AAEM MINIMONOGRAPH 35 ABSTRACT: Since 1985, when the technique of transcranial magnetic stimulation (TMS) was first developed, a wide range of applications in healthy and diseased subjects has been described. Comprehension of the physiological basis of motor control and cortical function has been improved. Modifications of the basic technique of measuring central motor conduction time (CMCT) have included measurement of the cortical silent period, paired stimulation in a conditioning test paradigm, repetitive transcranial magnetic stimulation (rTMS), and peristimulus time histograms (PSTH). These meth- ods allow dissection of central motor excitatory versus inhibitory interplay on the cortical motor neuron and its presynaptic connections at the spinal cord, and have proven to be powerful investigational techniques. TMS can be used to assess upper and lower motor neuron dysfunction, monitor the effects of many pharmacological agents, predict stroke outcome, document the plasticity of the motor system, and assess its maturation and the effects of aging, as well as perform intraoperative monitoring. The recent use of rTMS in the treatment of depression and movement disorders is novel, and opens the way for other potential therapeutic applications. © 2002 American Association of Electrodiagnostic Medicine. Muscle Nerve 25: 160–175, 2002 DOI 10.1002/mus.10038

MAGNETIC STIMULATION OF THE CENTRAL AND PERIPHERAL NERVOUS SYSTEMS

MARKUS WEBER, MD,1 and ANDREW A. EISEN, MD, FRCP(C)2

1 Department of Neurology, Kantonsspital, CH-9007 St. Gallen, Switzerland 2 Neuromuscular Diseases Unit, Vancouver General Hospital, Vancouver, British Columbia, Canada

Early experiments in humans used high-voltage, discusses methodological aspects, reviews the differ- short-duration electrical stimulation applied to the ent techniques and measurements in use, and criti- scalp overlying the motor cortex, a rather uncom- cally analyzes its utility in clinical practice and basic fortable procedure and inappropriate for routine neuroscience. clinical use. In 1985, Barker and colleagues5 intro- duced the technique of transcranial magnetic stimu- lation (TMS) which led to a new era of research in ANATOMY AND PHYSIOLOGY OF THE CORTICAL MOTOR NEURONAL SYSTEM motor control and cortical function. Since that time, interest in TMS has steadily increased and a vast lit- Motor function in humans is subserved by several erature has already accumulated. distinct yet interconnected anatomical regions. They This minimonograph considers current concepts include the primary motor cortex, also known as of the anatomical and physiological basis of TMS, Brodmann area 4, the premotor areas and supple- mentary motor cortex, basal ganglia, thalamus, cer- Abbreviations: AHC, anterior horn cell; ALS, amyotrophic lateral sclero- ebellum, brain stem, and reticular formation. The sis; CM, cortical motor neuronal; CMAP, compound muscle action poten- primary motor cortex is different from other regions tial; CMCT, central motor conduction time; EMG, electromyographic; GABA, gamma-aminobutyric acid; ISI, interstimulus interval; MEP, motor of the cerebral cortex in that it is thicker but has a evoked potential; MS, multiple sclerosis; PLS, primary lateral sclerosis; lower cell density. The main output cells are the PSTH, peristimulus time histograms; rTMS, repetitive transcranial mag- netic stimulation; SMNs, spinal motoneurons; T, tesla; TMS, transcranial large pyramidal cells in lamina V and smaller cells in magnetic stimulation lamina III. Their dendrites show a preferential ori- Key words: cortical function; cortical motor neuronal system; motor con- trol; motor evoked potential; peristimulus time histogram; transcranial entation parallel to the main axis of the precentral magnetic stimulation gyrus. Correspondence to: AAEM, 421 First Avenue SW, Suite 300 East, Roch- ester, MN 55902; e-mail: [email protected] The spinal motoneurons (SMNs) of the cord are © 2002 American Association of Electrodiagnostic Medicine. Published the “final common pathway” of the motor system to by John Wiley & Sons, Inc. which the higher centers and pyramidal cells make

160 Magnetic Stimulation MUSCLE & NERVE February 2002 direct or, more commonly, indirect connections via (Fig. 1, left).70 This results in a single descending multiple descending tracts. In non-human primates volley recordable from the pyramidal tract, which and other mammals, these descending tracts con- has been termed the D wave or direct wave.42,43,62,105 verge on the spinal motoneuron. In humans, the Increasing the stimulus intensity activates input cells, sophistication and complexity of motor control, par- causing indirect, transsynaptic activation of pyrami- ticularly in the face and distal aspects of the limbs, dal tract neurons. A series of recordable volleys, has largely sacrificed many of these indirect tracts named I waves to indicate their indirect origin, fol- with the expansion of the cortical motor neuronal low the initial D wave. The I waves are separated by (CM) system. This CM system originates from large intervals of about 1.5 to 2 ms. Anesthesia and cooling pyramidal cells in the primary motor cortex and is of the motor cortex have a profound depressant ef- the only descending motor pathway that makes fect on the I waves but not on the D wave.63 Epidural monosynaptic connections with the SMNs. (See Por- recordings of multiple descending volleys from the ter and Lemon,107 for a comprehensive review of spinal cord of conscious human patients have pro- corticospinal function in humans.) Each cortical mo- vided evidence that transcranial electrical stimula- tor neuron synapses with many SMNs, and each SMN tion activates the motor cortex in humans and ani- receives input from many different CM cells. This mals in the same way.19,49 arrangement of convergence and divergence is most The same experiments have also confirmed that abundant for the distal muscles—especially those of threshold transcranial magnetic stimuli over the the hand and facial musculature. It is what affords hand area of the motor cortex preferentially activate humans their amazing degree of fractionated con- the pyramidal cells indirectly (transsynaptically) trol and allows for a large repertoire of different through excitatory interneurons (Fig. 1, right). The movements served by the same muscle. CM control is onset latency of the compound muscle action poten- largely responsible for delicate control of force, pre- tial (CMAP) from small hand muscles is approxi- cision grip, angulation, rate of change of movement, mately 2 ms later (Fig. 1, bottom right, solid line) and muscle tension. It is likely that the CM system is than the electrically induced response. However, vital to the acquisition of new motor skills, which, with higher stimulus intensities or certain lateral coil once learned, are probably transferred to more cau- positions, the latency may shorten, consistent with D dal parts of the nervous system, including the spinal wave activation (Fig. 1, bottom right, dotted line). cord. Glutamate is the primary excitatory neuro- The different activation of pyramidal cells probably transmitter of the CM system. is related to the orientation of the induced current. The CM system is subject to excitatory and inhibi- Electrical stimulation causes the current to flow in all tory modulation. The stellate or basket cells are lo- directions parallel and radial to the surface, thus cated primarily in laminae III and V. Their axon penetrating the radially oriented pyramidal cells. terminals form predominantly inhibitory, gamma- TMS, however, induces current flow parallel to the aminobutyric acid (GABA) synapses on dendritic surface of the brain, preferentially exciting horizon- shafts, somata, and/or proximal axonal segment of tally oriented neurons. The result is that radially ori- the pyramidal neuron (cortical motor neuron)39 ented neurons will have a higher threshold for mag- and are horizontally orientated. These interneurons netic stimulation than electric stimulation. This is modulate the response of pyramidal neurons to ex- why coil orientation is important; even a slight posi- citatory inputs. tional change of the magnetic coil on the scalp can profoundly affect the size and latency of the motor NATURE OF TRANSCRANIAL evoked potential (MEP).14,85,96 The response of MAGNETIC STIMULATION lower limb muscles has a similar latency with electri- Since the introduction of TMS, there has been a cal and magnetic stimulation. This suggests that both debate over which structures are activated by the techniques have the same activation site in the ros- magnetic stimulus. A rapidly changing magnetic tral pyramidal axons as they leave the cortex and field is generated that induces electrical currents readily produce D wave activity.115 Whether TMS ac- within the cortex.5,43 Short-latency contractions are tivates a bi- or polysynaptic pathway in healthy sub- evoked in contralateral limb muscles. The latency is jects is presently unclear. Studies on monkeys have in keeping with a monosynaptic connection.4,43,61,86 failed to identify disynaptic excitation of motoneu- A single low-intensity anodal electrical stimulus rons from the pyramidal tract.83 delivered to the exposed surface of the cortex in TMS also activates the local circuit inhibitory in- monkeys preferentially activates pyramidal tract, terneurons. Several ipsi- and contralateral inhibitory neurons directly in the region of the axon hillock phenomena have been revealed with double (condi-

Magnetic Stimulation MUSCLE & NERVE February 2002 161 FIGURE 1. Preferential excitation of pyramidal cells at the axon hillock by electrical stimulation (left side) versus transsynaptical activation by magnetic stimulation (right side). Higher stimulus intensities result in a shorter latency of the magnetically evoked motor evoked potential (MEP) (dotted lines). Preferential D or I wave activation (spinal cord volleys) is indicated by solid lines (see also text). tioning) stimulus paradigms and rTMS, which will be neurons will be recruited, thus shortening the onset discussed below. latency. The increase of the CMAP amplitude indi- cates recruitment of a greater number of spinal mo- Facilitation. When TMS is performed with the tar- toneurons. This could also be due to increased spi- get muscle steadily contracting, it shows different re- nal excitability, increased synchronization of spinal sults than when the muscle is relaxed. Muscle con- motoneuron firing, or an increasing number of I traction has three main effects72: the threshold for waves bringing more AHCs to threshold. evoking the motor response is reduced, the latency of the MEP is shortened, and the amplitude of the MAGNETIC STIMULATOR AND COILS MEP is markedly increased.46,61,122 These facilitatory The components of a magnetic stimulator consist of effects can also be induced simply by the subject’s a capacitor and an inductor (the stimulating coil). thinking about the maneuver or contraction of an- The energy for stimulation is derived from charging other muscle (either on the same or opposite side), a bank of capacitors up to about 4 kV, which when but the extent of facilitation is less than that induced discharged induces a current of up to 5,000 A that by contraction of the target muscle.72,121 The under- passes through the copper stimulating coil, creating lying mechanisms for facilitation are not entirely un- a brief but intense magnetic field. Tissues, skull, and derstood but likely include increased cortical and scalp present little or no impedance to a magnetic spinal excitability.122,137 With voluntary contraction, field of rapidly changing intensity. The direction of the resting potential of the anterior horn cell (AHC) current flow in the coil is opposite to the direction of is closer to threshold, requiring less temporal sum- the induced currents in the nervous tissue. However, mation of descending volleys, which means that the the magnetic field declines rapidly with distance. discharge can occur at an earlier I or D wave, thus The intensity of the magnetic field is represented shortening the onset latency. Furthermore, with in- by flux lines around the coil and is measured in tesla creasing force, according to the Henneman size (T). The stimulating current, which is maximal in an principle, larger and faster conducting spinal moto- annulus underneath the coil, may be either biphasic

162 Magnetic Stimulation MUSCLE & NERVE February 2002 or monophasic. Because the direction and phases of of a target muscle. Measurements include the corti- current flow determine which neuronal elements are cal threshold, latency and central conduction time, activated within the cortex, a biphasic impulse may amplitude, and MEP/CMAP ratio. stimulate different populations of cells than a mono- phasic impulse. The responses to monophasic CORTICAL THRESHOLD stimuli tend to be unilateral, whereas responses to In a relaxed target muscle, the cortical threshold multiphasic stimuli may be bilateral. If the initial reflects the global excitability of the motor pathway, current flow in a circular coil positioned over the including large pyramidal cells, cortical excitatory vertex is clockwise, the left hemisphere will be acti- and inhibitory interneurons, and spinal motoneu- vated. Reversing the direction of the initial current rons. Even slight voluntary contraction of the target will activate the right hemisphere. Large round coils muscle reduces the cortical threshold. Threshold to produce fields that penetrate the deepest, and the magnetic stimulation is usually defined as the stimu- magnetic fields are distributed through a larger vol- lus required to elicit reproducible responses of 50 to ume of tissue, resulting in nonfocal stimulation. Cen- 100 µV in about 50% of 10 to 20 consecutive trials.65 tered over the vertex, the circumference of the coil When motor potentials are recorded from a mod- overlies the hand area of the motor cortex. Smaller estly activated target muscle, the response should be coils, especially butterfly or figure-eight shaped, around 200 to 300 µV so that it can be distinguished elicit more focal stimulation with activation occur- reliably from background activity. The position of a ring beneath the intersection site, but produce a circular coil centered over the vertex is less critical relatively weak and less penetrating magnetic field. than positioning a figure-eight coil. The optimal coil For small hand muscles, the optimal stimulation site position and orientation of the figure-eight coil may is some 5 cm lateral to the vertex on the interaural even be different for each intrinsic hand muscle.14 line with the figure-eight coil orientated 45° to the Mills and Nithi88 have recently developed a more parasagittal plane. reliable measure of threshold. A single stimulus at 20% of maximum stimulator output is given and METHODS AND MEASUREMENTS single trials at 10% increments are then performed For routine studies, the magnetic stimulator is con- until a response is obtained. The intensity is then nected with standard needle electromyographic decreased 1% at a time until 10 stimuli fail to give (EMG) equipment (Fig. 2). A synchronization pulse any response. This is referred to as the lower thresh- occurring at the moment of the stimulator’s dis- old. The stimulus intensity is then increased by 1% charge serves as an external trigger that starts a increments until all of 10 stimuli induce a response sweep that will display the recorded motor response of greater than 20 µV in amplitude with a latency of

FIGURE 2. Principle of TMS and calculation of central motor conduction time (CMCT). MEP1 is recorded after transcranial magnetic stimulation (S1), MEP2 after cervical stimulation (S2). CMCT is estimated by onset latency of T1 minus onset latency of T2.

Magnetic Stimulation MUSCLE & NERVE February 2002 163 17 to 30 ms. This is referred to as the upper thresh- the AHC to the intervertebral foramen and the old. Using this approach, the lower threshold mea- CMCT will be estimated as slightly too long. This is sured 38 ± 8.6% and the upper threshold 46.6 ± not the case when using the F-wave method. The 9.4%. Threshold in adults is independent of age, conduction time from spinal motor neuron to gender, and hemisphere, but varies with different muscle is given by the formula (F + M−1)/2 where F target muscles.88,131 It is lowest for hand muscles and is the shortest F-wave latency, M the onset of the highest for proximal arm muscles, leg muscles, and direct muscle response, and 1 ms is allowed for the axial muscles. This is in keeping with the more ex- turnaround time at the AHC. Latency varies with tensive cortical representation of hand versus more height and arm length, therefore, central motor con- proximal muscles. duction is slightly faster in women than men.36,54,88 Latency and central conduction increase in a linear LATENCY AND CENTRAL CONDUCTION TIME fashion with increasing age, but the correlation is Latency and central conduction time depend on weak.50,88 whether the MEP was recorded at rest or with acti- vation, which shortens the latency by several milli- AMPLITUDE AND MEP/CMAP RATIO seconds. However, the latency does not change The absolute amplitude of the MEP depends on much once 20% or more of maximum voluntary complex interactions between the CM and the AHC contraction is used. Thus, if latency is the only con- at the moment of stimulation. It reflects the sum of sideration, force does not need to be accurately con- upper and lower motor neuron activity. There can trolled and the subject can be asked to moderately be considerable inter-trial as well as intra-individual contract the muscle. The onset of the MEP is usually variation especially when stimulating with threshold readily identifiable. The shortest of four to five re- or slightly suprathreshold intensities. With increas- sponses should be measured. In some diseases, the ing stimulus intensity, the response becomes more MEP may be markedly reduced in amplitude and, stable.71 Many factors account for this variability, when facilitation is used, partially buried in the back- most of which are difficult or impossible to control ground EMG. This often makes the onset latency in the clinical setting. Coil position is critical; mini- difficult to recognize, and superimposing a number mal angulation of the coil even at the same site may of potentials may then be helpful. To calculate the drastically change the amplitude of subsequent re- central motor conduction time (CMCT), conduc- sponses. As discussed above, even modest muscle tion in the peripheral segment of the motor pathway contraction greatly facilitates the response and it is (AHC to muscle) is estimated and then subtracted imperative to state whether the response was elicited from the onset latency of the MEP (Fig. 2). with the target muscle relaxed or under voluntary For cervical root stimulation, the most active part contraction. If facilitation is used for amplitude mea- of the coil is positioned just rostral to the spinous surements, force or overall muscle activity should be process of C7 in the midline or within 2 cm lateral to estimated. This can be accomplished by either using this position. Because a peripheral nerve is being isometric strain gauges or rectifying and integrating stimulated, it is of no consequence which way the the background needle EMG to provide a measure coil faces. The lumbosacral roots can be stimulated as a percentage of the maximum.48 The amplitude is by positioning the coil with the midpoint of its lead- usually measured peak-to-peak. ing inner edge midline over the particular vertebral Because of its variability, the absolute amplitude body of interest. There is no need to obtain a maxi- is of limited clinical value. However, in the authors’ mal response. The primary aim is to elicit several experience, a side-to-side difference of 50% or superimposable responses from which an accurate greater can be regarded as abnormal in patients onset latency can be measured. However, for periph- without lower motor neuron disease. The MEP/ eral electrical stimulation of motor nerves, latency CMAP ratio takes account of the lower motor neu- depends critically on the axons stimulated in a sub- ron contribution and is a more useful indicator of maximal response. As response amplitude increases, disease originating in the cortex. However, the ratio latency almost always shortens. The same may apply is very variable, ranging in normal subjects from 10 for magnetic root stimulation, but this has not yet to 100%. The recently developed triple stimulation been systematically investigated. Nevertheless, stimu- technique provides a more accurate and less variable lating the nerve roots either magnetically or electri- estimate of upper motor neuron activation. It has cally excites the nerve roots in the region of the been applied in a variety of upper motor neuron intervertebral foramen.78,79 The onset latency does disorders,15,81,82,113 but the technique can be un- not, therefore, include the conduction time from comfortable for patients.

164 Magnetic Stimulation MUSCLE & NERVE February 2002 CORTICAL MAPPING and control subjects. In other words, the cortical rep- The motor cortex is organized in terms of move- resentation of the reading finger has become en- ments rather than muscles. Individual muscles have larged at the expense of other fingers. Also, the ac- multiple representations (convergence) and a given quisition of new fine motor skills in normal subjects is associated with reorganization of the motor cortex CM may provide input to several spinal motoneu- 37,99 rons of different muscles (divergence). Because output map. Different areas of the motor map TMS preferentially activates fast-conducting cortico- enlarge depending on the newly acquired skill. Spi- spinal fibers, maps reflect only the output function nal cord injury also results in enlargement of output maps projecting to muscles proximal to the lesion and distribution of the most direct fast-conducting 38 cortical motor neuronal fibers. For mapping the to- level. Reorganization of the motor cortex output pographic structure of cortical motor areas, a but- map has also been shown with altered sensory input associated with immobilization, ischemic nerve terfly (figure-eight) coil is usually used, because the 15,27,56,110,111 more focused field gives a more accurate map. The block, dystonia, stroke, and facial palsy. surface of the cortex is marked out in 1 cm squares Mapping of cortical areas other than the motor using Cz (international 10 to 20 system) as the zero– cortex is also possible. Depending on the exact coil zero mark. Points are extended anteriorly and pos- position and current direction, TMS of the occipital cortex evokes phosphenes in different areas of the teriorly along the sagittal plane and over the left and 5 right hemispheres in the coronal plane. With a stan- visual field. TMS of the sensorimotor cortex occa- dard stimulus magnitude, the coil is systematically sionally triggers somatotopically organized paresthe- moved over the motor cortex, which then produces sias but may also block detection of an electrically a map of different MEP amplitudes at each site.131 evoked sensory stimulus. The greatest MEP amplitude is evoked in the center THE CORTICAL SILENT PERIOD of the map (optimal position) and declines as the As mentioned earlier, TMS also produces inhibitory coil is moved away from it. With the figure-eight coil, phenomena, the most consistent being the presence the optimal position to elicit responses in small hand of a long period of needle EMG silence during a muscles is 5 to 7 cm lateral to the vertex on the sustained voluntary contraction (Fig. 3). This is akin 90,112 interaural line. Other measurements include to the silent period obtained by stimulating a periph- the number of excitable scalp positions and the cen- eral motor nerve during contraction of a muscle.109 44,89,132 ter of gravity. A frameless stereotactic system The duration of the silent period, usually defined as 74 allows more precise coil placement. the time from the beginning of the MEP to the re- turn of voluntary needle EMG activity, is linearly re- PLASTICITY OF THE MOTOR CORTEX lated to stimulus intensity but independent of the level of background contraction. For clinical consis- Motor mapping experiments that use the magnetic tency, measurements should be made with defined coil in conscious humans have now clearly docu- stimulus intensities in relation to individual motor mented plasticity of the motor cortex and its ability to thresholds. Silent periods are longest in small hand reorganize in certain circumstances.24,33,34,76,91,100,114 muscles (200 to 300 ms) and less prominent in proxi- Piano practice for a few days tends to increase the mal arm muscles and leg muscles. Weak stimuli can size of the cortical motor area for relevant muscles. depress EMG activity while eliciting no motor re- In congenital atresia of the forearm and hand akin sponse, indicating that the threshold for this inhibi- to that seen in thalidomide teratogenicity, the proxi- tory effect is less than for the excitatory effect. Spinal mal arm develops a larger than normal representa- inhibitory mechanisms such as Renshaw inhibition tion. are considered to contribute only to the first 50 ms When hemispherectomy is performed early in to 60 ms of the TMS-induced cortical silent period, life, ipsilateral motor representation becomes much whereas most of the suppression is due to different more pronounced, and cortical stimulation induces cortical inhibitory mechanisms.28,53,109 The neuro- bilateral responses.38 In adults, learning-induced nal elements responsible for these effects are topo- representational plasticity has been demonstrated in graphically close to the corticospinal neurons and blind Braille readers.102,104 In long-standing Braille are most likely the local inhibitory interneurons, readers, the representation of the first dorsal inter- which use GABA as their transmitter. osseous muscle of the reading finger is much larger than the homologous muscle on the other side, PAIRED CORTICAL STIMULATION whereas the adductor digiti minimi of the reading Two transcranial magnetic stimuli delivered in a con- hand is smaller than that of the non-reading hand ditioning test paradigm can be used to assess intra-

Magnetic Stimulation MUSCLE & NERVE February 2002 165 old test stimulus, short-latency excitatory effects (ISI 1 to 6 ms) can be demonstrated. These effects show periodicity reminiscent of the I waves recorded di- rectly from the pyramidal tract. The paradigm can be used to assess drug effects and pathological con- ditions.29,77,139

REPETITIVE TRANSCRANIAL MAGNETIC STIMULATION Repetitive transcranial magnetic stimulation (rTMS) is only possible with special stimulators that have technical features allowing the generation of fast rates of stimulation. The technique permits modula- tion of corticospinal excitability.26 The effects, rang- ing from inhibition to facilitation, depend on the stimulation parameters (stimulus intensity, inter- stimulus interval, number of stimuli, and interval be- tween successive trains) and may last beyond the du- ration of the rTMS itself.26,103 Lasting effects of high- frequency rTMS (greater than 1 HZ) on clinical symptoms have been seen in Parkinson’s disease and depressed patients, whereas low-frequency rTMS can transiently improve symptoms in patients with task- specific dystonia.55,101,119 Further clinical applica- tions include treatment of focal epilepsy, cortical my- oclonus, spasticity, and obsessive-compulsive disorders. Other effects outside the motor areas in- clude interference with language, cognitive pro- cesses, and memory.98 The different therapeutic FIGURE 3. Silent period in a normal subject and an ALS patient. benefits of rTMS are not easy to explain but may Recordings were made from a modestly contracting abductor include neuromodulatory effects from released neu- digiti minimi muscle. rotransmitters and changes in cerebral blood flow. High frequency and intensity rTMS may cause cortical inhibitory and excitatory mechanisms.94 The epileptic seizures. Secondarily generalized seizures effects depend on the type of stimulus (electrical or following rTMS have been reported in healthy sub- magnetic), the scalp site at which stimuli are applied, jects and patients with epilepsy and depression, but the intensity of both the conditioning and test there is no evidence for the development of epilepsy stimuli, the muscle activity, and the interstimulus in- after an rTMS-provoked seizure.129 Spread of excita- terval (ISI). With the muscle at rest, the response of tion in the cortex, as evidenced by CMAPs appearing a suprathreshold stimulus is inhibited by a sub- in muscles remote from the target muscle and threshold conditioning stimulus at intervals of 1 to 5 needle EMG activity that persists after the stimulus ms and facilitated from about 10 to 20 ms.75 The ends, is considered an indicator of induced epileptic inhibitory effect is reduced with voluntary contrac- activity.8,103 These observations were used as the ba- tion. The inhibition is due to the effects of local- sis on which the maximum safe combinations of circuit inhibitory interneurons and also the result of stimulus intensity, frequency, and duration of single inhibitory collaterals from excited corticospinal fi- trains of rTMS were defined.130 bers.75 Threshold pairs of stimuli of equal strength result in inhibition of the test response at ISI of 5 to DIRECT STIMULATION OF CORTICOSPINAL 30 ms, and facilitation at ISI of 40 to 90 ms. A dif- TRACT AXONS ferent pattern occurs with higher stimulus intensi- It is usually not possible to stimulate the pyramidal ties: ISI of 25 to 50 ms cause facilitation, and ISI of 60 tract axons directly (postsynaptically) with a magnet. to 200 ms cause inhibition. Using pairs of threshold However, Ugawa et al.128 demonstrated that this is stimuli (0.9 to 1.1 times threshold or a suprathresh- feasible with a double cone type of coil that is ca- old conditioning stimulus) followed by a subthresh- pable of delivering stimuli to deep structures. The

166 Magnetic Stimulation MUSCLE & NERVE February 2002 coil is placed over the inion stimulating ipsilateral to the side of recording. Latency to the hand muscles is about 16.5 ms, compared to about 20 ms after cor- tical and 12.5 ms after cervical root stimulation, which suggests that the corticospinal pathway is stimulated at the level of the pyramidal decussation. Indicators that the response indeed originates in the tract and not pyramidal cells include consistency of latency and shape of the response and a latency that is identical to that evoked by electrical stimulation at the same site. This method is useful to confirm that prolonged latency of a MEP elicited by cortical stimulation is due to slowed conduction within the spinal tracts and not the result of increased temporal or spatial dispersion resulting from impaired intracortical ini- tiation of the descending volley at the level of the pyramidal cell.127

PERISTIMULUS TIME HISTOGRAMS The cortical motor neuronal system (the CM and its target spinal motoneuron) can be investigated using peristimulus time histograms (PSTHs) (Fig. 4). The firing probability of a voluntarily activated motor unit is modulated when it is subjected to a series of transcranial magnetic stimuli.18,42 The PSTH re- corded from forearm and hand muscles typically shows a marked increase in the firing probability occurring at about 20 to 25 ms after the stimulus, FIGURE 4. Peristimulus time histograms (PSTHs). Well- synchronized primary peak of short duration in a control subject which is referred to as the primary peak. The onset (top) and Kennedy’s disease (middle). Dispersed primary peak latency of the primary peak is in keeping with a vol- consisting of a double peak in ALS (bottom). ley descending through the fast-conducting mono- synaptic (corticospinal) pathway. The configuration Peripheral nerves are most readily stimulated by of the primary peak (amplitude, duration, and dis- the magnetic coil at sites where there is an abrupt persion) reflects the rising phase of the composite change in the volume conductor or at sites of nerve excitatory postsynaptic potential at the AHC induced bending.78,79 This may explain the paradoxical ease by the descending cortical volley. This technique has with which proximal rather than distal nerves are been applied to several diseases but has been of par- stimulated. This can be helpful for stimulating the ticular value in amyotrophic lateral sclerosis deeply placed phrenic nerve in the neck. The direc- (ALS).134 tion of current flow is critical in cortex stimulation USE OF THE MAGNETIC COIL FOR PERIPHERAL but not in peripheral nerve stimulation. However, NERVE STIMULATION the CMAP latency may change by a fraction of a Use of the magnetic coil for study of the peripheral millisecond when the current direction is reversed, nervous system is presently limited by its inability to possibly as a result of the “cathodal-anodal reversal” deliver a controlled focal stimulus. For example, it is effect and/or shallower rise-time in the strength of difficult to stimulate the median and ulnar nerve the magnetic field with reversed flow. independently when the coil is placed over the wrist. It is also difficult to elicit maximum amplitude SAFETY CONSIDERATIONS AND SIDE EFFECTS CMAPs in a reproducible manner as is possible with Since Barker’s development of the first commercial conventional electrical stimulation. However, more magnetic stimulator, many thousands of patients advances in coil design show promise in improving and normal individuals throughout the world have the precision and thus the utility of magnetic periph- undergone magnetic stimulation without ill effect. eral nerve stimulation.11 Adverse effects of single pulse magnetic stimulation

Magnetic Stimulation MUSCLE & NERVE February 2002 167 of the motor cortex are extremely rare. Induction of elination. In children, CMCT linearly declines with epileptic seizures and kindling have caused the most age. Adult values for central motor conduction can concern but there have only been a few reports of usually be attained by 4 years of age. However, cor- seizures occurring at or shortly after the time of mag- tical threshold remains high until the end of the first netic stimulation.35,64,66 Formal studies on known decade.92 The disparity between attainment of adult epileptics have failed to induce either clinical sei- values of central motor conduction velocity and cor- zures or electroencephalographic epileptiform activ- tical threshold is consistent with the notion that cen- ity.139 However, it has become clear that rTMS, de- tral myelination is completed before synaptogenesis. pending on the stimulation parameters (see above), On the other end of the age spectrum, MEP ampli- can evoke seizures in normal subjects and patients tude declines and central motor conduction time with neurological disease.130 Other concerns have gradually increases with increasing age. Injury to the included possible brain cell damage with cognitive motor cortex in young children can be followed by and other dysfunction and complications from dis- excellent functional recovery of the affected limb(s). lodging neurosurgically inserted metal clips. It is In such situations, magnetic stimulation of the unaf- possible that magnetically induced currents could fected cortex induces not only the usual contralater- damage the internal electronics of biomedical de- al response but also a large ipsilateral one. This is vices such as cardiac pacemakers. The coils should probably the result of corticospinal sprouting within therefore not be placed in the vicinity of cardiac the pyramidal tract leading to an elaboration of ip- pacemakers. Cardiac muscle can only be stimulated silateral projections.18 It has been morphologically with the magnetic coil if it is placed directly over the confirmed that sprouting of central fibers can occur, open heart; applying a magnetic coil over the lateral using labeling of corticospinal cells. chest wall in the process of stimulating the intercos- tal nerves has not caused cardiac irregularities. Nev- MAGNETIC STIMULATION IN DISEASE ertheless, it is advisable to avoid stimulating directly Many abnormalities revealed by magnetic stimula- over the precordium. tion are not disease specific and, like most other Formal psychometric testing before and after neurophysiological tests, the results must be consid- single pulse magnetic stimulation has indicated no ered in the light of clinical data. Frequently the cor- associated cognitive impairment. Endocrine assess- relation between clinical deficit and degree of MEP ment of the pituitary-hypothalamic axis after TMS abnormalities is rather poor. In general, demyelin- has shown no consistent changes.67,84 ation of central motor pathways is associated with Magnetic stimulation can activate the auricular more marked conduction slowing and prolongation muscles, especially in young children. The noise of central conduction times. On the other hand, in level of older stimulators raised the concern of tem- neuronal disease, the MEP, if recordable, is of re- porary hearing impairment but no lasting effects duced amplitude but usually is only modestly pro- were found with the small number of stimuli applied longed in latency. to most subjects.97 For routine clinical practice, 10 to 15 stimuli are usually more than sufficient to achieve Amyotrophic Lateral Sclerosis. Earlier studies in the desired information. Several hundred subthresh- ALS using electrical stimulation of the cortex old stimuli, as required for PSTHs, are equally safe. showed modest prolongation of central motor con- As a general guide, previous cranial neurosurgery, duction time, frequently marked MEP attenuation, the wearing of an electrically sensitive biomedical and, in some cases, absence of the MEP.68 TMS re- device such as a cardiac pacemaker or intrathecal veals similar abnormalities. The prominent abnor- pump, and a history of seizures are relative contra- mality is an absent or small MEP that is frequently indications. dispersed (Fig. 5). This correlates with the occur- rence of dispersed primary peaks in the PSTH (Fig. MATURATION OF THE CORTICAL MOTOR 4, bottom), which may reflect hyperexcitability of NEURONAL SYSTEM AND AGE-RELATED CHANGES CM connections. In general, the correlation of cen- Magnetic stimulation is ideal for the study of the tral motor conduction prolongation with other MEP maturation of motor pathways. Adult values for cen- abnormalities and with clinical upper motor neuron tral motor conduction velocity are attained a few signs (hyperreflexia, finger flexion, impaired fine years after central sensory conduction. In both in- finger movement) is poor.117 stances, adult values for peripheral conduction are Various neurophysiological methods employing reached earlier than central conduction, implying TMS have also indicated hyperexcitability of the mo- that peripheral myelination precedes central my- tor cortex in ALS.47,58,73,87,93 The threshold required

168 Magnetic Stimulation MUSCLE & NERVE February 2002 Patients with primary lateral sclerosis (PLS) show significantly elevated thresholds to TMS and longer central conduction time to both upper and lower limbs. However, using PSTHs, it can be demon- strated that the onset latency of the primary peak in ALS and PLS does not significantly differ, implying that TMS activates the same population of CM con- nections in ALS and PLS.

Multiple Sclerosis. Demyelination induces conduc- tion block, slowed conduction, and inability to faith- fully sustain rapid trains of impulses. These charac- teristic physiological disturbances in multiple sclerosis (MS), individually or in combination, ac- count for prolongation of CMCT, reduced MEP/ CMAP ratio, increased variability of onset latency of the MEP (latency jitter), and dispersed morphology (Fig. 6).12,16,62 Slowing of central motor conduction, the most commonly seen abnormality, can be very marked and correlates to some degree with the pres- ence of upper motor neuron signs and clinical defi- cit.69 A common site of demyelination in MS is the corpus callosum, and interhemispheric conduction through the corpus callosum is significantly slowed 13 FIGURE 5. Small complex MEP recorded from the abductor digiti in this disease. Ipsilateral cortical stimulation minimi in ALS at 45% stimulator output. Threshold 35%. Note the causes transcallosal inhibition of a contracting target late component with a latency of 45.6 ms which corresponds to muscle, and this fact can be used to measure con- the second component of the double primary peak (see Fig. 4) duction through the corpus callosum. A significantly and may reflect activation of slow conducting cortical motor neu- ronal connections. increased excitability threshold in resting or preac- tivated muscles is frequent. This is usually associated with prolonged central conduction but may also oc- to stimulate the motor cortex with a magnetic coil is cur as an isolated abnormality. MEP studies may de- reduced early in the disease, especially in patients tect subclinical involvement of motor pathways and with preserved muscle bulk and prominent fascicu- the overall sensitivity is comparable to visual evoked lations.89 Other TMS studies suggest that cortical in- potentials.57 hibitory mechanisms are also impaired in ALS.95 For example, the cortical silent period, a measure of cor- ticospinal inhibition, is shortened compared to nor- mal subjects (Fig. 3, bottom), and a subthreshold conditioning stimulus delivered shortly before a su- prathreshold test stimulus fails to inhibit the test re- sponse in ALS.60 PSTHs in patients with ALS show a diversity of abnormalities ranging from the primary peak being small (or absent) to being large and increased in temporal dispersion.133 Over time, the dispersion in- creases and double primary peaks occur, suggesting activation of slow-conducting indirect pathways. In- direct evidence suggests that these abnormalities are supraspinal in origin and are not the result of AHC disease. In Kennedy’s disease (bulbar-spinal muscu- FIGURE 6. Delayed (upper limit 25 ms) and small MEP with lar atrophy), the primary peak of the PSTH is nor- variability of onset latency and dispersed morphology from the mal (Fig. 4, middle), which confirms that the abnor- abductor pollicis brevis in multiple sclerosis at 90% stimulator mal PSTH in ALS is due to supraspinal disease.133 output. Threshold 80%.

Magnetic Stimulation MUSCLE & NERVE February 2002 169 Studies with PSTHs in MS have shown delayed Hereditary Spastic Paraplegia and Spinocerebellar and dispersed primary peaks consisting of multiple Ataxias. In patients with hereditary spastic paraple- subpeaks.12 A similar abnormality is seen in ALS but gia, lower limb responses are almost always abnor- the underlying mechanism is different. In MS, con- mal: absent, reduced, or delayed. Upper limb re- duction through the descending motor tracts is de- sponses, however, are usually normal even in the layed, whereas in ALS, conduction slowing and tem- presence of clinical upper motor neuron signs. A poral dispersion is caused by selective loss of large, similar pattern can be seen in patients with heredi- fast-conducting pyramidal neurons.135 tary motor and sensory neuropathy with pyramidal signs. The CMCT to small hand muscles in Friedreich’s Movement Disorders. Conduction time through ataxia is most often prolonged.41 Moreover, the MEP the descending motor pathways is normal in Parkin- is frequently of small amplitude and dispersed. The son’s disease, Huntington’s disease, primary dysto- sensitivity is even greater when recording from lower nia, essential tremor, and myoclonus.7,23 Determin- limb muscles. In other cerebellar ataxias, abnormali- ing the cortical threshold in Parkinson’s disease has ties are less severe and less frequent, with the highest produced inconsistent results: decreased, normal, rate of impairment being seen in spinocerebellar and elevated thresholds have all been reported.23 ataxias. Prolongation of central motor conduction is The cortical silent period is shortened or normal also a common finding in patients with human T-cell and, when short, the abnormality can be reversed lymphotrophic virus type I–associated tropical spas- after levodopa therapy.108 Corticocortical inhibition, tic paraparesis. Responses in the lower limbs typically tested at short conditioning test intervals and with show marked prolongation. Upper limb responses the muscle at rest, is reduced in Parkinson’s disease. may be normal or show slowing of central conduc- On the other hand, interstimulus intervals of be- tion less prominent than recordings from leg tween 40 to 75 ms show greater than normal inhibi- muscles.138 tion of the test response.9 The physiological abnor- malities in Parkinson’s disease revealed by TMS Epilepsy and Drugs. Attempts have been made to probably result from a combination of increased in- use TMS for localization of epileptic foci, but it ap- hibition and reduced excitation occurring at both pears that the epileptic focus cannot be localized cortical and subcortical levels.7 In dystonia and Hun- with sufficient resolution using this approach.22,32,136 tington’s disease, double stimulation paradigms have One would expect that cortical excitability might be produced conflicting findings, most likely due to dif- increased in patients with epilepsy but threshold ferent stimulation parameters.2,59,118 Nevertheless, it measurements have revealed conflicting results. In- is likely that future studies will reveal useful insight tracortical inhibition in epilepsy is reduced, but this into the pathophysiological mechanisms and mode is a nonspecific finding which can be seen in many of drug action. other disorders. It is unclear whether changes in cor- tical excitability are due to medication or to epilepsy Stroke. In stroke patients, the response after corti- itself. Antiepileptic drugs which act on sodium chan- cal stimulation is often absent.106 In patients in nels (carbamazepine, phenytoin, lamotrigine) in- whom a response is obtained, the MEP is quite often crease motor threshold but do not have a significant of small amplitude and dispersed. CMCT is usually effect on intracortical inhibition. In contrast, antiep- only slightly prolonged. The cortical threshold is ileptic drugs or medication modulating activity of commonly found to be raised.25 In a formal study, GABA receptors (e.g., benzodiazepines) have no sig- the duration of the silent period was markedly nificant effect on motor threshold but enhance in- longer on the affected side when compared with a tracortical inhibition and suppress intracortical fa- control group. This parameter also seems to detect cilitation.139 In patients evaluated for epilepsy mild, subclinical disturbances.1 TMS appears to be a surgery, rTMS applied to the dominant hemisphere good predictor of stroke outcome.3,33,60,123,125,126 A can produce speech arrest but does not always cor- recordable MEP in early stages correlates with a fa- respond directly with Wada test results.10,51 vorable outcome, whereas an absent response pre- dicts poor recovery. Patients with delayed but pres- Radiculopathy and Spondylotic Myelopathy. Mag- ent MEPs recover more slowly than those with netic stimulation over the spinal enlargements ex- normal MEPs, but are similar at 12 months. The cites the nerve roots a few centimeters distal to the CMCT correlates well with the grade of weakness. AHC in the vicinity of the intervertebral foramen.30 The finding of an increased threshold correlates The response latency is reproducible but the stimu- with the presence of brisk tendon jerks. lus is usually submaximal. This precludes standard-

170 Magnetic Stimulation MUSCLE & NERVE February 2002 ization based on the amplitude of the response and focality and inability to elicit a potential of consistent detection of a more distal conduction block. The maximal amplitude. This precludes, amongst other value of magnetic root stimulation to evaluate ra- things, accurate detection of conduction block. diculopathies is thus limited. As with other conduc- However, there are at least two situations in which tion techniques used to evaluate radiculopathies (F use of the coil is advantageous. In children who do waves, somatosensory evoked potentials, H reflexes, not tolerate electrical stimulation, magnetic stimula- and magnetic stimulation), conduction block is dif- tion often allows measurement of conduction veloc- ficult to interpret given the variability of MEP ampli- ities sufficient to differentiate between a demyelinat- tude and uncertainty in obtaining a maximum am- ing and an axonal neuropathy. Secondly, in plitude potential. demyelinating neuropathies, cortical stimulation A high percentage of abnormalities in the MEP can reveal marked conduction slowing in the most has been described in spondylotic myelopathy.45 proximal nerve segments. The F wave can often do The CMCT is frequently prolonged, the threshold the same, but when the neuropathy is severe it may raised, and the response dispersed and of small am- be absent. plitude. Abnormalities of central conduction may precede clinical evidence of myelopathy. Slowed CRANIAL NERVES central conduction may be an early manifestation of The intracranial portions of the motor cranial nerves cord compression before it is evident on magnetic V, VII, XI, and XII are readily stimulated with a mag- resonance imaging.124 netic coil.6,21,129 To elicit responses from muscles innervated through the cranial nerves at their intra- Plexopathy. Although magnetic stimulation can be cranial-extramedullary portion, the magnetic coil used in plexopathies, the technique has not been should be positioned over the occiput ipsilateral to able to substitute for electrical stimulation. With the recording site. Evidence indicates that the nerves magnetic stimulation at the plexus level, supramaxi- are excited close or just distal to their exit foramina. mal responses are not always possible and the precise Because proximity of the stimulus to the surface re- site of stimulation is uncertain.40,52,112,116 However, cording electrodes can be a problem, a concentric magnetic stimulation provides certain advantages in needle electrode is often preferred for recording the some types of plexus lesions. For example, a elicited muscle response. An intra-oral “permucosal” neurapraxic lesion of the upper trunk of the bra- recording device is also helpful to reduce artifact. chial plexus cannot be detected by electrical stimu- The central, crossed, corticopontine portion of lation of Erb’s point, which is usually below the le- the motor cranial nerve conduction is more difficult sion. Accurate localization would require direct to assess. The coil is optimally placed 4 cm lateral to electrical stimulation of the spinal roots through a the vertex on a line joining the vertex (Cz) and the monopolar needle. This can be achieved noninva- external auditory meatus. Activation of the target sively by magnetic stimulation. Eliciting a response muscle is usually required to obtain a response. from the deltoid or biceps is clear evidence of nerve continuity, and significant slowing of onset latencies DIAPHRAGMATIC CONDUCTION would indicate focal demyelination.95 MEP ampli- Diaphragmatic recording is used routinely to diag- tudes after magnetic plexus stimulation are variable nose and monitor patients with impaired respiratory and one cannot comment as to the presence or ab- function. Although electrical stimulation of the sence of conduction block. Once there has been sig- phrenic nerve is well established, cervical magnetic nificant axonal loss, needle EMG is the best method stimulation of the phrenic nerves is less painful and of determining axonal continuity and reinnervation. achieves a more constant degree of diaphragmatic Stimulation of the lumbosacral plexus and cauda recruitment.120,140 An unexplained phenomenon is equina is also possible with magnetic stimulation, the greater transdiaphragmatic twitch pressure that but, like stimulation of the brachial plexus, it some- occurs with magnetic rather than electrical stimula- times fails to elicit reproducible, maximal re- tion. This may be due to coactivation of extradia- sponses.17,31,80 This again limits its value in the as- phragmatic muscles. Normal values for the latency to sessment of lumbosacral radiculopathies and the diaphragm using electrical stimulation in the plexopathies.31 neck are between 7 and 8 ms, but data are less ho- mogeneous for magnetic stimulation. Peripheral Neuropathies. As previously mentioned, The diagnosis of impaired central respiratory magnetic stimulation is presently limited with regard drive can often be accomplished by transcortical to the peripheral nervous system because of lack of magnetic stimulation of the motor cortex with re-

Magnetic Stimulation MUSCLE & NERVE February 2002 171 cording of the diaphragm and phrenic nerve re- 6. Benecke R, Meyer BU, Schonle P, Conrad B. Transcranial mag- netic stimulation of the human brain: responses in muscles sponses. These studies are of particular value in criti- supplied by cranial nerves. Exp Brain Res 1988;71:623–632. cally ill patients in whom both central and periph- 7. Berardelli A. Transcranial magnetic stimulation in move- eral lesions may impair respiration.140 Phrenic nerve ment disorders. Electroencephalogr Clin Neurophysiol pacing is becoming a more frequent substitute for Suppl 1999;51:276–280. 8. Berardelli A, Inghilleri M, Rothwell JC, Romeo S, Curra A, positive pressure ventilation via tracheotomy in pa- Gilio F, Modugno N, Manfredi M. Facilitation of muscle tients with high cervical cord lesions or central hy- evoked responses after repetitive cortical stimulation in poventilation. Although its indications are infre- man. Exp Brain Res 1998;122:79–84. 9. Berardelli A, Rona S, Inghilleri M, Manfredi M. Cortical quent, TMS may help to determine which patients inhibition in Parkinson’s disease. A study with paired mag- may benefit from this treatment.120 netic stimulation. Brain 1996;119:71–77. 10. Bhatnagar SC, Mandybur GT, Buckingham HW, Andy OJ. Language representation in the human brain: evidence INTRAOPERATIVE MONITORING OF MOTOR from cortical mapping. Brain Lang 2000;74:238–259. EVOKED POTENTIALS 11. Binkofski F, Classen J, Benecke R. Stimulation of peripheral The ability to evoke MEPs during surgery has been a nerves using a novel magnetic coil. Muscle Nerve 1999;22: 751–757. useful addition to the battery of neurophysiological 12. Boniface SJ, Mills KR, Schubert M. Responses of single spinal tests that can be used to monitor and prevent the motoneurons to magnetic brain stimulation in healthy sub- development of clinical deficits during surgery.5,20 jects and patients with multiple sclerosis. Brain 1991;114: 643–662. MEP monitoring is particularly relevant in surgery 13. Boroojerdi B, Hungs M, Mull M, Topper R, Noth J. Inter- that may damage the motor pathways independently hemispheric inhibition in patients with multiple sclerosis. of the sensory pathways. Examples of this include Electroencephalogr Clin Neurophysiol 1998;109:230–237. 14. Brasil-Neto JP, Cohen LG, Panizza M, Nilsson J, Roth BJ, resection of spinal cord tumors, cross-clamping of Hallett M. Optimal focal transcranial magnetic activation of cerebral blood vessels, and resection of tumors and the human motor cortex: effects of coil orientation, shape of arteriovenous malformations involving the motor the induced current pulse, and stimulus intensity. J Clin Neurophysiol 1992;9:132–136. cortex and subcortical motor pathways. However, 15. Brasil-Neto JP, Valls-Sole´ J, Pascual-Leone A, Cammarota A, intraoperative monitoring is expensive, requiring Amassian VE, Cracco R, Maccabee P, Cracco J, Hallett M, Co- dedicated teams and equipment, and its cost-effec- hen LG. Rapid modulation of human cortical motor outputs following ischaemic nerve block. Brain 1993;116:511–525. tiveness needs to be considered. 16. Britton TC, Meyer BU, Benecke R. Variability of cortically Electrical stimulation of the cortex with record- evoked motor responses in multiple sclerosis. Electroen- ing over the spine (preferably using intradural elec- cephalogr Clin Neurophysiol 1991;81:186–194. 17. Brodak PP, Bidair M, Joseph A, Szollar S, Juma S. Magnetic trodes) is the most reliable method because it is in- stimulation of the sacral roots. Neurourol Urodyn 1993;12: dependent of the anesthetic used. However, it is 533–540. possible to use magnetic stimulation with some an- 18. Brouwer B, Ashby P. Altered corticospinal projections to esthetics such as ketamine or fentanyl and obtain lower limb motoneurons in subjects with cerebral palsy. Brain 1991;114:1395–1407. reasonably reliable results. If trains of stimuli are 19. Burke D, Hicks R, Gandevia SC, Stephen J, Woodforth I, used rather than a single stimulus, the facilitation Crawford M. Direct comparison of corticospinal volleys in produced helps overcome the effects of some anes- human subjects to transcranial magnetic and electrical stimulation. J Physiol 1993;470:383–393. thetics. 20. Burke D, Hicks RG. Surgical monitoring of motor pathways. J Clin Neurophysiol 1998;15:194–205. 21. Campos A, Barona R, Escudero J, Montalt J, Escudero M. REFERENCES Hypoglossal nerve conduction study by transcranial mag- netic stimulation in normal subjects. Otolaryngol Head 1. Ahonen JP, Jehkonen M, Dastidar P, Molnar G, Hakkinen V. Neck Surg 1995;112:520–525. Cortical silent period evoked by transcranial magnetic stimu- 22. Cantello R, Civardi C, Cavalli A, Varrasi C, Tarletti R, Mo- lation in ischemic stroke. Electroencephalogr Clin Neuro- naco F, Migliaretti G. Cortical excitability in cryptogenic lo- physiol 1998;109:224–229. calization-related epilepsy: interictal transcranial magnetic 2. Amadio S, Panizza M, Pisano F, Maderna L, Miscio C, Nils- stimulation studies. Epilepsia 2000;41:694–704. son J, Volonte MA, Comi G, Galardi G. Transcranial mag- 23. Cantello R, Gianelli M, Bettucci D, Civardi C, De Angelis MS, netic stimulation and silent period in spasmodic torticollis. Mutani R. Parkinson’s disease rigidity: magnetic motor Am J Phys Med Rehabil 2000;79:361–368. evoked potentials in a small hand muscle. Neurology 1991; 3. Arac N, Sagduyu A, Binai S, Ertekin C. Prognostic value of 41:1449–1456. transcranial magnetic stimulation in acute stroke. Stroke 24. Caramia MD, Cicinelli P, Paradiso C, Mariorenzi R, Zarola F, 1994;25:2183–2186. Bernardi G, Rossini PM. Excitability changes of muscular 4. Barker AT, Freeston IL, Jabinous R, Jarratt JA. Clinical evalu- responses to magnetic brain stimulation in patients with cen- ation of conduction time measurements in central motor tral motor disorders. Electroencephalogr Clin Neurophysiol pathways using magnetic stimulation of human brain. Lan- 1991;81:243–250. cet 1986;1:1325–1326. 25. Catano A, Houa M, Caroyer JM, Ducarne H, Noel P. Mag- 5. Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic netic transcranial stimulation in acute stroke: early excita- stimulation of human motor cortex. Lancet 1985;1:1106– tion threshold and functional prognosis. Electroencepha- 1107. logr Clin Neurophysiol 1996;101:233–239.

172 Magnetic Stimulation MUSCLE & NERVE February 2002 26. Chen R, Classen J, Gerloff C, Celnik P, Wassermann EM, 45. de Noordhout AM, Myressiotis S, Delvaux V, Born JD, Del- Hallett M, Cohen LG. Depression of motor cortex excitabil- waide PJ. Motor and somatosensory evoked potentials in cer- ity by low-frequency transcranial magnetic stimulation. Neu- vical spondylotic myelopathy. Electroencephalogr Clin Neu- rology 1997;48:1398–1403. rophysiol 1998;108:24–31. 27. Chen R, Hallett M. The time course of changes in motor 46. de Noordhout M, Pepin JL, Gerard P, Delwaide PJ. Facilita- cortex excitability associated with voluntary movement. Can tion of responses to motor cortex stimulation: effects of iso- J Neurol Sci 1999;26:163–169. metric voluntary contraction. Ann Neurol 1992;32:365–370. 28. Chen R, Lozano AM, Ashby P. Mechanism of the silent pe- 47. Desiato MT, Caramia MD. Towards a neurophysiological riod following transcranial magnetic stimulation. Evidence marker of amyotrophic lateral sclerosis as revealed by from epidural recordings. Exp Brain Res 1999;128:539–542. changes in cortical excitability. Electroencephalogr Clin 29. Chen R, Samii A, Canos M, Wassermann EM, Hallett M. Neurophysiol 1997;105:1–7. Effects of phenytoin on cortical excitability in humans. Neu- 48. Devanne H, Lavoie BA, Capaday C. Input-output properties rology 1997;49:881–883. and gain changes in the human corticospinal pathway. Exp 30. Chokroverty S, Picone MA, Chokroverty M. Percutaneous Brain Res 1997;114:329–338. magnetic coil stimulation of human cervical vertebral col- 49. Di Lazzaro V, Oliviero A, Profice P, Insola A, Mazzone P, umn: site of stimulation and clinical application. Electroen- Tonali P, Rothwell JC. Direct recordings of descending vol- cephalogr Clin Neurophysiol 1991;81:359–365. leys after transcranial magnetic and electric motor cortex 31. Chokroverty S, Sachdeo R, Dilullo J, Duvoisin RC. Magnetic stimulation in conscious humans. Electroencephalogr Clin stimulation in the diagnosis of lumbosacral radiculopathy. J Neurophysiol Suppl 1999;51:120–126. Neurol Neurosurg Psychiatry 1989;52:767–772. 50. Eisen A, Shytbel W, Murphy K, Hoirch M. Cortical magnetic 32. Cicinelli P, Mattia D, Spanedda F, Traversa R, Marciani MG, stimulation in amyotrophic lateral sclerosis. Muscle Nerve Pasqualetti P, Rossini PM, Bernardi G. Transcranial mag- 1990;13:146–151. netic stimulation reveals an interhemispheric asymmetry of 51. Epstein CM, Woodard JL, Stringer AY, Bakay RA, Henry TR, cortical inhibition in focal epilepsy. Neuroreport 2000;11: Pennell PB, Litt B. Repetitive transcranial magnetic stimula- 701–707. tion does not replicate the Wada test. Neurology 2000;55: 33. Cicinelli P, Traversa R, Rossini PM. Post-stroke reorganiza- 1025–1027. tion of brain motor output to the hand: a 2–4 month follow- 52. Evans BA. Magnetic stimulation of the peripheral nervous up with focal magnetic transcranial stimulation. Electroen- system. J Clin Neurophysiol 1991;8:77–84. cephalogr Clin Neurophysiol 1997;105:438–450. 53. Fuhr P, Agostino R, Hallett M. Spinal motor neuron excit- 34. Classen J, Schnitzler A, Binkofski F, Werhahn KJ, Kim YS, ability during the silent period after cortical stimulation. Kessler KR, Benecke R. The motor syndrome associated with Electroencephalogr Clin Neurophysiol 1991;81:257–262. exaggerated inhibition within the primary motor cortex of 54. Furby A, Bourriez JL, Jacquesson JM, Mounier-Vehier F, patients with hemiparetic. Brain 1997;120:605–619. Guieu JD. Motor evoked potentials to magnetic stimulation: 35. Classen J, Witte OW, Schlaug G, Seitz RJ, Holthausen H, technical considerations and normative data from 50 sub- Benecke R. Epileptic seizures triggered directly by focal jects. J Neurol 1992;239:152–156. transcranial magnetic stimulation. Electroencephalogr Clin 55. George MS, Wassermann EM, Kimbrell TA, Little JT, Willi- Neurophysiol 1995;94:19–25. ams WE, Danielson AL, Greenberg BD, Hallett M, Post RM. 36. Claus D. Central motor conduction: method and normal Mood improvement following daily left prefrontal repetitive results. Muscle Nerve 1990;13:1125–1132. transcranial magnetic stimulation in patients with depres- 37. Cohen LG, Brasil-Neto JP, Pascual-Leone A, Hallett M. Plas- sion: a placebo-controlled crossover trial. Am J Psychiatry ticity of cortical motor output organization following deaf- 1997;154:1752–1756. ferentation, cerebral lesions, and skill acquisition. Adv Neu- 56. Girlanda P, Quartarone A, Battaglia F, Picciolo G, Sinicropi rol 1993;63:187–200. S, Messina C. Changes in spinal cord excitability in patients 38. Cohen LG, Roth BJ, Wassermann EM, Topka H, Fuhr P, affected by ulnar neuropathy. Neurology 2000;55:975–978. Schultz J, Hallett M. Magnetic stimulation of the human 57. Guerit JM, Monje AA. The sensitivity of multimodal evoked cerebral cortex, an indicator of reorganization in motor potentials in multiple sclerosis. A comparison with magnetic pathways in certain pathological conditions. J Clin Neuro- resonance imaging and cerebrospinal fluid analysis. Electro- physiol 1991;8:56–65. encephalogr Clin Neurophysiol 1988;70:230–238. 39. Conde F, Lund JS, Jacobowitz DM, Baimbridge KG, Lewis DA. 58. Hanajima R, Ugawa Y. Impaired motor cortex inhibition in Local circuit neurons immunoreactive for calretinin, calbindin patients with ALS: evidence from paired transcranial mag- D-28k or parvalbumin in monkey prefrontal cortex: distribu- netic stimulation. Neurology 1998;51:1771–1772. tion and morphology. J Comp Neurol 1994;341:95–116. 59. Hanajima R, Ugawa Y. Intracortical inhibition of the motor 40. Cros D, Gominak S, Shahani B, Fang J, Day B. Comparison cortex in movement disorders. Brain Dev 2000;22(suppl 1): of electric and magnetic coil stimulation in the supraclavic- 132–135. ular region. Muscle Nerve 1992;15:587–590. 60. Heald A, Bates D, Cartlidge NE, French JM, Miller S. Lon- 41. Cruz-Martinez A, Palau F. Central motor conduction time by gitudinal study of central motor conduction time following magnetic stimulation of the cortex and peripheral nerve stroke. 1. Natural history of central motor conduction. Brain conduction follow-up studies in Friedreich’s ataxia. Electro- 1993;116:1355–1370. encephalogr Clin Neurophysiol 1997;105:458–461. 61. Hess CW, Mills KR, Murray NM. Responses in small hand 42. Day BL, Dressler D, Maertens de Noordhout A, Marsden CD, muscles from magnetic stimulation of the human brain. J Nakashima K, Rothwell JC, Thompson PD. Electric and mag- Physiol 1987;388:397–419. netic stimulation of human motor cortex: surface EMG and 62. Hess CW, Mills KR, Murray NM, Schriefer TN. Magnetic single motor unit responses. J Physiol 1989;412:449–473. brain stimulation: central motor conduction studies in mul- 43. Day BL, Rothwell JC, Thompson PD, Dick JP, Cowan JM, tiple sclerosis. Ann Neurol 1987;22:744–752. Berardelli A, Marsden CD. Motor cortex stimulation in in- 63. Hicks R, Burke D, Stephen J, Woodforth I, Crawford M. tact man. 2. Multiple descending volleys. Brain 1987;110: Corticospinal volleys evoked by electrical stimulation of hu- 1191–1209. man motor cortex after withdrawal of volatile anaesthetics. J 44. de Carvalho M, Miranda PC, Luis ML, Ducla-Soares E. Cor- Physiol 1992;456:393–404. tical muscle representation in amyotrophic lateral sclerosis 64. Homberg V, Netz J. Generalised seizures induced by trans- patients: changes with disease evolution. Muscle Nerve 1999; cranial magnetic stimulation of motor cortex. Lancet 1989; 22:1684–1692. 2:1223.

Magnetic Stimulation MUSCLE & NERVE February 2002 173 65. Hufnagel A, Elger CE. Responses of the epileptic focus to 85. Mills KR, Boniface SJ, Schubert M. Magnetic brain stimula- transcranial magnetic stimulation. Electroencephalogr Clin tion with a double coil: the importance of coil orientation. Neurophysiol Suppl 1991;43:86–99. Electroencephalogr Clin Neurophysiol 1992;85:17–21. 66. Hufnagel A, Elger CE, Durwen HF, Boker DK, Entzian W. 86. Mills KR, Murray NM, Hess CW. Magnetic and electrical Activation of the epileptic focus by transcranial magnetic transcranial brain stimulation: physiological mechanisms stimulation of the human brain. Ann Neurol 1990;27:49–60. and clinical applications. Neurosurgery 1987;20:164–168. 67. Hufnagel A, Elger CE, Klingmuller D, Zierz S, Kramer R. 87. Mills KR, Nithi KA. Corticomotor threshold is reduced in Activation of epileptic foci by transcranial magnetic stimula- early sporadic amyotrophic lateral sclerosis. Muscle Nerve tion: effects on secretion of prolactin and luteinizing hor- 1997;20:1137–1141. mone. J Neurol 1990;237:242–246. 88. Mills KR, Nithi KA. Corticomotor threshold to magnetic 68. Ingram DA, Swash M. Central motor conduction is abnor- stimulation: normal values and repeatability. Muscle Nerve mal in motor neuron disease. J Neurol Neurosurg Psychiatry 1997;20:570–576. 1987;50:159–166. 69. Ingram DA, Thompson AJ, Swash M. Central motor conduc- 89. Miranda PC, de Carvalho M, Conceicao I, Luis ML, Ducla- tion in multiple sclerosis: evaluation of abnormalities re- Soares E. A new method for reproducible coil positioning in vealed by transcutaneous magnetic stimulation of the brain. transcranial magnetic stimulation mapping. Electroen- J Neurol Neurosurg Psychiatry 1988;51:487–494. cephalogr Clin Neurophysiol 1997;105:116–123. 70. Kernell D, Chien-Ping W. Post-synaptic effects of cortical 90. Mortifee P, Stewart H, Schulzer M, Eisen A. Reliability of stimulation on forelimb motoneurones in the baboon. J transcranial magnetic stimulation for mapping the human Physiol 1967;191:673–690. motor cortex. Electroencephalogr Clin Neurophysiol 1994; 71. Kiers L, Cros D, Chiappa KH, Fang J. Variability of motor 93:131–137. potentials evoked by transcranial magnetic stimulation. Elec- 91. Muellbacher W, Artner C, Mamoli B. The role of the intact troencephalogr Clin Neurophysiol 1993;89:415–423. hemisphere in recovery of midline muscles after recent 72. Kiers L, Fernando B, Tomkins D. Facilitatory effect of think- monohemispheric stroke. J Neurol 1999;246:250–256. ing about movement on magnetic motor-evoked potentials. 92. Muller K, Homberg V, Lenard HG. Magnetic stimulation of Electroencephalogr Clin Neurophysiol 1997;105:262–268. motor cortex and nerve roots in children. Maturation of 73. Kohara N, Kaji R, Kojima Y, Kimura J. An electrophysiologi- cortico-motoneuronal projections. Electroencephalogr Clin cal study of the corticospinal projections in amyotrophic lat- Neurophysiol 1991;81:63–70. eral sclerosis. Clin Neurophysiol 1999;110:1123–1132. 93. Naka D, Mills KR. Further evidence for corticomotor hyper- 74. Krings T, Naujokat C, von Keyserlingk DG. Representation excitability in amyotrophic lateral sclerosis. Muscle Nerve of cortical motor function as revealed by stereotactic trans- 2000;23:1044–1050. cranial magnetic stimulation. Electroencephalogr Clin Neu- rophysiol 1998;109:85–93. 94. Nakamura H, Kitagawa H, Kawaguchi Y, Tsuji H. Intracorti- cal facilitation and inhibition after transcranial magnetic 75. Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, stimulation in conscious humans. J Physiol 1997;498: Ferbert A, Wroe S, Asselman P, Marsden CD. Corticocortical 817–823. inhibition in human motor cortex. J Physiol 1993;471: 501–519. 95. Oge AE, Boyaciyan A, Gurvit H, Yazici J, Degirmenci M, 76. Liepert J, Miltner WH, Bauder H, Sommer M, Dettmers C, Kantemir E. Magnetic nerve root stimulation in two types of Taub E, Weiller C. Motor cortex plasticity during constraint- brachial plexus injury: segmental demyelination and axonal induced movement therapy in stroke patients. Neurosci Lett degeneration. Muscle Nerve 1997;20:823–832. 1998;250:5–8. 96. Pascual-Leone A, Cohen LG, Brasil-Neto JP, Hallett M. Non- 77. Liepert J, Schwenkreis P, Tegenthoff M, Malin JP. The glu- invasive differentiation of motor cortical representation of tamate antagonist riluzole suppresses intracortical facilita- hand muscles by mapping of optimal current directions. tion. J Neural Transm 1997;104:1207–1214. Electroencephalogr Clin Neurophysiol 1994;93:42–48. 78. Maccabee PJ, Amassian VE, Eberle LP, Cracco RQ. Magnetic 97. Pascual-Leone A, Cohen LG, Shotland LI, Dang N, Pikus A, coil stimulation of straight and bent amphibian and mam- Wassermann EM, Brasil-Neto JP, Valls-Sole´ J, Hallett M. No malian peripheral nerve in vitro: locus of excitation. J Phys- evidence of hearing loss in humans due to transcranial mag- iol 1993;460:201–219. netic stimulation. Neurology 1992;42:647–651. 79. Maccabee PJ, Amassian VE, Eberle LP, Rudell AP, Cracco 98. Pascual-Leone A, Gates JR, Dhuna A. Induction of speech RQ, Lai KS, Somasundarum M. Measurement of the electric arrest and counting errors with rapid-rate transcranial mag- field induced into inhomogeneous volume conductors by netic stimulation. Neurology 1991;41:697–702. magnetic coils: application to human spinal neurogeometry. 99. Pascual-Leone A, Nguyet D, Cohen LG, Brasil-Neto JP, Cam- Electroencephalogr Clin Neurophysiol 1991;81:224–237. marota A, Hallett M. Modulation of muscle responses 80. Macdonell RA, Cros D, Shahani BT. Lumbosacral nerve root evoked by transcranial magnetic stimulation during the ac- stimulation comparing electrical with surface magnetic coil quisition of new fine motor skills. J Neurophysiol 1995;74: techniques. Muscle Nerve 1992;15:885–890. 1037–1045. 81. Magistris MR, Rosler KM, Truffert A, Landis T, Hess CW. A 100. Pascual-Leone A, Peris M, Tormos JM, Pascual AP, Catala clinical study of motor evoked potentials using a triple stimu- MD. Reorganization of human cortical motor output maps lation technique. Brain 1999;122:265–279. following traumatic forearm amputation. Neuroreport 1996; 82. Magistris MR, Rosler KM, Truffert A, Myers JP. Transcranial 7:2068–2070. stimulation excites virtually all motor neurons supplying the target muscle. A demonstration and a method improving the 101. Pascual-Leone A, Rubio B, Pallardo F, Catala MD. Rapid-rate study of motor evoked potentials. Brain 1998;121:437–450. transcranial magnetic stimulation of left dorsolateral pre- 83. Maier MA, Illert M, Kirkwood PA, Nielsen J, Lemon RN. frontal cortex in drug-resistant depression. Lancet 1996;348: Does a C3–C4 propriospinal system transmit corticospinal 233–237. excitation in the primate? An investigation in the macaque 102. Pascual-Leone A, Torres F. Plasticity of the sensorimotor cor- monkey. J Physiol 1998;511:191–212. tex representation of the reading finger in Braille readers. 84. Masur H, Ludolph AC, Hilker E, Hengst K, Knuth U, Rolf Brain 1993;116:39–52. LH, Bals-Pratsch M. Transcranial magnetic stimulation: in- 103. Pascual-Leone A, Valls-Sole´ J, Wassermann EM, Hallett M. fluence on plasma levels of hormones of the anterior pitu- Responses to rapid-rate transcranial magnetic stimulation of itary gland and of cortisol? Funct Neurol 1991;6:59–63. the human motor cortex. Brain 1994;117:847–858.

174 Magnetic Stimulation MUSCLE & NERVE February 2002 104. Pascual-Leone A, Wassermann EM, Sadato N, Hallett M. The 122. Thompson PD, Day BL, Rothwell JC, Dressler D, Maertens role of reading activity on the modulation of motor cortical de Noordhout A, Marsden CD. Further observations on the outputs to the reading hand in Braille readers. Ann Neurol facilitation of muscle responses to cortical stimulation by 1995;38:910–915. voluntary contraction. Electroencephalogr Clin Neuro- 105. Patton HD, Amassian VE. Single- and multiple-unit analysis physiol 1991;81:397–402. of cortical stage of pyramidal tract activation. J Neurophysiol 123. Traversa R, Cicinelli P, Bassi A, Rossini PM, Bernardi G. 2001;345–363. Mapping of motor cortical reorganization after stroke. A 106. Pennisi G, Rapisarda G, Bella R, Calabrese V, Maertens de brain stimulation study with focal magnetic pulses. Stroke Noordhout A, Delwaide PJ. Absence of response to early 1997;28:110–117. transcranial magnetic stimulation in ischemic stroke pa- 124. Travlos A, Pant B, Eisen A. Transcranial magnetic stimula- tients: prognostic value for hand motor recovery. Stroke tion for detection of preclinical cervical spondylotic myelop- 1999;30:2666–2670. athy. Arch Phys Med Rehabil 1992;73:442–446. 107. Porter R, Lemon RN. Corticospinal function and voluntary 125. Trompetto C, Assini A, Buccolieri A, Marchese R, Abbruzz- movement. Monographs of the Physiological Society, No. 45. ese G. Motor recovery following stroke: a transcranial mag- Oxford: Clarendon Press; 1993. p 122–209. netic stimulation study. Clin Neurophysiol 2000;111: 108. Priori A, Berardelli A, Inghilleri M, Accornero N, Manfredi 1860–1867. M. Motor cortical inhibition and the dopaminergic system. 126. Turton A, Wroe S, Trepte N, Fraser C, Lemon RN. Contra- Pharmacological changes in the silent period after transcra- lateral and ipsilateral EMG responses to transcranial mag- nial brain stimulation in normal subjects, patients with Par- netic stimulation during recovery of arm and hand function kinson’s disease and drug-induced parkinsonism. Brain after stroke. Electroencephalogr Clin Neurophysiol 1996; 1994;117:317–323. 101:316–328. 109. Prout AJ, Eisen AA. The cortical silent period and amyotro- 127. Ugawa Y, Kanazawa I. Motor-evoked potentials: unusual find- phic lateral sclerosis. Muscle Nerve 1994;17:217–223. ings. Clin Neurophysiol 1999;110:1641–1645. 110. Ridding MC, Sheean G, Rothwell JC, Inzelberg R, Kujirai T. 128. Ugawa Y, Uesaka Y, Terao Y, Suzuki M, Sakai K, Hanajima R, Changes in the balance between motor cortical excitation Kanazawa I. Clinical utility of magnetic corticospinal tract and inhibition in focal, task specific dystonia. J Neurol Neu- stimulation at the foramen magnum level. Electroencepha- rosurg Psychiatry 1995;59:493–498. logr Clin Neurophysiol 1996;101:247–254. 111. Rijntjes M, Tegenthoff M, Liepert J, Leonhardt G, Kotterba 129. Urban PP, Hopf HC, Connemann B, Hundemer HP, S, Muller S, Kiebel S, Malin JP, Diener HC, Weiller C. Cor- Koehler J. The course of cortico-hypoglossal projections in tical reorganization in patients with facial palsy. Ann Neurol the human brainstem. Functional testing using transcranial 1997;41:621–630. magnetic stimulation. Brain 1996;119:1031–1038. 112. Rosenberg JN, Turchetta J. Magnetic coil stimulation of the 130. Wassermann EM. Risk and safety of repetitive transcranial brachial plexus. Arch Phys Med Rehabil 1993;74:928–932. magnetic stimulation: report and suggested guidelines from 113. Rosler KM, Truffert A, Hess CW, Magistris MR. Quantifica- the International Workshop on the Safety of Repetitive tion of upper motor neuron loss in amyotrophic lateral scle- Transcranial Magnetic Stimulation, June 5–7, 1996. Electro- rosis. Clin Neurophysiol 2000;111:2208–2218. encephalogr Clin Neurophysiol 1998;108:1–16. 114. Rossini PM, Caltagirone C, Castriota-Scanderbeg A, Cicinelli 131. Wassermann EM, McShane LM, Hallett M, Cohen LG. Non- P, Del Gratta C, Demartin M, Pizzella V, Traversa R, Romani invasive mapping of muscle representations in human motor GL. Hand motor cortical area reorganization in stroke: a cortex. Electroencephalogr Clin Neurophysiol 1992;85:1–8. study with fMRI, MEG and TCS maps. Neuroreport 1998;9: 2141–2146. 132. Wassermann EM, Wang B, Zeffiro TA, Sadato N, Pascual- 115. Rothwell JC, Thompson PD, Day BL, Dick JP, Kachi T, Leone A, Toro C, Hallett M. Locating the motor cortex on Cowan JM, Marsden CD. Motor cortex stimulation in intact the MRI with transcranial magnetic stimulation and PET. man. 1. General characteristics of EMG responses in differ- Neuroimage 1996;3:1–9. ent muscles. Brain 1987;110:1173–1190. 133. Weber M, Eisen A. Assessment of upper and lower motor 116. Schmid UD, Walker G, Schmid-Sigron J, Hess CW. Transcu- neurons in Kennedy’s disease: implications for corticomoto- taneous magnetic and electrical stimulation over the cervical neuronal PSTH studies. Muscle Nerve 1999;22:299–306. spine: excitation of plexus roots-rather than spinal roots. 134. Weber M, Eisen A, Nakajima M. Corticomotoneuronal activ- Electroencephalogr Clin Neurophysiol Suppl 1991;43: ity in ALS: changes in the peristimulus time histogram over 369–384. time. Clin Neurophysiol 2000;111:169–177. 117. Schriefer TN, Hess CW, Mills KR, Murray NM. Central mo- 135. Weber M, Eisen A, Stewart HG, Andersen PM, Hirota N. The tor conduction studies in motor neurone disease using mag- physiological basis of conduction slowing in ALS patients netic brain stimulation. Electroencephalogr Clin Neuro- homozygous for the D90A CuZn-SOD mutation. Muscle physiol 1989;74:431–437. Nerve 2001;24:89–97. 118. Schwenkreis P, Vorgerd M, Malin JP, Tegenthoff M. Assess- 136. Werhahn KJ, Lieber J, Classen J, Noachtar S. Motor cortex ment of postexcitatory inhibition in patients with focal dys- excitability in patients with focal epilepsy. Epilepsy Res 2000; tonia. Acta Neurol Scand 1999;100:260–264. 41:179–189. 119. Siebner HR, Tormos JM, Ceballos-Baumann AO, Auer C, 137. Wilson SA, Thickbroom GW, Mastaglia FL. An investigation Catala MD, Conrad B, Pascual-Leone A. Low-frequency re- of the late excitatory potential in the hand following mag- petitive transcranial magnetic stimulation of the motor cor- netic stimulation of the motor cortex. Electroencephalogr tex in writer’s cramp. Neurology 1999;52:529–537. Clin Neurophysiol 1995;97:55–62. 120. Similowski T, Mehiri S, Duguet A, Attali V, Straus C, De- 138. Young RE, Morgan OS, Forster A. Motor pathway analysis in renne JP. Comparison of magnetic and electrical phrenic HAM/TSP using magnetic stimulation and F-waves. Can J nerve stimulation in assessment of phrenic nerve conduction Neurol Sci 1998;25:48–54. time. J Appl Physiol 1997;82:1190–1199. 139. Ziemann U, Steinhoff BJ, Tergau F, Paulus W. Transcranial 121. Stedman A, Davey NJ, Ellaway PH. Facilitation of human first magnetic stimulation: its current role in epilepsy research. dorsal interosseous muscle responses to transcranial mag- Epilepsy Res 1998;30:11–30. netic stimulation during voluntary contraction of the con- 140. Zifko U, Remtulla H, Power K, Harker L, Bolton CF. Trans- tralateral homonymous muscle. Muscle Nerve 1998;21: cortical and cervical magnetic stimulation with recording of 1033–1039. the diaphragm. Muscle Nerve 1996;19:614–620.

Magnetic Stimulation MUSCLE & NERVE February 2002 175