −2, 60
32, −16, 69

5.91
5.50

63
14

k

Z

k

Talairach
coordinates
(x, y, z)

Z

k

Talairach
coordinates
(x, y, z)

Z

k

Talairach
coordinates
(x, y, z)

Z

34

51, −11, 52

6.88

256

32, −21, 49

5.95

57

57, −17, 41
32,−21,50

5.77

11

38, −19, 47
59, 2, 9

7.36
6.01

685
20

25
17

40, −8, 64

6.69

352

32, −9, 49

5.98

82

7.08
6.83
6.49
6.02
5.78
5.38
6.31
6.01
5.13
7.64
6.80
Inf
5.69
Inf
7.75
5.19
Inf

576
55
90
18
12
17
72
38
21
946
77
1671
18
1352
1184
18
855

L premotor (6)
L parietal (40)
L insula
L cerebellum
Vermis

7.60

360

−63,9,23
−40,−34,66
−63, 9, 23

5.46
5.55
5.46

10

−22, −57, −19

6.31

99

−22, −57, −14

6.54

434

−32, −63, −18

6.40

56

36, −19, 49
53, 15, 36
32, −53, 62
30, −33, 40
34, −31, 37
28, −36, 48
59, 4, 11
38, −17, 12
37, 10, 0
26, −67, −20
−8, 15, 66
−51, −19, 51
−36, −33, 70
−59, 9, 27
−44 −32, 50
−57, 23, −11
−16, −65, −15

0, −67, −13

6.99

16

−14, −47, −14
−22, −59, −11

5.83
5.55

20
13

−4, −66, −3

7.17

34

−13, −65, −14

44, 2, −7
22, −54, −26

5.17

35

−44, −25, 40

5.27

12

−56, −22, 29

5.48

16

−20, −50, −23
−20, −71, −17
14, −51, −14

6.64
5.46
5.31

415
35
12

−26, −51, −18

6.71

#5

Talairach
coordinates
(x, y, z)

R insula

R cerebellum
SMA (6)
L sensory−motor (1–4)

#4

227

5.54

19

37

44, −50, 52
45,−48,52

6.57

45

36, −11, 12

6.09

22

40, −54, −24

5.92

19

−65, −13, 15

6.29

13

k

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

R sensory–motor (1–4)

Controls

BA, Brodmann area; R, right; L, left; SMA, supplementary motor area; Z, voxel-level corrected P value N.05; k, number of voxels, extent threshold K N10 voxels.

1363

1364

Table 2
Locations of significant activation in controls and patient during right-hand movement
Region (BA)

Controls

Patient

#1

#2
Z

k

L sensory-motor (1–4)

−51, −13, 49

5.47

31

L pre-motor (6)
L parietal (40)
L cingulated
(25–31)
L insula

−43, −13, 58
−56,−22,29

5.39
5.48

29
16

L cerebellum
SMA (6)

−22, −73, −18
−4, 14, 56
2, 16, 54

5.42
5.44
5.30

27
27
11

#4

Talairach
coordinates
(x, y, z)

Z

k

Talairach
coordinates
(x, y, z)

Z

−46,−24,66
−57, −19, 47

6.13
5.92

73
51

−44,−11,61

5.86

−42, −12, 63

5.79

−28, −50, −19

5.95

k

#5

Talairach
coordinates
(x, y, z)

Z

29

−40, −18, 58

6.44

11

125

−38, −12, 61

6.16

78

−20, −53, −19

6.64

271

90

k

Talairach
coordinates
(x, y, z)

Z

k

Talairach
coordinates
(x, y, z)

Z

k

−40, −14, 62

5.80

29

−55, −16, 34

7.22

1199

6.99
7.75
5.64
5.29
6.53
6.00
7.11
6.23
5.63
5.00
5.07
6.99
6.92
6.45
5.54
6.23

323
1184
53
11
333
63
230
130

13

−36,−16,65
−44, −32, 50
−10, −65, 12
−8, 15, −9
−63, 8, 10
−57, −17, 16
−24, −63, −20
−6, 13, 62
2, 9, 62
2, 11, 58
34, −21, 49
36, −16, 65
16, −4, 72
53, 13, 34
55, 6, 11
24, −53, −19
2, −65, −13

5.47

13

R sensory-motor (1–4)
R premotor (6)

R insula
R cerebellum

14, −59, −16

7.01

522

Vermis

12, −61, −10

6.22

33

12, −61, −15

5.73

49

20, −67, −17

6.46

270

0,−67,−13

5.48

77

22, −59, −15
36, −83, −21
0, −63, −10
−14, −71, −13
−6, −80, −11

6.65
5.75
6.25
5.41
5.09

314
91
108
11
12

38, −52, −24

5.64

10
250
38
272
84
484

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

Talairach
coordinates
(x, y, z)

#3

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

1365

Fig. 2. Activation areas obtained during the left hand movement in the patient (A) and in a representative control subject (B) superimposed on a template volume
rendering (both fMRI data and brain surface are normalized). Images show a predominant ipsilateral activation and a larger contralateral activation in the patient
as compared to the control subject.

areas, including the cingulated gyrus. Smaller clusters of
activation were seen in the ispilateral hemisphere, in the
sensory-motor area, in a region anterior to the central sulcus
(premotor area) and in the insula (Fig. 1 and Fig. 3, Table 2).
4.2. Transcranial Magnetic Stimulation
4.2.1. Controls
In 1 out of 5 control subjects only iMEPs could be elicited
by stimulating the left hemisphere over pre-motor areas
(3 cm lateral and 3 cm anterior to Cz), whilst no iMEPs could

be obtained during stimulation of M1. Such ipsilateral
responses showed remarkably smaller amplitudes than those
obtained in the patient and than cMEPs in all cases (Table 3).
No iMEPs were obtained during stimulation of the
right hemisphere.
4.2.2. Patient
In the patient, large ipsilateral MEPs were recorded in
response to stimulation of both hemispheres (Fig. 4 and
Fig. 5). In contrast to normal subjects, the elicitation of such
responses, was obtained not only by stimulating premotor

Fig. 3. Activation areas obtained during the right hand movement in the patient (A) and in a representative control subject (B) superimposed on a template
volume rendering (both fMRI data and brain surface are normalized). Images show a larger contralateral activation in the patient as compared to the control and a
small cluster of ipsilateral activation which is not evident in the control subject.

1366

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

Table 3
TMS in control subjects and in the patient
M1

Controls

Patient

Left hemisphere

Right hemisphere

Left hemisphere
(unaffected)

Right hemisphere
(affected)

Parameters measured

iMEPs

iMEPs

iMEPs

iMEPs

iMEPs

cMEPs

iMEPs

cMEPs

Amplitude (mV)
Latency (ms)
MEP duration (ms)
SP duration (ms)

/
/
/
/

8.1±0.8
19.4±1.5
39.7±5
135±10

/
/
/
/

7.9±0.6
19.2±1.7
40.2±4.1
139±8.7

4.2±0.5
25.8±0.8
16.6±2.5
20.7±9

12.4±0.5
19.6±0.8
36.6±3.7
142±19

2.2±0.7
20.8±0.5
25±10
16.6±8

13.5±1.9
18±1.2
38.9±5.0
147±25

Parameters measured

iMEPs

cMEPs

iMEPs

cMEPs

iMEPs

cMEPs

iMEPs

cMEPs

Amplitude (mV)
Latency (ms)
MEP duration (ms)
SP duration (ms)

0.81±1.7
22.8±3
22.3±1
21±11

6.6±1.3
18.9±1.7
33.3±2
145±10

/
/
/
/

6.2±1.1
18.4±1.5
32.9±3
146±9.2

2.6±0.3
20.7±0.4
19.6±2.5
29.1±11

13.2±1.3
19.1±0.5
37.0±5.0
140±15

/
/
/
18.3±5

10±0.2
17.7±0.3
34.9±5.1
138±18

Premotor

/, recordings with MEPs not identifiable from background EMG activity. Measurements with means±S.D. are obtained during muscle contraction.

areas (position b), but also during stimulation of primary
motor areas (position a) when exploring the unaffected left
hemisphere, thus paralleling the emergent f MRI results. The
stimulation of the right affected hemisphere gave rise to
ipsilateral iMEPs from M1 whilst from the premotor site a
poorly synchronized ipsilateral activity followed by a silent
period could only be recorded (Fig. 4).
Large cMEPs, were obtained in both positions when
stimulating both hemispheres. In particular, it is important to
point up that amplitudes of cMEPs were all up to two times
larger than normal and that cMEPs recorded in response to

Fig. 4. TMS of the right (AVM) hemisphere. Traces displayed in the upper
panel (A) relate to the stimulation of M1, where cMEPs (upper traces)
morphology appear greatly poliphasic, reflecting the impulse scattering
through the diseased/reorganized cortex; ipsilateral MEPs (lower traces) are
also recorded, in contrast to normal, but with a lower amplitude than
ipsilateral MEPs obtained from left (unaffected) hemisphere, as displayed in
Fig. 5. Traces displayed in the lower panel (B) relate to stimulation of the
premotor cortex, where cMEPs (upper traces) are still poliphasic, although
less than in M1; ipsilateral MEPs (lower traces) are not so clearly identified
from the background EMG activity.

the affected right hemisphere showed a polyphasic morphology, i.e., loss of the normal triphasic morphology (Fig. 4); by
stimulating the unaffected side cMEPs retained a normal
morphology (Fig. 5; Table 3).

5. Discussion
In the present study, both f MRI and TMS testify a motor
cortex reorganization in the context of a right rolandic
arteriovenous malformation, consisting in a pattern of

Fig. 5. TMS of the left (unaffected) hemisphere. Traces displayed in the
upper panel (A) relate to the stimulation of M1, where iMEPs (upper traces)
amplitude is the largest ever obtained amongst all iMEPs, recorded
throughout all the stimulation sites on both hemispheres. This striking
degree of ipsilaterality matches the predominant ipsilateral activation
obtained during left hand movement in fMRI (Fig. 2). Contralateral MEPs
(lower traces) were larger than normal, and even more so than iMEPs. Traces
displayed in the lower panel (B) relate to stimulation of the premotor cortex,
where iMEPs (upper traces) showed a lower amplitude than those recorded
from M1; also in this position, cMEPs (lower traces) were both greater than
normal and than ipsilateral ones.

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

predominant ipsilateral activity when moving the hand
opposite to the affected hemisphere. The combination of the
two techniques here allowed a wider view of motor
information regarding our patient, who did not display
motor deficits while harboring the AVM where motor areas
are normally located and functionally activated.
The paradigm of crossed association, known as the
Valsalva doctrine can be altered due to damage or disease,
and the manifestation of ipsilateral activity reflecting a
functional relationship between one hemisphere and the
limbs on the same side of the body has been demonstrated in
neurological patients during motor recovery [15–19].
Activation in ipsilateral motor cortex extended well
beyond M1, including pre-motor, supplementary motor,
parietal areas and insula. These results and the concurring
generation of iMEPs both inside and outside M1 reflect an
increased outflow of motor areas stimulation [20,21].
Ipsilateral activity expressed both by iMEPs and by
f MRI activation foci could be also recorded in some
control subjects, but on the left side only. This is in
agreement with several studies and possibly reflects a
physiological background of “ipsilaterality” with a bias
towards left hemisphere bilateral motor control, showing
that ipsilateral hemispheric activation is significantly
greater with non-dominant (left) than dominant (right)
hand motor tasks [22–24].
The presence of ipsilateral corticospinal connections
appears to be a normal state in ontogeny, as demonstrated
by a TMS study in healthy children. After cerebral damage,
ipsilateral tracts may become unmasked by a lack of
inhibitory control from the affected hemisphere or callosal
output [25]. Contralateral activity was also recorded in our
case, during movement of the left hand. Reorganization of
contralateral cortical areas following injury to the adult
brain has been described in several diseases, including
stroke, multiple sclerosis, amyotrophyc lateral sclerosis and
brain tumors. For lesions affecting the sensory-motor
cortex, cortical changes classically involve enlargement of
the limb cortical representation, anterior or posterior shift of
the center of maximal activation, recruitment of other areas
such as the insula and the parietal regions [16,26–31].
It has thus been proposed that the cortical generation and
control of movements at the cortical level are mediated by
parallel outputs from the nonprimary cortex as well as from
the primary motor cortex [32]. As a consequence, several
cortical neuron populations have the potential to influence
the control of voluntary movements, independent of the
proper M1.
Recovery of function after central nervous system
damage depends on the maturity of the brain at the moment
the damage is incurred, being generally greater when brain
damage occurs early in life. These observations suggest that
remaining areas of the brain are able to take over behavioral
functions that normally occur in the damaged areas and that
the brain possesses a greater ability to compensate in its
immature than in its mature state. Our results are in

1367

accordance with findings of large-scale reorganization
observed in primate studies. Rouiller et al. [33] demonstrated
that both the stage of development at the time of an induced
lesion and the extent of the lesion can play a major role in
triggering different plastic changes contributing to the
preservation of motor functions. In the immature brain, in
which corticospinal projections have not yet reached their
targets in the cervical cord, lesions in M1 mainly resulted in a
functional shift to the ipsilateral, intact hemisphere. When
lesions in M1 were induced in the mature brain, in contrast, a
complete new map of hand and arm representation was
found in a region adjacent to the lesion.
Although AVM are generally considered inborn errors of
vascular morphogenesis, the exact period in which brain
vascular malformations occur is still controversial. There is
increasing evidence that they may develop postnatally,
thereby representing a complex endothelial cell dysfunction
[34]. The extensive contribution of the ispilateral, nonaffected hemisphere to the execution of a motor task in our
case suggests that the AVM may have been present early in
the life of the patient. Furthermore, AVMs are often
followed by later microhemorrhages or hemodynamic
alterations that may have been responsible for later
seizure-like activity.
Another aspect raised from our study is that no BOLD
signal was present within the nidus of the AVM. As already
observed by previous investigators, this condition could be
explained by the histopathologic findings that the AVM's
feeding vessels are separated from brain parenchyma and,
therefore, do not participate to metabolic changes that occur
during neuronal activity.
Alkhadi et al. [2] suggest that the conflicting results
regarding the detection of BOLD signal within the nidus, as
reported in some previous works, have to be related to the
difficulty in distinguishing the exact border of the nidus so
that the intervening brain interposed between the distal
feeding and the proximal draining vessels could be mistaken
for intranidal activation.
In our study, the aspects of cortical reorganization were
concordantly demonstrated by both f MRI and TMS.
f MRI is a technique able to map functional activation by
detecting changes in regional blood oxygenation levels,
blood flow and volume. It is therefore extremely influenced
by hemodynamic perturbations. A different characteristic of
the BOLD response in the vascular network supporting the
neuronal population of the damaged hemisphere may have
contributed to the results. With respect to f MRI, TMS
defines reorganization in terms of neural activity by
inducing the activation of cortical regions, which manifests
itself, directly, under the form of motor responses recorded
in the target muscles. f MRI and TMS provide complementary information as to the status of the motor cortex. To
optimize the meaning of the combined exploration, one has
to consider, however, that activation visualized in f MRI
parametric maps extended obviously beyond the foci
explored with TMS. The hemodynamic response tends to

1368

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369

be more widespread in space and lasts longer in time as
compared with the neuronal activity. This may be due to the
characteristics of f MRI that preferentially provides information about intracortical processing or input to brain regions
which are relevant for an actual task rather than the output of
a region [35]. Therefore, when comparing the results of the
two techniques, it should be taken into account the structural
difference of the respective dominions of investigation:
more spatial with f MRI that is like an open window; more
temporal with TMS, where each event instantaneously
related to the motor output is observed in a strictly focused
dimension of space, like through a keyhole.

References
[1] Fleetwood IG, Steinberg GK. Arteriovenous malformations. Lancet
2002;359(9309):863–73.
[2] Alkadhi H, Kollias SS, Crelier GR, Golay X, Hepp-Reymond MC,
Valavanis A. Plasticity of the human motor cortex in patients with
arteriovenous malformations: a functional MR imaging study. AJNR
Am J Neuroradiol 2000;21(8):1423–33.
[3] Latchaw RE, Hu X, Ugurbil K, Hall WA, Madison MT, Heros RC.
Functional magnetic resonance imaging as a management tool for
cerebral arteriovenous malformations. Neurosurgery 1995;37(4):
619–25 [discussion 625-616].
[4] Lehericy S, Biondi A, Sourour N, Vlaicu M, du Montcel ST, Cohen L,
et al. Arteriovenous brain malformations: is functional MR imaging
reliable for studying language reorganization in patients? Initial
observations. Radiology 2002;223(3):672–82.
[5] Maldjian J, Atlas SW, Howard II RS, Greenstein E, Alsop D, Detre
JA, et al. Functional magnetic resonance imaging of regional brain
activity in patients with intracerebral arteriovenous malformations
before surgical or endovascular therapy. J Neurosurg 1996;84(3):
477–83.
[6] Schlosser MJ, McCarthy G, Fulbright RK, Gore JC, Awad IA. Cerebral
vascular malformations adjacent to sensorimotor and visual cortex.
Functional magnetic resonance imaging studies before and after
therapeutic intervention. Stroke 1997;28(6):1130–7.
[7] Thulborn KR, Davis D, Erb P, Strojwas M, Sweeney JA. Clinical
f MRI: implementation and experience. Neuroimage 1996;4(3 Pt 3):
S101–107.
[8] Valavanis A. The role of angiography in the evaluation of cerebral
vascular malformations. Neuroimaging Clin N Am 1996;6(3):
679–704.
[9] Valavanis A, Yasargil MG. The endovascular treatment of brain
arteriovenous malformations. Adv Tech Stand Neurosurg 1998;24:
131–214.
[10] Friston KJ, Frith CD, Frackowiak RS. Principal component analysis
learning algorithms: a neurobiological analysis. Proc Biol Sci 1993;
254(1339):47–54.
[11] Friston KJ, Holmes AP, Poline JB, Grasby PJ, Williams SC,
Frackowiak RS, et al. Analysis of f MRI time-series revisited.
Neuroimage 1995;2(1):45–53.
[12] Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain:
3 dimensional proportional system; an approach to cerebral imaging.
Stuttgart: George Thieme Verlag; 1988.
[13] Caramia MD, Telera S, Palmieri MG, Wilson-Jones M, Scalise A, Iani
C, et al. Ipsilateral motor activation in patients with cerebral gliomas.
Neurology 1998;51(1):196–202.
[14] Caramia MD, Palmieri MG, Giacomini P, Iani C, Dally L, Silvestrini
M. Ipsilateral activation of the unaffected motor cortex in patients with
hemiparetic stroke. Clin Neurophysiol 2000;111(11):1990–6.

[15] Lee M, Reddy H, Johansen-Berg H, Pendlebury S, Jenkinson M, Smith
S, et al. The motor cortex shows adaptive functional changes to brain
injury from multiple sclerosis. Ann Neurol 2000;47(5):606–13.
[16] Nyberg G, Andersson J, Antoni G, Lilja A, Pellettieri L, Valind S,
et al. Activation PET scanning in pretreatment evaluation of
patients with cerebral tumours or vascular lesions in or close to
the sensorimotor cortex. Acta Neurochir (Wien) 1996;138(6):
684–94.
[17] Reddy H, Narayanan S, Arnoutelis R, Jenkinson M, Antel J, Matthews
PM, et al. Evidence for adaptive functional changes in the cerebral
cortex with axonal injury from multiple sclerosis. Brain 2000;123
(Pt 11):2314–20.
[18] Rocca MA, Falini A, Colombo B, Scotti G, Comi G, Filippi M.
Adaptive functional changes in the cerebral cortex of patients with
nondisabling multiple sclerosis correlate with the extent of brain
structural damage. Ann Neurol 2002;51(3):330–9.
[19] Vinas FC, Zamorano L, Mueller RA, Jiang Z, Chugani H, Fuerst D,
et al. Diaz FG. [15O]-water PET and intraoperative brain mapping: a
comparison in the localization of eloquent cortex. Neurol Res 1997;19
(6):601–8.
[20] Leblanc E, Meyer E, Zatorre R, Tampieri D, Evans A. Functional
PET scanning in the preoperative assessment of cerebral arteriovenous malformations. Stereotact Funct Neurosurg 1995;65(1-4):
60–4.
[21] Morioka T, Yamamoto T, Mizushima A, Tombimatsu S, Shigeto H,
Hasuo K, et al. Comparison of magnetoencephalography, functional
MRI, and motor evoked potentials in the localization of the sensorymotor cortex. Neurol Res 1995;17(5):361–7.
[22] Bastings EP, Gage HD, Greenberg JP, Hammond G, Hernandez L,
Santago P, et al. Co-registration of cortical magnetic stimulation and
functional magnetic resonance imaging. Neuroreport 1998;9(9):
1941–6.
[23] Johansen-Berg H, Rushworth MF, Bogdanovic MD, Kischka U,
Wimalaratna S, Matthews PM. The role of ipsilateral premotor cortex
in hand movement after stroke. Proc Natl Acad Sci U S A 2002;99(22):
14518–23.
[24] Lotze M, Kaethner RJ, Erb M, Cohen LG, Grodd W, Topka H.
Comparison of representational maps using functional magnetic
resonance imaging and transcranial magnetic stimulation. Clin
Neurophysiol 2003;114(2):306–12.
[25] Caramia MD, Iani C, Bernardi G. Cerebral plasticity after stroke as
revealed by ipsilateral responses to magnetic stimulation. Neuroreport
1996;7(11):1756–60.
[26] Baumann SB, Noll DC, Kondziolka DS, Schneider W, Nichols TE,
Mintun MA, et al. Comparison of functional magnetic resonance
imaging with positron emission tomography and magnetoencephalography to identify the motor cortex in a patient with an arteriovenous
malformation. J Image Guid Surg 1995;1(4):191–7.
[27] Carpentier AC, Constable RT, Schlosser MJ, de Lotbiniere A,
Piepmeier JM, Spencer DD, et al. Patterns of functional magnetic
resonance imaging activation in association with structural lesions in
the rolandic region: a classification system. J Neurosurg 2001;94(6):
946–54.
[28] Morioka T, Mizushima A, Yamamoto T, Tobimatsu S, Matsumoto S,
Hasuo K, et al. Functional mapping of the sensorimotor cortex:
combined use of magnetoencephalography, functional MRI, and motor
evoked potentials. Neuroradiology 1995;37(7):526–30.
[29] Mueller WM, Yetkin FZ, Hammeke TA, Morris III GL, Swanson SJ,
Reichert K, et al. Functional magnetic resonance imaging mapping of
the motor cortex in patients with cerebral tumors. Neurosurgery 1996;
39(3):515–20 [discussion 520-511].
[30] Nishiyama Y, Yamamoto Y, Fukunaga K, Satoh K, Ohkawa M,
Kunishio K, et al. Visualization of the motor activation area using
SPECT in neurosurgical patients with lesions near the central sulcus.
J Nucl Med 2000;41(3):411–5.
[31] Schreiber A, Hubbe U, Ziyeh S, Hennig J. The influence of gliomas
and nonglial space-occupying lesions on blood-oxygen-level-

F. Caramia et al. / Magnetic Resonance Imaging 27 (2009) 1360–1369
dependent contrast enhancement. AJNR Am J Neuroradiol 2000;21(6):
1055–63.
[32] Dum RP, Strick PL. The origin of corticospinal projections from the
premotor areas in the frontal lobe. J Neurosci 1991;11(3):667–89.
[33] Rouiller EM, Yu XH, Moret V, Tempini A, Wiesendanger M, Liang F.
Dexterity in adult monkeys following early lesion of the motor cortical
hand area: the role of cortex adjacent to the lesion. Eur J Neurosci
1998;10(2):729–40.

1369

[34] Stapleton SR, Kiriakopoulos E, Mikulis D, Drake JM, Hoffman HJ,
Humphreys R, et al. Combined utility of functional MRI, cortical
mapping, and frameless stereotaxy in the resection of lesions in
eloquent areas of brain in children. Pediatr Neurosurg 1997;26(2):
68–82.
[35] Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A.
Neurophysiological investigation of the basis of the f MRI signal.
Nature 2001;412(6843):150–7.