The wave peak of interest is called P100. In healthy subjects it is a positive deflection 100 ms poststimulus (Figure 31.1). In the clinical example shown, taken from a patient with a presumptive diagnosis of multiple sclerosis, the normal P100 wave from the right-eye test indicated that both optic tracts and both optic radiations were clear. The P100 wave from the left eye was both delayed and of reduced amplitude, suggesting presence of one or more plaques of myelin degeneration in the left optic nerve. (Note: On screen and in printouts, it is now customary for the waveforms to be ‘flipped’, with positive responses registering as upward deflections.)

Figure 31.1 Visual evoked potentials. The patient’s right eye has been tested and is now shielded. The left eye is fixated on the spot in the center of the checkerboard during pattern reversal episodes. The pattern from the right eye is normal, showing a positive deflection at 100 ms poststimulus. In the recording from the left eye the P100 is both delayed and reduced in amplitude. The combined results indicate presence of a lesion in the left eye or left optic nerve.

Conduction defects caused by demyelination are more often expressed in the form of latency delays of the kind shown, than in the form of amplitude abnormalities.

In the absence of any evidence for MS elsewhere, an abnormal P100 from one eye may be caused by an ocular disease such as glaucoma or by compression or ischemia of the optic nerve.

Bilateral abnormal P100 recordings can indicate pathology in one or both optic radiations. In such a situation, it is usual to take recordings from electrode pairs placed a few centimeters to one side of the midline and then a few centimeters to the other side. Should the (say) right optic radiation be at fault, any P100 abnormality is likely to be more pronounced in recordings from that side of the midline.

Brainstem auditory evoked potentials

Remarkably, it is possible to follow the sequence of electrical events in the auditory pathway, step by step, from cochlea to primary auditory cortex. Following placement of temporal scalp recording electrodes, 0.1 ms click sounds are presented at approximately 10 Hz to each ear in turn through conventional audiometric earphones. Click intensity is adjusted to 65–70 decibels above click hearing threshold for the ear being tested. The contralateral ear is ‘masked’ by white noise.

A sequence of seven averaged-out waves (I–VII) constitutes the BAER (brainstem auditory evoked response). They are accounted for in the caption to Figure 31.2.

Figure 31.2 Brainstem auditory evoked potentials.

Sources of evoked potentials: 1 distal cochlear nerve (cochlear hair cells); 2 cochlear nerve (proximal); 3 from cochlear nucleus; 4 from lateral lemniscus; 5 from inferior colliculus (inferior brachium); 6 from medial geniculate body (auditory radiation); 7 primary auditory cortex.

Pathology anywhere along the auditory pathway results in reduction or abolition of the wave above that level. The technique is the most sensitive screening test available for acoustic neuroma. A diagnostic feature here is I–III latency separation. (Latency refers to the time interval between stimulus and response; separation refers to extension of the interval between waves I and III, caused by delay during passage along the affected cochlear nerve during a characteristically reduced amplitude wave II.)

In about 30% of patients who have multiple sclerosis (MS) with no clinical evidence of brainstem lesions, the BAER is abnormal. Most frequent abnormalities are reduced amplitude of wave V and overall slowing of conduction indicated by increased interwave intervals.

Another clinical application of the BAER technique is the assessment of cochlear function in infants under suspicion of congenital deafness.

Assessment of brainstem auditory evoked potentials is also important in the medicolegal domain, to assess veracity of claims of deafness induced by environmental noise in industry.

Evidence for a ‘Where’ auditory pathway

When recording electrodes are specifically deployed over the temporoparietal region, and brief sounds are emitted from loudspeakers placed in the left and right visual fields, a cortical response can be detected over the posterior part of the temporal plane, close to the temporoparietal junction. The right posterior temporal plane gives a stronger response, suggesting a right-sided dominance for auditory as well as visual space analysis.

Somatosensory evoked potentials

Somatosensory evoked potentials are the waveforms recorded at surface landmarks en route from the point of stimulation of a peripheral nerve to the contralateral somatic sensory cortex. The rate and amplitude of impulse conduction provide valuable information about the status of myelinated nerve fibers in both peripheral nerves and central pathways.

The nerve of choice for stimulation in the upper limb is the median at the wrist, in the lower limb the common peroneal at the knee. Repetitive electrical pulses are delivered to the nerve through a surface or needle electrode. The larger myelinated fibers are stimulated. Computer averaging is required to distinguish the stimulated responses from background noise, notably within the CNS. In the example shown in Figure 31.3, impulse traffic along the median nerve is detected by a sequence of active electrodes attached to the skin for the purpose of recording speed and amplitude of nerve conduction in sequential segments. as follows:

1 over the brachial plexus, to assess the median nerve segment extending from wrist to anterior triangle of neck;

2 over the spine of vertebra C2, for the segment 1 plus transit through nerve roots and ipsilateral posterior column (fasciculus cuneatus);

3 over the contralateral scalp at some distance away from the somatic sensory cortex, to ‘pick up’ stimulus traffic ascending the medial lemniscus;

4 directly over the hand area of the somatic sensory cortex, to detect activity in the thalamocortical projection.

In the various peripheral neuropathies mentioned in Chapter 9 the first segment (wrist to brachial plexus) reveals slowing, usually with a reduction of amplitude. The second segment (brachial plexus to nucleus gracilis) may be affected in the first few milliseconds of its time course as a result of posterior nerve root compression by osteophytes in patients with cervical spondylosis. A little later, the curve may be affected by posterior column disease (Ch. 15). Abnormality in the third segment (contralateral medial lemniscus) is found in 9 out of 10 patients suffering from MS in the presence of sensory symptoms, and in 6 out of 10 in the absence of sensory symptoms.

Motor Evoked Potentials

Motor evoked potentials are motor unit action potentials (MUAPs) detected in surface EMG recordings following controlled excitation of the corticospinal tract. The technique was mentioned in Chapter 18 because it revealed that the pyramidal tract pathway to sternomastoid spinal motor neurons is essentially crossed, rather than being ipsilateral as previously thought. The most frequent objective is to determine central motor conduction time along the corticospinal tract. The procedure is both safe and painless. It uses a subtraction approach comparable in principle to that used to determine peripheral nerve conduction times (Figure 31.3).

Figure 31.3 Short latency somatosensory evoked potentials derived from stimulation (St.) of the median nerve at the wrist. The pathway has four segments. 1 is purely PNS. 2 is PNS from brachial plexus to spinal cord and CNS within the cord. 3 and 4 are purely CNS. In the recordings from a patient with multiple sclerosis (MS), trace 1 is normal; trace 2 shows reduced amplitude of the negative (upward) and positive peaks within the posterior column; trace 3 shows slowing as well as reduced amplitude.

(Adapted in part from Adams and Victor 1993.)

The procedure is known as transcranial magnetic stimulation (TMS). Figure 31.4 illustrates the concept in action. Stimulation is by means of a magnet in the form of a circular coil about 10 cm in diameter. To stimulate the pyramidal cells of the corticospinal tract supplying left anterior horn motor neurons, the magnet is hand held a little to the right side of the vertex and the patient maintains the selected limb muscle (biceps brachii in this example) in a state of slight contraction. A few very brief (200 ms) currents are pulsed at an intensity comfortably above the threshold required to elicit a twitch. The patient feels only a small ‘tap’ sensation on the scalp. The procedure is then repeated with the magnet touching the skin of the neck overlying the spine of vertebra C5, again eliciting a ‘tap’ sensation. It is generally agreed that the second pulse depolarizes the axons of anterior nerve roots exiting the vertebral canal.

Figure 31.4 Laboratory estimation of central motor conduction time. The (red) coil over the vertex has delivered a 200 ms current to pyramidal tract neurons serving upper limb spinal motor neurons. A large compound motor action potential (CMAP) is elicited in the biceps. The (green) coil over the cervical spine has delivered a weaker pulse to cervical anterior nerve roots eliciting a smaller CMAP. The difference between the two latencies (stimulus–response intervals) represents the central motor conduction time (CMCT).

The latencies and amplitudes of the compound motor action potentials (CMAPs) are measured. The right biceps can then be activated in the same manner. The same procedure can be performed for the lower limb, the spinal stimulus being delivered in the lumbar region.

In neurophysiology units, central motor conduction time is estimated where there is reason to suspect the presence of plaques of multiple sclerosis in the white matter of brain or spinal cord; and where muscle wasting in the arms and/or legs leads to suspicion that upper as well as lower motor neurons may be degenerating; and in patients where moderate muscle weakness on one side, associated with brisk tendon reflexes, raises suspicion of a stroke.

Motor training

A remarkable degree of plasticity in the healthy motor cortex has been demonstrated by TMS studies. Figure 31.5 represents outcomes of five-finger piano-playing exercises. A small magnetic coil was used over the scalp to locate the modules primarily involved in flexion and extension of the fingers of the right hand. This small scalp area was marked in three sets of volunteers, and the baseline size was measured for each subject. Group A imagined doing the five-finger exercise for 2 hours per day for 5 days; Group B did the exercises for the same periods; Group C did not participate in any way prior to attempting the task once on day 5. As indicated in the figure, merely thinking about performance led to a major increase in the number of modules that activated the fingers when stimulated on days 3 and 5. Group B – the actual performers – showed the greatest increase of participating motor modules. The performance skills on day 5 were substantially better in Group B than Group A, and Group A’s performance was better than that of Group C.

Figure 31.5 Five-finger exercise. As explained in the main text, Group A volunteers imagined performing a daily five-finger piano exercise for the right hand, whereas Group B actually did perform it. Transcranial magnetic stimulation (TMS) showed remarkable enlargement of the area of motor cortex activating the finger flexors and extensors in both groups.

(Based on data in Pascual-Leone et al., 1995.)

There is general agreement that dramatic alterations such as those shown in this group experiment are best explained in terms of unmasking of pre-existing connections, as in the case of rapid expansion of the cortical sensory territory of one thalamocortical projection following experimental inactivation of a neighboring projection. The most likely mechanism of additional pyramidal cell recruitment appears to be one of disinhibition, probably by the premotor cortex, involving activation of sequential pairs of GABAergic neurons in the manner illustrated in Figure 6.2.

In this general context, it has also been shown that performance improvement in weight lifting is optimal when subjects mentally rehearse weight lifting during the days between performing the exercises.

Finally, Box 31.1 includes an experiment in which TMS has been used to assess the supposed usefulness of acupuncture in improving motor performance.

Box 31.1 Acupuncture

‘There is sufficient evidence of acupuncture’s value to expand its use into conventional medicine and to encourage further studies of its physiology and clinical value.’ (NIH Consensus Conference on Acupuncture, 1997, cited by Stux et al., 1997).

Sensation

For relief of pain, fine (0.25 mm) needles are inserted bilaterally through appropriate acupoints, coming to rest among superficial muscle fibers underlying the subcutaneous fat (Figure 31.1.1). The needle is then briefly spun to excite Types II and III sensory nerve fibers in its immediate neighborhood. A subjective sense of numbness or heaviness in the acupoint area is usually reported.

The accepted explanation of the rapid pain relief produced by this form of stimulation is release of enkephalin by (a) spinal antinociception via the segmental reflex mentioned in ‘rubbing the sore spot’ combined with (b) supraspinal antinociception via spinoreticular activation of the PAG–MRN–RST–Enk pathway shown in Figure 24.8. Pain relief is not achieved if the subject has received a prior injection of the opiate antagonist naloxone.

The term acupuncture analgesia refers to low-frequency (1 Hz) electrical stimulation of needles inserted at appropriate acupoints (electroacupuncture). The remarkable result is analgesia so complete as to permit open surgery in alert awake patients. Clearly the ascending reticular activating system is not paralyzed, unlike the case with general anesthesia. Animal experiments indicate that electroacupuncture, in addition to the above-mentioned effects, produces co-release of β-endorphin from the arcuate nucleus of the hypothalamus and ACTH from the adenohypophysis.

With availability of fMRI as a research tool, attention is being focused on the effects of acupuncture on higher-level sensory functions. Figure 31.1.2 is composite, reproducing the remarkable results of two quite separate experiments. In each volunteer, needling was performed bilaterally, one acupoint being traditional for relief of disorders of vision, the other for disorders of hearing. Areas of increased cortical blood flow closely resemble those associated with retinal/cochlear activation. Surprisingly, the volunteers did not report visual/auditory hallucinations. And an obvious question persists concerning the mysteriously specific anatomical pathways from the acupoints to the appropriate areas of cortex.