Electroencephalography and clinical Neurophysiology , 85 (1992) 110-115 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0924-980X/92/$05.00
110
ELMOCO 90999
Percutaneous magnetic stimulation of the motor cortex in migraine A. Maertens de Noordhout, J.L. Pepin, J. Schoenen and P.J. Delwaide Headache Clinic, University Department of Neurology, H~pital de la Citadelle, 4000 Liege (Belgium) (Accepted for publication: 14 June 1991)
Summary We have used transcranial magnetic stimulation of the motor cortex interictally in 12 patients with unilateral classic migraine with sensorimotor auras and 10 patients with common migraine and unilateral headache. In classic migraine, the threshold of activation of the FDI muscle by the cortical stimulus was significantly increased on the side of the auras, when compared to the unaffected side (P < 0.01) and to normal subjects ( P < 0.01). The amplitude of EMG responses was also reduced in FDI on the affected side when compared to normals (P < 0.02). Responses obtained in common migraine patients were normal on both sides. We suggest that some permanent subclinical dysfunction of the motor cortex might play a role in the pathogenesis of attacks of classic migraine with sensorimotor auras. Key words: Cortical stimulation; Migraine; Cortical excitability; Motor evoked potentials
Whether migraine is originally a vascular or a neuronal disorder is debated. Cerebral blood flow studies during induced attacks of classic migraine have shown the existence of spreading cortical oligaemia, usually starting in the occipital cortex and extending anteriorly at a speed of 2-3 m m / m i n (Olesen et al. 1981; Lauritzen et al. 1983). The reduced regional cerebral blood flow (rCBF) observed in these studies developed independently of the vascular territories of the cerebral cortex, suggesting a neuronal rather than vascular cause for this phenomenon. Such events could be similar to the "spreading cortical depression" (SCD) which can be experimentally induced in animals (Leap 1944). Spreading oligaemia was absent ictally in common migraine (Olesen et al. 1981). It was then proposed that permanent dysfunction of some cortical neurones might be responsible for the migraine attacks. Using 31p_NMR spectroscopy, Ramadan et al. (1989) have recently measured low brain magnesium levels in migraineurs, which could facilitate the induction of SCD. Indeed, it has been shown that glutamate induced SCD is facilitated by low Mg 2÷ levels (Van Harreveld 1984), which increase the sensitivity of NMDA receptors (Ault et al. 1980; Nowak et al. 1984). In recent years, methods of percutaneous electrical (Merton and Morton, 1980) and magnetic (Barker et al. 1985) stimulation of the motor cortex have been
Correspondence to: Dr. Alain Maertens de Noordhout, University Department of Neurology, H6pital de la Citadelle, Bd du 12e de Ligne, 4000 Liege (Belgium). Tel.: 3241256401.
developed. These allow non-invasive study of the function of the motor cortex and fast-conducting corticospinal pathways in fully conscious man. The magnetic device delivers painless stimuli, which facilitates its use for clinical studies. We used this new technique in migraine in order to disclose possible modifications of motor cortex function between attacks.
Patients and methods
Twelve patients with pure unilateral classic migraine or migraine with aura (10 women, 2 men; mean age: 36.3 _+ 13.7 years; mean disease duration: 13.7 ___8.6 years) and 10 patients (7 women, 3 men; mean age: 38.8 _+ 10.6 years; mean disease duration: 16.3 +_8.5 years) suffering from unilateral common migraine or migraine without aura were studied between attacks and compared with 20 age-matched controls (16 women, 4 men; mean age: 39.6 +_ 13.8 years). Subjects gave fully informed consent and approval of the Local Ethical Committee was obtained. All classic migraine patients experienced sensorimotor auras with visual and speech disturbances in some of them. Neurological symptoms always occurred on the same side of the body, but the subsequent headache was not always strictly contralateral to the side of the aura. Details of the history of the individual patients are given in Table I. Patients with common migraine had attacks of hemicrania always located on the same side of the scalp. Characteristics of patients with common migraine are summarized in Table II. Patients included in either
111
CORTICAL STIMULATION IN MIGRAINE TABLE I Clinical history of patients with classic migraine. Patient
Age (years)
Sex
Disease duration
Attacks frequency
D.G.
37
F
18
4/year
L.S.
15
M
2
3/year
A.M.
39
F
21
5/year
N.G.
45
F
6
2/year
P.E.
69
F
30
10/year
Z.M.
33
F
8
2/year
L.N.
39
F
21
8/year
M.M.
30
F
15
3/year
M.P.
18
F
2
5/year
M.I.
31
F
16
M.C.
39
M
8
5/year
L.J.
40
F
17
6/year
2/month
group fulfill6d the new criteria of the Headache Classification Committee of the IHS (1988) for migraine with aura (classic migraine) or migraine without aura (common migraine). None of the patients had abnormal CT, permanent E E G abnormalities or any other pathological condition. The experimental procedure was conducted at least 1 week after the last migraine attack and in all but 1 patient (L.M., common migraine) there has been a headache-free interval of at least 1 week after the test. None received prophylactic antimigraine therapy at the time of the test, or any drugs that might
Side and type of aura
Side of pain
R. paraesthesia dysphasia L. paresis L. paraesthesia L. scotoma R. paraesthesia R. scotoma L. paresis L. hypoaesthesia L. paraesthesia L. scotoma R. paresis R. scotoma dysphasia L. paraesthesia L. scotoma L. paresis L. paraesthesia L. paraesthesia L. scotoma R. paraesthesia R. scotoma dysphasia L. paraesthesia L. scotoma L. paresis L. paraesthesia L. scotoma
L. and R. R.
L. and R. R. R. L. and R.
R. L. and R. R. L.
L. and R. L. and R.
modify cortical excitability (hypnotics, barbiturates, anticonvulsants). The only treatments allowed were ergotamine or NSAIDs against migraine attacks and oral contraceptives. Magnetic stimulation of the motor cortex was achieved with a Novametrix Magstim 200 with a 9 cm mean diameter stimulating coil held over the vertex, its edge overlying the arm area of the motor cortex. The maximal transient magnetic field produced in this coil is 1.5 Tesla. To stimulate the right hemisphere, the direction of the current in the coil seen from above was
TABLE II Clinical history of patients with unilateral common migraine. Patient
Age (years)
Sex
Disease duration
Attacks frequency
Side of pain
B.L. S.B. D.C. M.A. V.A. F.M. R.J. S.M. L.M. T.T.
64 40 41 31 43 30 41 32 40 26
F F F F M F F M M F
35 15 16 15 20 5 13 8 24 12
3/month 2/month 3/month 4/month 2/month 3/month 1/month 2/month 1/month 3/month
R R L R L R L L L R
112
A. M A E R T E N S D E N O O R D H O U T E T AL.
clockwise and vice versa. In fact, since the last version of the Novametrix stimulator became available in 1990, it appears that the direction of the current in the coil was the reverse of what was stated before (Hess et al. 1987). Responses evoked in the first dorsal interosseous (FDI) muscle of the hand were recorded with surface Ag-AgCI electrodes placed on the belly and tendon of the muscle, some 3 cm apart from each other. Signals were filtered (3 Hz-3000 Hz) and amplified with 2 different gains (100 /zV/V and 2 m V / V , Digitimer D150) to allow accurate detection of the activation threshold and maximal amplitude of the EMG responses. Traces were then stored on floppy disks (Digitimer D200) for off-line analysis. The threshold of activation of FD! by cortical stimuli was measured on both sides, at rest and during isometric background contraction of the target muscle (20% max). Steady voluntary contraction of the muscle results in a marked reduction of its activation threshold (Hess et al. 1987), mainly due to increased excitability of spinal motoneurones (Day et al. 1987). The threshold of activation was defined as the intensity of stimulation able to produce an EMG response in FDI to at least 50% of a series of 16 stimuli. Patients were comfortably seated in an armchair, eyes open. Shortest latency and mean amplitude of the first negative peak of responses to a series of 8 maximal cortical stimuli (CTXmax) were measured on both sides at rest and during background voluntary contraction. Latencies and amplitudes were measured manually, with cursors. Amplitude of the responses was eX-
pressed as percentage of the EMG potentials evoked in FDI by supramaximal stimulation of the ulnar nerve (Mmax), which simultaneously activates all motoneurones supplying the muscle. Calculation of the CTXmax/Mmax ratio ensures that possible amplitude differences between sides are not due to asymmetrical positioning of the recording electrodes or unilateral lesion of the peripheral nervous system. Conduction time in the central motor pathways (CMCT) was calculated by the method proposed by Rossini et al. (1985).
Results
In normal subjects, the mean intensity of the magnetic field (expressed as percentage of the maximal output of the stimulator) able to activate the FDI without voluntary contraction was 48 + 6%. The mean side-to-side difference of activation threshold was 6.8 + 3.5% of the mean value of both sides. This difference was not correlated with handedness. During voluntary isometric contraction (20% max), FDI activation threshold was 33 + 3%. The mean interside difference was 7.9 + 5.9% of the mean. Central motor conduction time (CMCT) from cortex to C7-Thl spinal segment was 6.5 + 0.6 msec on both sides. During voluntary contraction, the C T X m a x / M m a x amplitude ratio was 50.6 + 16.7%. In classic migraine patients, FDI activation thresholds at rest and during voluntary contraction were respectively 55 + 9% and 41 + 8% of the maximal out-
T A B L E III Characteristics of responses recorded in classic migraine patients. Threshold (% max)
CTXmax/Mmax (%)
at rest
contraction
aft. side
nl. side
29 30 41 29 30 29 32 34 28 39 39 37
41 37 27 41 24 25 36 35 36 64 54 19
44 28 38 60 31 37 36 37 20 69 53 44
33 * * 5
37 + 13
41N.S. 14
aft. side
nl. side
aft. side
nl. side
Patient D.G. L.S. A.M. N.G. P.E. Z.M. L.N. M.M. M.P. M.I. M.C. L.J.
51 59 59 42 60 53 63 65 42 50 53 68
51 50 50 33 44 45 54 46 42 51 52 49
28 38 45 40 49 41 34 51 30 40 39 51
Mean S.D.
55 *'+ 9
47 6
41 **'+ 8
Normals Mean S.D.
48 ÷ 6
Wilcoxon: * P < 0.02; * * P < 0.01; N.S. not significant. Unpaired t: + P < 0.05.
33 + 3
51 17
÷
CORTICAL STIMULATION IN MIGRAINE
113 P.E.
put of the stimulator on the side of the body affected by the aura. On the normal side, these values were 47 + 6% and 33 + 5%. The mean interside difference of activation threshold was 15.9 + 12.8% of the mean at rest and 20 + 16.5% during background contraction of the target muscle. The differences of activation thresholds at rest and during contraction between both sides were statistically significant ( P < 0.01, paired t). The mean values recorded on the side affected by the aura were also significantly different from those obtained in normal subjects at rest and during contraction ( P < 0.01, unpaired t). However, 4 patients (33%) showed no side-to-side difference of FDI activation threshold, although their histories and symptoms were not different from those of the other cases. If individual values are considered, the FDI activation threshold at rest was above 3 S.D. from normals in only 1 case (8%), and in 4 cases (33%) during background contraction. Mean CMCT was 6.6 + 0.7 msec on both sides. This value does not differ from that of normals. CTXmax/Mma x ratio was 36.6 12.8% on the affected side and 41.4 + i3.8% on the other (Table III). The former was significantly different from the ratio calculated in the control group ( P < 0.02, unpaired t). A typical example of responses to magnetic cortical stimulation in classic migraine is shown in Fig. 1. No correlation was found between the frequency of attacks or duration of the disease and increased FDI activation threshold on the side of the auras (Spearman rank correlation test). In common migraine, FDI activation thresholds at rest and during backgound contraction were 45 + 6%
.C.t..AS.S.~.~ . _aS_ ).~
f.d.i.
!_
_
|
I __
[~ II
/t
sideof
~ --
/
-
~
A ~ ~ \
aura CORTICAL STIMULATION n~i~eal
~
~
__ .--
2rn---;
s~eof aura ULNARNERVE
-
L.~ ~'/ S
normalSTIMULATION side
Fig. 1. Maximal EMG responses produced interictally in FDI of a classic migraine patient by magnetic cortical (upper traces) and ulnar nerve (bottom traces) stimuli. Responses to cortical stimulation are smaller on the side of the body affected by the aura (4 individual responses superimposed in each trace), while those to ulnar nerve stimuli are identical on both sides. Cortical stimulation was achieved during steady (20% max) background contraction of FDI.
and 33 + 5% of the maximal output of the stimulator on the side of the body contralateral to the side of hemicrania. On the other side, these values were 44 + 7% and 33 + 6% respectively. Interside threshold difference was 3.3 + 3.2% of the mean at rest and 6.1 + 3.5% during contraction. CMCT was 6.5 + 0.7 msec on both sides and C T X m a x / M m a x ratio was 49.4 _+ 9.7% on the"affected side" and 50.2 _+ 9.3% on the other (Table IV). No side-to-side statistically significant difference could be disclosed nor differences from values of the control group. Fig. 2 summarizes the mean
TABLE IV Characteristics of responses recorded in common migraine patients. Threshold (Tesla)
CTXmax/Mma~ (%)
at rest
contraction
aft. side
nl. side
32 21 29 43 40 33 33 35 33 29
50 55 38 42 63 46 48 51 36 65
44 56 46 44 56 48 49 48 39 72
33 6
49 10
50 9
aft. side
hi. side
aft. side
nl. side
Patient B.L. S.B. D.C. M.A. V.A. F.M. R.J. S.M. L.M. T.T.
42 29 52 50 47 42 43 48 45 43
41 27 51 50 52 45 43 45 44 43
33 23 31 42 35 32 32 33 31 31
Mean S.D.
45 6
44 7
33 5
Normal$ Mean S.D.
N.S.
N.S.
N.S.
48 6 N.S.
33 3 N.S.
51 17 N.S.
114
A. MAERTENS DE NOORDHOUT ET AL.
FDI
ACTIVATION % maximal
THRESHOLD output
lO0 rest
80 contraction
60
i L
40
20
I
;ltlJic xf. side
C|IIIiC nL side
normals
common nf.side
common nl.stdc
Fig. 2. Activation thresholds of the FDI muscle at rest (white columns) and during isometric contraction (dashed columns) in the different groups studied. Thresholds at rest and during contraction are significantly increased in classic migraine on the side of the auras when compared to the other side or to normals. No abnormality is observed in common migraine.
values of FDI activation threshold measured in the different groups tested. Finally, the experimental procedure was well tolerated by all subjects and no adverse reactions were reported. Additionally, cortical shocks proved ineffective to induce migraine attacks in this population.
Discussion
The present study was conducted on particular subpopulations of migraine patients. Although they fulfilled IHS criteria for classic (with aura) or common (without aura) migraine, they are in no way typical samples of either affection. The possibility of using the other side as control led us to choose such cases with unilateral symptoms. In the classic migraine group, a significantly stronger magnetic field was needed to activate the first dorsal interosseous muscle of the hand on the side affected by the aura than on the contralateral one. The mean FDI activation threshold on the affected side was also higher than in an age-matched control group. This held true with the target muscle at rest or during voluntary contraction. However, individual values were within the normal range in most patients. For this reason, cortical stimulation should not be considered as a paraclifiical diagnostic tool in this affection. However, the present findings might have pathophysiological implications, which will be detailed further. No significant interside threshold difference was noted in common migraine, nor discrepancy with normal values.
The threshold of activation of a given muscle by motor cortex stimuli depends on several factors. It has been demonstrated that slightly supraliminal magnetic stimuli applied at the vertex excite the fast-conducting pyramidal neurones of the primary motor cortex indirectly, via some cortical excitatory interneurones (Day et al. 1987, 1989). The axons of pyramidal neurones seem to be monosynaptically connected with spinal motoneurones innervating intrinsic hand muscle fibres. The pathway tested by magnetic cortical stimulation is thus at least disynaptic. Increased activation threshold and reduced amplitude of the responses of a given muscle can reflect hypoexcitability of spinal motoneurones, pyramidal cells or cortical interneurones. Additional factors such as unilateral skull thickening or brain atrophy might theoretically play a role, hut they can be ruled out in migraine. It seems unlikely that the excitability of spinal motoneurones is depressed unilaterally in classic migraine, though this possibility has not been studied. However, if this were the cause of interside threshold differences observed in our patients, this difference should be accentuated during voluntary contraction, which normally brings spinal motoneurones to a higher state of excitability. Thus, a higher activation threshold observed in classic migraine rather reflects reduced excitability of large diameter pyramidal cells or cortical interneurones, or hyperactivity in cortical inhibitory structures. There is no definite argument to suggest that cortical hypoexcitability results from brain ischaemia induced by previous attacks. Indeed, reduction of rCBF sufficient to produce structural brain damage has been occasionally measured in episodes of classic migraine (Skyhoj Olsen et al. 1987). However, in this limited series, no correlation was found between electrophysiological abnormalities and duration of migraine history or frequency of attacks. However, episodes of brain ischaemia may have occurred in some patients who were not necessarily those experiencing the most frequent attacks. Our observations contrast with evoked potential studies in migraine, which suggested cortical hyperrather than hypoexcitabiliy (Kennard et al. 1978; Kasprzyk et al. 1984; Milone et al. 1984). At present, we have no straightforward explanation for this discrepancy, except that cerebral structures tested were not the same. In spite of the significant differences found in this study between classic and common migraine, we do not support the idea that these are separate entities with distinct pathophysiological mechanisms. The first argument against this view is certainly the fact ~that all prophylactic and acute treatments are equally active in both forms of migraine. Although rCBF studies seem to yield different results in migraine with and without aura (Olesen et al. 1981), there is thus far no specific
CORTICAL STIMULATION IN MIGRAINE b i o c h e m i c a l or histological m a r k e r specific for either form. F o r these reasons, it is m o r e likely that m i g r a i n e r e p r e s e n t s a c o n t i n u u m with m i g r a i n e w i t h o u t a u r a at o n e e n d a n d m i g r a i n o u s i n f a r c t i o n at the other. T h e y w o u l d be different expressions of a u n i q u e vascular a n d / o r n e u r o n a l dysfunction, driven by d i f f e r e n t individual levels of sympathetic activity ( W e l c h et al. 1988). O u r study also shows that, in a d d i t i o n to C M C T a n d a m p l i t u d e of E M G responses, t h r e s h o l d m e a s u r e m e n t s can detect otherwise u n s u s p e c t e d d y s f u n c t i o n of the m o t o r system.
References Ault, B., Evans, R.H., Francis, A.A., Oakes, D.J. and Watkins, J.C. Selective depression of excitatory amino acid-induced depolarization by magnesium ions in isolated spinal cord preparation. J. Physiol. (Lond.), 1980, 307: 413-428. Barker, A.T., Jalinous, R. and Freeston, I.L. Non invasive magnetic stimulation of the human motor cortex. Lancet, 1985, i: 11061107. Day, B.L., Rothwell, J.C., Thompson, P.D., Dick, J.P.R., Cowan, J.M.A., Berardelli, A. and Marsden, C.D. Motor cortex stimulation in intact man. II. Multiple descending volleys. Brain, 1987, 110: 1191-1210. Day, B.L., Dressier, D., Maertens de Noordhout, A., Marsden, C.D., Nakashima, K., Rothwell, J.C. and Thompson, P.D. Electric and magnetic stimulation of human motor cortex:surface EMG and single motor unit studies. J. Physiol. (Lond.), 1989, 412: 449-473. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia, 1988, 8 (Suppl. 7). Hess, C.W., Mills, K.R. and Murray, N.M.F. Responses in small hand muscles from magnetic stimulation of the human brain. J. Physiol. (Lond.), 1987, 388: 397-420.
115 Kasprzyk, M., Brzecki, A., Podemski, M., Paradowski, B. and Bogdanska, R. Diagnostic value of pattern-reversal visual evoked potentials in vasomotoric headaches. Cephalalgia, 1984, 5 (Suppl. 1): 466. Kennard, C., Gawel, M., Rudolph, N.M. and Clifford Rose, F. Visual evoked potentials in migraine subjects. In: A.P. Friedman, M. Granger and M. Critchley (Eds.), Research and Clinical Studies in Migraine, Vol 6. Karger, Basel, 1978: 73-80. Lauritzen, M., Skyhoj Olsen, T., Lassen, N.A. and Paulson, O.B. The changes of regional cerebral blood flow during the course of classical migraine attacks. Ann. Neurol., 1983, 13: 633-641. Le~o, A.A.P. Spreading depression of activity in cerebral cortex. J. Neurophysiol., 1944, 7: 359-390. Merton, P.A. and Morton, H.B. Stimulations of the cerebral cortex in the intact human subject. Nature, 1980, 285: 227. Milone, F., D'Andrea, G., Cananzi, A. and Zanini, R. Somatosensory evoked potentials patterns in migraine sufferers. Cephalalgia, 1984, 5 (Suppl 1): 470-471. Nowak, L., Bregestowski, P., Ascher, P., Herbert, A. and Prochiantz, A. Magnesium gates glutamate-activated channels in mouse central neurones. Nature, 1984, 307: 462-465. Olesen, J., Larsen, B. and Lauritzen, M. Focal hyperemia followed by spreading oligemia and impaired activation of rCBF in classic migraine. Ann. Neurol., 1981, 9: 344-352. Ramadan, N.M., Halvorson, H., Vande-Linde, A., Levine, S.R., Helpern, J.A. and Welch, K.M.A. Low brain magnesium in migraine. Headache, 1989, 29: 590-593. Rossini, P., Di Stefano, E. and Stanzione, P. Nerve impulse propagation along central and peripheral fast conducting motor and sensory pathways in man. Electroenceph. clin. Neurophysiol., 1985, 60: 320-334. Skyhoj Olsen, T., Friberg, L. and Lassen, N.A. Ischemia may be the primary cause of the neurological deficits in classic migraine. Ann. Neurol., 1987, 45: 154-159. Van Harreveld, A. The nature of the chick's magnesium-sensitive retinal spreading depression. J. Neurobiol., 1984, 15: 333-344. Welch, K.M.A., NageI-Leiby, S. and D'Andrea, G. The biological and behavioural basis of migraine. In: J. Olesen and L. Edvinsson (Eds.), Basic Mechanisms of Headache. Elsevier, Amsterdam, 1988: 447-456.
110
ELMOCO 90999
Percutaneous magnetic stimulation of the motor cortex in migraine A. Maertens de Noordhout, J.L. Pepin, J. Schoenen and P.J. Delwaide Headache Clinic, University Department of Neurology, H~pital de la Citadelle, 4000 Liege (Belgium) (Accepted for publication: 14 June 1991)
Summary We have used transcranial magnetic stimulation of the motor cortex interictally in 12 patients with unilateral classic migraine with sensorimotor auras and 10 patients with common migraine and unilateral headache. In classic migraine, the threshold of activation of the FDI muscle by the cortical stimulus was significantly increased on the side of the auras, when compared to the unaffected side (P < 0.01) and to normal subjects ( P < 0.01). The amplitude of EMG responses was also reduced in FDI on the affected side when compared to normals (P < 0.02). Responses obtained in common migraine patients were normal on both sides. We suggest that some permanent subclinical dysfunction of the motor cortex might play a role in the pathogenesis of attacks of classic migraine with sensorimotor auras. Key words: Cortical stimulation; Migraine; Cortical excitability; Motor evoked potentials
Whether migraine is originally a vascular or a neuronal disorder is debated. Cerebral blood flow studies during induced attacks of classic migraine have shown the existence of spreading cortical oligaemia, usually starting in the occipital cortex and extending anteriorly at a speed of 2-3 m m / m i n (Olesen et al. 1981; Lauritzen et al. 1983). The reduced regional cerebral blood flow (rCBF) observed in these studies developed independently of the vascular territories of the cerebral cortex, suggesting a neuronal rather than vascular cause for this phenomenon. Such events could be similar to the "spreading cortical depression" (SCD) which can be experimentally induced in animals (Leap 1944). Spreading oligaemia was absent ictally in common migraine (Olesen et al. 1981). It was then proposed that permanent dysfunction of some cortical neurones might be responsible for the migraine attacks. Using 31p_NMR spectroscopy, Ramadan et al. (1989) have recently measured low brain magnesium levels in migraineurs, which could facilitate the induction of SCD. Indeed, it has been shown that glutamate induced SCD is facilitated by low Mg 2÷ levels (Van Harreveld 1984), which increase the sensitivity of NMDA receptors (Ault et al. 1980; Nowak et al. 1984). In recent years, methods of percutaneous electrical (Merton and Morton, 1980) and magnetic (Barker et al. 1985) stimulation of the motor cortex have been
Correspondence to: Dr. Alain Maertens de Noordhout, University Department of Neurology, H6pital de la Citadelle, Bd du 12e de Ligne, 4000 Liege (Belgium). Tel.: 3241256401.
developed. These allow non-invasive study of the function of the motor cortex and fast-conducting corticospinal pathways in fully conscious man. The magnetic device delivers painless stimuli, which facilitates its use for clinical studies. We used this new technique in migraine in order to disclose possible modifications of motor cortex function between attacks.
Patients and methods
Twelve patients with pure unilateral classic migraine or migraine with aura (10 women, 2 men; mean age: 36.3 _+ 13.7 years; mean disease duration: 13.7 ___8.6 years) and 10 patients (7 women, 3 men; mean age: 38.8 _+ 10.6 years; mean disease duration: 16.3 +_8.5 years) suffering from unilateral common migraine or migraine without aura were studied between attacks and compared with 20 age-matched controls (16 women, 4 men; mean age: 39.6 +_ 13.8 years). Subjects gave fully informed consent and approval of the Local Ethical Committee was obtained. All classic migraine patients experienced sensorimotor auras with visual and speech disturbances in some of them. Neurological symptoms always occurred on the same side of the body, but the subsequent headache was not always strictly contralateral to the side of the aura. Details of the history of the individual patients are given in Table I. Patients with common migraine had attacks of hemicrania always located on the same side of the scalp. Characteristics of patients with common migraine are summarized in Table II. Patients included in either
111
CORTICAL STIMULATION IN MIGRAINE TABLE I Clinical history of patients with classic migraine. Patient
Age (years)
Sex
Disease duration
Attacks frequency
D.G.
37
F
18
4/year
L.S.
15
M
2
3/year
A.M.
39
F
21
5/year
N.G.
45
F
6
2/year
P.E.
69
F
30
10/year
Z.M.
33
F
8
2/year
L.N.
39
F
21
8/year
M.M.
30
F
15
3/year
M.P.
18
F
2
5/year
M.I.
31
F
16
M.C.
39
M
8
5/year
L.J.
40
F
17
6/year
2/month
group fulfill6d the new criteria of the Headache Classification Committee of the IHS (1988) for migraine with aura (classic migraine) or migraine without aura (common migraine). None of the patients had abnormal CT, permanent E E G abnormalities or any other pathological condition. The experimental procedure was conducted at least 1 week after the last migraine attack and in all but 1 patient (L.M., common migraine) there has been a headache-free interval of at least 1 week after the test. None received prophylactic antimigraine therapy at the time of the test, or any drugs that might
Side and type of aura
Side of pain
R. paraesthesia dysphasia L. paresis L. paraesthesia L. scotoma R. paraesthesia R. scotoma L. paresis L. hypoaesthesia L. paraesthesia L. scotoma R. paresis R. scotoma dysphasia L. paraesthesia L. scotoma L. paresis L. paraesthesia L. paraesthesia L. scotoma R. paraesthesia R. scotoma dysphasia L. paraesthesia L. scotoma L. paresis L. paraesthesia L. scotoma
L. and R. R.
L. and R. R. R. L. and R.
R. L. and R. R. L.
L. and R. L. and R.
modify cortical excitability (hypnotics, barbiturates, anticonvulsants). The only treatments allowed were ergotamine or NSAIDs against migraine attacks and oral contraceptives. Magnetic stimulation of the motor cortex was achieved with a Novametrix Magstim 200 with a 9 cm mean diameter stimulating coil held over the vertex, its edge overlying the arm area of the motor cortex. The maximal transient magnetic field produced in this coil is 1.5 Tesla. To stimulate the right hemisphere, the direction of the current in the coil seen from above was
TABLE II Clinical history of patients with unilateral common migraine. Patient
Age (years)
Sex
Disease duration
Attacks frequency
Side of pain
B.L. S.B. D.C. M.A. V.A. F.M. R.J. S.M. L.M. T.T.
64 40 41 31 43 30 41 32 40 26
F F F F M F F M M F
35 15 16 15 20 5 13 8 24 12
3/month 2/month 3/month 4/month 2/month 3/month 1/month 2/month 1/month 3/month
R R L R L R L L L R
112
A. M A E R T E N S D E N O O R D H O U T E T AL.
clockwise and vice versa. In fact, since the last version of the Novametrix stimulator became available in 1990, it appears that the direction of the current in the coil was the reverse of what was stated before (Hess et al. 1987). Responses evoked in the first dorsal interosseous (FDI) muscle of the hand were recorded with surface Ag-AgCI electrodes placed on the belly and tendon of the muscle, some 3 cm apart from each other. Signals were filtered (3 Hz-3000 Hz) and amplified with 2 different gains (100 /zV/V and 2 m V / V , Digitimer D150) to allow accurate detection of the activation threshold and maximal amplitude of the EMG responses. Traces were then stored on floppy disks (Digitimer D200) for off-line analysis. The threshold of activation of FD! by cortical stimuli was measured on both sides, at rest and during isometric background contraction of the target muscle (20% max). Steady voluntary contraction of the muscle results in a marked reduction of its activation threshold (Hess et al. 1987), mainly due to increased excitability of spinal motoneurones (Day et al. 1987). The threshold of activation was defined as the intensity of stimulation able to produce an EMG response in FDI to at least 50% of a series of 16 stimuli. Patients were comfortably seated in an armchair, eyes open. Shortest latency and mean amplitude of the first negative peak of responses to a series of 8 maximal cortical stimuli (CTXmax) were measured on both sides at rest and during background voluntary contraction. Latencies and amplitudes were measured manually, with cursors. Amplitude of the responses was eX-
pressed as percentage of the EMG potentials evoked in FDI by supramaximal stimulation of the ulnar nerve (Mmax), which simultaneously activates all motoneurones supplying the muscle. Calculation of the CTXmax/Mmax ratio ensures that possible amplitude differences between sides are not due to asymmetrical positioning of the recording electrodes or unilateral lesion of the peripheral nervous system. Conduction time in the central motor pathways (CMCT) was calculated by the method proposed by Rossini et al. (1985).
Results
In normal subjects, the mean intensity of the magnetic field (expressed as percentage of the maximal output of the stimulator) able to activate the FDI without voluntary contraction was 48 + 6%. The mean side-to-side difference of activation threshold was 6.8 + 3.5% of the mean value of both sides. This difference was not correlated with handedness. During voluntary isometric contraction (20% max), FDI activation threshold was 33 + 3%. The mean interside difference was 7.9 + 5.9% of the mean. Central motor conduction time (CMCT) from cortex to C7-Thl spinal segment was 6.5 + 0.6 msec on both sides. During voluntary contraction, the C T X m a x / M m a x amplitude ratio was 50.6 + 16.7%. In classic migraine patients, FDI activation thresholds at rest and during voluntary contraction were respectively 55 + 9% and 41 + 8% of the maximal out-
T A B L E III Characteristics of responses recorded in classic migraine patients. Threshold (% max)
CTXmax/Mmax (%)
at rest
contraction
aft. side
nl. side
29 30 41 29 30 29 32 34 28 39 39 37
41 37 27 41 24 25 36 35 36 64 54 19
44 28 38 60 31 37 36 37 20 69 53 44
33 * * 5
37 + 13
41N.S. 14
aft. side
nl. side
aft. side
nl. side
Patient D.G. L.S. A.M. N.G. P.E. Z.M. L.N. M.M. M.P. M.I. M.C. L.J.
51 59 59 42 60 53 63 65 42 50 53 68
51 50 50 33 44 45 54 46 42 51 52 49
28 38 45 40 49 41 34 51 30 40 39 51
Mean S.D.
55 *'+ 9
47 6
41 **'+ 8
Normals Mean S.D.
48 ÷ 6
Wilcoxon: * P < 0.02; * * P < 0.01; N.S. not significant. Unpaired t: + P < 0.05.
33 + 3
51 17
÷
CORTICAL STIMULATION IN MIGRAINE
113 P.E.
put of the stimulator on the side of the body affected by the aura. On the normal side, these values were 47 + 6% and 33 + 5%. The mean interside difference of activation threshold was 15.9 + 12.8% of the mean at rest and 20 + 16.5% during background contraction of the target muscle. The differences of activation thresholds at rest and during contraction between both sides were statistically significant ( P < 0.01, paired t). The mean values recorded on the side affected by the aura were also significantly different from those obtained in normal subjects at rest and during contraction ( P < 0.01, unpaired t). However, 4 patients (33%) showed no side-to-side difference of FDI activation threshold, although their histories and symptoms were not different from those of the other cases. If individual values are considered, the FDI activation threshold at rest was above 3 S.D. from normals in only 1 case (8%), and in 4 cases (33%) during background contraction. Mean CMCT was 6.6 + 0.7 msec on both sides. This value does not differ from that of normals. CTXmax/Mma x ratio was 36.6 12.8% on the affected side and 41.4 + i3.8% on the other (Table III). The former was significantly different from the ratio calculated in the control group ( P < 0.02, unpaired t). A typical example of responses to magnetic cortical stimulation in classic migraine is shown in Fig. 1. No correlation was found between the frequency of attacks or duration of the disease and increased FDI activation threshold on the side of the auras (Spearman rank correlation test). In common migraine, FDI activation thresholds at rest and during backgound contraction were 45 + 6%
.C.t..AS.S.~.~ . _aS_ ).~
f.d.i.
!_
_
|
I __
[~ II
/t
sideof
~ --
/
-
~
A ~ ~ \
aura CORTICAL STIMULATION n~i~eal
~
~
__ .--
2rn---;
s~eof aura ULNARNERVE
-
L.~ ~'/ S
normalSTIMULATION side
Fig. 1. Maximal EMG responses produced interictally in FDI of a classic migraine patient by magnetic cortical (upper traces) and ulnar nerve (bottom traces) stimuli. Responses to cortical stimulation are smaller on the side of the body affected by the aura (4 individual responses superimposed in each trace), while those to ulnar nerve stimuli are identical on both sides. Cortical stimulation was achieved during steady (20% max) background contraction of FDI.
and 33 + 5% of the maximal output of the stimulator on the side of the body contralateral to the side of hemicrania. On the other side, these values were 44 + 7% and 33 + 6% respectively. Interside threshold difference was 3.3 + 3.2% of the mean at rest and 6.1 + 3.5% during contraction. CMCT was 6.5 + 0.7 msec on both sides and C T X m a x / M m a x ratio was 49.4 _+ 9.7% on the"affected side" and 50.2 _+ 9.3% on the other (Table IV). No side-to-side statistically significant difference could be disclosed nor differences from values of the control group. Fig. 2 summarizes the mean
TABLE IV Characteristics of responses recorded in common migraine patients. Threshold (Tesla)
CTXmax/Mma~ (%)
at rest
contraction
aft. side
nl. side
32 21 29 43 40 33 33 35 33 29
50 55 38 42 63 46 48 51 36 65
44 56 46 44 56 48 49 48 39 72
33 6
49 10
50 9
aft. side
hi. side
aft. side
nl. side
Patient B.L. S.B. D.C. M.A. V.A. F.M. R.J. S.M. L.M. T.T.
42 29 52 50 47 42 43 48 45 43
41 27 51 50 52 45 43 45 44 43
33 23 31 42 35 32 32 33 31 31
Mean S.D.
45 6
44 7
33 5
Normal$ Mean S.D.
N.S.
N.S.
N.S.
48 6 N.S.
33 3 N.S.
51 17 N.S.
114
A. MAERTENS DE NOORDHOUT ET AL.
FDI
ACTIVATION % maximal
THRESHOLD output
lO0 rest
80 contraction
60
i L
40
20
I
;ltlJic xf. side
C|IIIiC nL side
normals
common nf.side
common nl.stdc
Fig. 2. Activation thresholds of the FDI muscle at rest (white columns) and during isometric contraction (dashed columns) in the different groups studied. Thresholds at rest and during contraction are significantly increased in classic migraine on the side of the auras when compared to the other side or to normals. No abnormality is observed in common migraine.
values of FDI activation threshold measured in the different groups tested. Finally, the experimental procedure was well tolerated by all subjects and no adverse reactions were reported. Additionally, cortical shocks proved ineffective to induce migraine attacks in this population.
Discussion
The present study was conducted on particular subpopulations of migraine patients. Although they fulfilled IHS criteria for classic (with aura) or common (without aura) migraine, they are in no way typical samples of either affection. The possibility of using the other side as control led us to choose such cases with unilateral symptoms. In the classic migraine group, a significantly stronger magnetic field was needed to activate the first dorsal interosseous muscle of the hand on the side affected by the aura than on the contralateral one. The mean FDI activation threshold on the affected side was also higher than in an age-matched control group. This held true with the target muscle at rest or during voluntary contraction. However, individual values were within the normal range in most patients. For this reason, cortical stimulation should not be considered as a paraclifiical diagnostic tool in this affection. However, the present findings might have pathophysiological implications, which will be detailed further. No significant interside threshold difference was noted in common migraine, nor discrepancy with normal values.
The threshold of activation of a given muscle by motor cortex stimuli depends on several factors. It has been demonstrated that slightly supraliminal magnetic stimuli applied at the vertex excite the fast-conducting pyramidal neurones of the primary motor cortex indirectly, via some cortical excitatory interneurones (Day et al. 1987, 1989). The axons of pyramidal neurones seem to be monosynaptically connected with spinal motoneurones innervating intrinsic hand muscle fibres. The pathway tested by magnetic cortical stimulation is thus at least disynaptic. Increased activation threshold and reduced amplitude of the responses of a given muscle can reflect hypoexcitability of spinal motoneurones, pyramidal cells or cortical interneurones. Additional factors such as unilateral skull thickening or brain atrophy might theoretically play a role, hut they can be ruled out in migraine. It seems unlikely that the excitability of spinal motoneurones is depressed unilaterally in classic migraine, though this possibility has not been studied. However, if this were the cause of interside threshold differences observed in our patients, this difference should be accentuated during voluntary contraction, which normally brings spinal motoneurones to a higher state of excitability. Thus, a higher activation threshold observed in classic migraine rather reflects reduced excitability of large diameter pyramidal cells or cortical interneurones, or hyperactivity in cortical inhibitory structures. There is no definite argument to suggest that cortical hypoexcitability results from brain ischaemia induced by previous attacks. Indeed, reduction of rCBF sufficient to produce structural brain damage has been occasionally measured in episodes of classic migraine (Skyhoj Olsen et al. 1987). However, in this limited series, no correlation was found between electrophysiological abnormalities and duration of migraine history or frequency of attacks. However, episodes of brain ischaemia may have occurred in some patients who were not necessarily those experiencing the most frequent attacks. Our observations contrast with evoked potential studies in migraine, which suggested cortical hyperrather than hypoexcitabiliy (Kennard et al. 1978; Kasprzyk et al. 1984; Milone et al. 1984). At present, we have no straightforward explanation for this discrepancy, except that cerebral structures tested were not the same. In spite of the significant differences found in this study between classic and common migraine, we do not support the idea that these are separate entities with distinct pathophysiological mechanisms. The first argument against this view is certainly the fact ~that all prophylactic and acute treatments are equally active in both forms of migraine. Although rCBF studies seem to yield different results in migraine with and without aura (Olesen et al. 1981), there is thus far no specific
CORTICAL STIMULATION IN MIGRAINE b i o c h e m i c a l or histological m a r k e r specific for either form. F o r these reasons, it is m o r e likely that m i g r a i n e r e p r e s e n t s a c o n t i n u u m with m i g r a i n e w i t h o u t a u r a at o n e e n d a n d m i g r a i n o u s i n f a r c t i o n at the other. T h e y w o u l d be different expressions of a u n i q u e vascular a n d / o r n e u r o n a l dysfunction, driven by d i f f e r e n t individual levels of sympathetic activity ( W e l c h et al. 1988). O u r study also shows that, in a d d i t i o n to C M C T a n d a m p l i t u d e of E M G responses, t h r e s h o l d m e a s u r e m e n t s can detect otherwise u n s u s p e c t e d d y s f u n c t i o n of the m o t o r system.
References Ault, B., Evans, R.H., Francis, A.A., Oakes, D.J. and Watkins, J.C. Selective depression of excitatory amino acid-induced depolarization by magnesium ions in isolated spinal cord preparation. J. Physiol. (Lond.), 1980, 307: 413-428. Barker, A.T., Jalinous, R. and Freeston, I.L. Non invasive magnetic stimulation of the human motor cortex. Lancet, 1985, i: 11061107. Day, B.L., Rothwell, J.C., Thompson, P.D., Dick, J.P.R., Cowan, J.M.A., Berardelli, A. and Marsden, C.D. Motor cortex stimulation in intact man. II. Multiple descending volleys. Brain, 1987, 110: 1191-1210. Day, B.L., Dressier, D., Maertens de Noordhout, A., Marsden, C.D., Nakashima, K., Rothwell, J.C. and Thompson, P.D. Electric and magnetic stimulation of human motor cortex:surface EMG and single motor unit studies. J. Physiol. (Lond.), 1989, 412: 449-473. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia, 1988, 8 (Suppl. 7). Hess, C.W., Mills, K.R. and Murray, N.M.F. Responses in small hand muscles from magnetic stimulation of the human brain. J. Physiol. (Lond.), 1987, 388: 397-420.
115 Kasprzyk, M., Brzecki, A., Podemski, M., Paradowski, B. and Bogdanska, R. Diagnostic value of pattern-reversal visual evoked potentials in vasomotoric headaches. Cephalalgia, 1984, 5 (Suppl. 1): 466. Kennard, C., Gawel, M., Rudolph, N.M. and Clifford Rose, F. Visual evoked potentials in migraine subjects. In: A.P. Friedman, M. Granger and M. Critchley (Eds.), Research and Clinical Studies in Migraine, Vol 6. Karger, Basel, 1978: 73-80. Lauritzen, M., Skyhoj Olsen, T., Lassen, N.A. and Paulson, O.B. The changes of regional cerebral blood flow during the course of classical migraine attacks. Ann. Neurol., 1983, 13: 633-641. Le~o, A.A.P. Spreading depression of activity in cerebral cortex. J. Neurophysiol., 1944, 7: 359-390. Merton, P.A. and Morton, H.B. Stimulations of the cerebral cortex in the intact human subject. Nature, 1980, 285: 227. Milone, F., D'Andrea, G., Cananzi, A. and Zanini, R. Somatosensory evoked potentials patterns in migraine sufferers. Cephalalgia, 1984, 5 (Suppl 1): 470-471. Nowak, L., Bregestowski, P., Ascher, P., Herbert, A. and Prochiantz, A. Magnesium gates glutamate-activated channels in mouse central neurones. Nature, 1984, 307: 462-465. Olesen, J., Larsen, B. and Lauritzen, M. Focal hyperemia followed by spreading oligemia and impaired activation of rCBF in classic migraine. Ann. Neurol., 1981, 9: 344-352. Ramadan, N.M., Halvorson, H., Vande-Linde, A., Levine, S.R., Helpern, J.A. and Welch, K.M.A. Low brain magnesium in migraine. Headache, 1989, 29: 590-593. Rossini, P., Di Stefano, E. and Stanzione, P. Nerve impulse propagation along central and peripheral fast conducting motor and sensory pathways in man. Electroenceph. clin. Neurophysiol., 1985, 60: 320-334. Skyhoj Olsen, T., Friberg, L. and Lassen, N.A. Ischemia may be the primary cause of the neurological deficits in classic migraine. Ann. Neurol., 1987, 45: 154-159. Van Harreveld, A. The nature of the chick's magnesium-sensitive retinal spreading depression. J. Neurobiol., 1984, 15: 333-344. Welch, K.M.A., NageI-Leiby, S. and D'Andrea, G. The biological and behavioural basis of migraine. In: J. Olesen and L. Edvinsson (Eds.), Basic Mechanisms of Headache. Elsevier, Amsterdam, 1988: 447-456.
