magnetic brain stimulation refers to a technique for electrical stimulation of part of the brain, through the intact skull, by the principle of electromagnetic induction.

The fact that the brain can be excited by electric current was first demonstrated by David Ferrier in England, and by Fritsch and Hitzig in Germany in the early 1880s. They removed part of the skull of anaesthetized dogs and monkeys, in order to expose the surface of the cerebral cortex, and then applied small electrical shocks to what is now termed the motor cortex, which is a strip of cortex running down the side of the cerebral hemisphere. Stimulation produced twitches in muscles on the opposite side of the body — in the legs if the electric shock was given near the midline of the brain, in the forelimbs if the stimulus was more lateral.

Mapping the brain in humans

Neurosurgeons, operating on human patients to remove tumours or areas of abnormal brain tissue giving rise to epileptic seizures, are obviously anxious to avoid unnecessary injury to particularly important regions of the brain, such as the motor cortex, damage of which causes permanent paralysis. In the 1930s, the Canadian surgeon, Wilder Penfield, exploited the technique of electrical stimulation of the exposed cortex to map the location of the motor cortex. The operation could then be conducted in such a way as to avoid this area if at all possible. Since the brain itself has no direct pain sensation, this kind of mapping was usually carried out under local anaesthesia, so the patient could also report any sensations that were produced. This enabled Penfield to explore many of the sensory areas of the cortex. For instance, he found that stimulation of the strip of cortex directly behind the motor cortex (the post-central gyrus) produced curious sensations (paraesthesiae) in skin and deep tissues of parts of the body, forming a ‘map’ of the body lined up with the neighbouring motor map. This was the first proof that the post-central gyrus is the location of the primary somatic sensory cortex in humans. In the same way, Penfield discovered that electrical stimulation of the visual or auditory areas of the cortex gave rise to corresponding sensory hallucinations, and stimulation in the lower part of the temporal lobe could elicit whole episodes of experience, involving many different sorts of sensory experience.

The barrier of the skull

Being able to define the motor cortex and other regions before surgery is extremely valuable. But, despite the extensive use of electrical brain stimulation in experiments on anaesthetized animals and during human neurosurgery, it was assumed that the overlying skull had to be removed in order to apply current to the brain. Then electroconvulsive therapy (ECT), developed in the 1930s for treatment of depression, showed that the brain could be stimulated by applying electric current through the skull. However, this technique is hardly amenable for application to people except for therapeutic reasons: the stimulus is not focal, it provokes seizures, and is nowadays always performed under general anaesthesia.

Apart from a very small number of largely unsuccessful attempts to refine the parameters of electrical stimulation and to make it more focal, the concept of transcranial stimulation was virtually forgotten for the next fifty years. The problem is that the high resistance of the scalp and skull prevents all but a fraction of the current applied to electrodes on the scalp from reaching the brain. Most of the current flows along the skin, causing local pain and contraction of scalp muscle. The solution came in two stages. First of all, in 1980, Merton and Morton found that it was possible to use a single high-voltage electrical pulse, applied through small scalp electrodes, to stimulate the motor cortex in conscious subjects. Because the stimulus was very brief, the pain was reduced. This technique rapidly became a standard way of testing the integrity of the nerve pathway from the motor area of the cerebral cortex to the motoneurons in the spinal cord. Damage to the pathway was indicated if the size of the resulting muscle twitch was smaller than usual, or if the delay between stimulation and contraction was unusually long. Nevertheless, because the stimulus also caused a large twitch in the scalp muscle and considerable discomfort, the range of applications was limited.

In 1985, a new development occurred that solved this problem. Barker and colleagues in Sheffield had developed a magnetic stimulator that they had tested successfully by using it to stimulate peripheral nerves through the skin. It relied on the principle of electromagnetic induction, and consisted of an insulated coil of copper wire connected to a large electrical capacitance. The capacitance was charged up to a high voltage and then short-circuited through the coil. A current of several thousand amps flowed for about 1 millisecond and generated a magnetic field at right angles to the coil. Like the current in the coil, this field changed rapidly, being zero before the pulse was given and reaching a strength of 1–2 Teslas (several million times the earth's magnetic field) after 0.5 milliseconds or less. According to Faraday's laws of electromagnetic induction, a changing magnetic field can induce electric current to flow in any conductive structure nearby. The saline environment of the body is no exception to this, so the magnetic field induced an electrical current inside the body, and it was this that stimulated the peripheral nerves. In effect, the magnetic field acted as a ‘carrier’ of an electrical stimulus from the external coil into the body tissues.

Because magnetic fields can penetrate bone as readily as any other body tissue, Barker and colleagues thought that their stimulator might be an ideal way of overcoming the problem of scalp pain with transcranial electric stimulation. They therefore brought the first version of their magnetic stimulator to London in 1985 so that Merton could test it out on his own head. On the very first occasion, the magnetic stimulator activated the motor cortex through his scalp and caused a twitch of the hand on the opposite side of his body. Better still, the scalp sensation was minimal, because there was no impedance to the passage of magnetic field into the brain. The electric current induced on the surface of the skin was similar to that induced in the brain, rather than being 10–100 times larger, as with the electrical method.

Applications

The magnetic technique is now the method of choice for all investigations (both clinical and experimental) using transcranial brain stimulation. However, there are two limitations: first, the stimulus spreads out some distance from the coil, so that several square centimetres of tissue may be stimulated at the same time; second, the effectiveness of the stimulus falls off rapidly with distance, so stimulation of deep brain structures is not possible at the present time.

The principal clinical application is to measure conduction in the corticospinal pathway from motor cortex to spinal cord. However, there is now speculation that the technique can be used therapeutically. New stimulators that can deliver repeated stimuli up to 50 times a second are available, and this has given rise to speculation that they could produce persistent changes in the strength of synapses between nerve fibres and nerve cells in the cortex. Such repetitive stimulation has been shown, in extensive studies on animals, to change the amount of neurotransmitter released by the fibre or the number of receptors in the receiving cell's membrane, altering the effectiveness of the synapses — long-term potentiation or depression. The hope is that, by applying repeated stimuli to a cortical region, it may become possible to encourage changes in protein synthesis that affect receptor sensitivity or transmitter release at cortical synapses. Trials are presently being conducted to test the effectiveness of the method in treating patients with depression.

Finally, the ability to stimulate the brain in conscious human subjects is proving to be a remarkable experimental tool that can complement other methods of human brain imaging. For example, PET and fMRI are imaging techniques that provide information about which areas of brain are active in a particular task, but give no information on whether a particular area is essential for performance of the task, or just associated with it. Similarly PET and fMRI provide poor information about the timing of events in the brain. Regions of activity are mapped out, but the precise sequence of events cannot be demonstrated. Both points can be addressed with magnetic stimulation. A single stimulus to any area of brain disrupts function in that area for a tenth of a second or so. Thus if an area is essential to a particular task, a magnetic stimulus will disrupt performance if applied at the appropriate time.

J. C. Rothwell

See also brain; cerebral cortex; imaging techniques; memory.