Introduction

The primary concerns of clinical neurophysiology laboratories are twofold: assessment of the functional state of the peripheral nervous system (PNS); and assessment of cerebral cortical function. PNS assessment entails the use of nerve conduction studies by stimulation of selected peripheral nerves while recording the waveforms of their response, and of electromyography by recording the waveforms generated by selected muscles during voluntary contraction. The combination of nerve conduction studies and electromyography is referred to as electrodiagnostic examination.

Nerve Conduction Studies

Nerve conduction studies are routinely employed as part of the clinical examination of suspected disorders of the peripheral nervous system. Through stimulation of nerves allied to recording of muscle fiber depolarizations, it is possible to determine whether the disorder involves the nerve, neuromuscular junction or muscle; also whether it is a focal or diffuse process involving sensory and/or motor axons and whether it is primarily affecting myelin or axons.

Nerve conduction in the upper limb

The role model for detection of distributed (as distinct from focal) disorders within the peripheral neuromuscular system in general is the median nerve. The median, a mixed motor and sensory nerve, has three key advantages for electrophysiological studies of a general nature:

1 It is readily accessible for stimulation and/or recording at elbow and wrist.

2 For motor nerve conduction studies the abductor pollicis brevis, supplied by the median, is readily available for either surface or needle elecromyography;

3 For sensory nerve conduction studies the skin of the index finger is ideal for recording action potentials traveling antidromically following median stimulation at the elbow or wrist. (As noted in Ch. 11, ‘antidromic’ means ‘running against’ the normal (orthodromic) direction of impulse conduction.)

A typical stimulating electrode is one with an anode and a cathode in the form of two blunt prongs which are applied to the skin surface overlying the nerve. In Figure 12.1 it has been placed over the median nerve at the wrist (just lateral to the cordlike palmaris longus tendon). The cathode is placed nearer to the recording site than the anode in order to prevent any conduction block by the anode. When sufficient current is passed from cathode to anode, transmembrane ionic movements initiate impulse propagation in both directions along the nerve. Large myelinated nerve fibers lying nearest the cathode are the first to become depolarized; these include the Aα diameter axons of anterior horn motor neurons. A pulse of 20–40 mA with a duration of 0.1 ms is usually sufficient to activate all motor units in abductor pollicis brevis.

Figure 12.1 Basic setup for recording compound motor action potentials (CMAPs). SA, stimulus artifact. The stimulating electrode (S) has been placed over the median nerve. The active electrode (Act) has been placed over the abductor pollicis brevis muscle and the reference (Ref) electrode has been placed distally.

Recording

An active surface recorder, in the form of a disk in this situation, is placed over the midregion of the muscle where the motor end plates are concentrated, i.e. the motor point. A second, reference electrode, is placed over a neutral site a short distance away. The amplifier used to magnify evoked motor responses is designed to record the potential differences between the two sites. The setup is arranged so that if the active electrode records a more negative response this will take the form of an upward deflection on the monitor.

At low level of stimulation, the only on-screen change in the tracing will be a small stimulus artifact on an otherwise flat tracing. As the current increases, small compound motor action potentials appear. These are produced by activation of large myelinated axons close to the stimulator; the depolarization wave traveling along each will in turn depolarize all of the muscle fibers in the territory of that axon. In the case of the intrinsic muscles of the hand, including abductor pollicis brevis, each motor unit has an innervation ratio of two or three hundred muscle fibers per motor neuron. In large muscles not specialized for fine movements (e.g., deltoid, gastrocnemius) the minimum deflection on the monitor will be several times larger, for two reasons: their motor innervation ratio is 1/1000 or more, and their larger muscle fibers generate action potentials of greater amplitude.)

It should be emphasized that the onscreen waveform is not produced by the contraction process itself, but by the extracellular potentials generated by depolarization of the muscle membranes and filtered through the tissues and skin. However, while this distinction needs to be remembered, most disorders of muscle will also affect the surface membrane depolarization and hence lead to abnormalities of the waveform morphology.

Increasing the applied voltage activates additional motor units until all are activated by each pulse. The required stimulus is called maximal. For good measure, the final stimulus is often supramaximal at 5–10% above maximal. The final waveform observed constitutes the compound motor action potential, or CMAP. It is produced by summation of the individual muscle fiber potentials (Figure 12.2).

Figure 12.2 Summation of CMAPs. Motor units are simply represented by interdigitating pairs of muscle fibers. Moderate (1), medium (2) and maximal (3) stimulation yield progressively larger waveforms on screen, despite being physiologically separate phenomena.

Routine measurements of the final CMAP are shown in Figure 12.3. They include the latency (time interval) between stimulus and depolarization onset, and the amplitude and duration of the negative phase of the waveform. (The final, positive phase is produced by inward ion movement during collective repolarization of the muscle fibers.)

Figure 12.3 Routine CMAP measurements.

Motor nerve conduction velocity (MNCV)

The setup required to determine motor nerve conduction velocity for the median nerve is straightforward, as shown in Figure 12.4. Here the nerve has first been activated at the wrist (S1) to generate and store a ‘wrist to muscle’ velocity record. The stimulator has then been placed over the median nerve at the elbow (S2) to provide an ‘elbow to muscle’ record. Speed being the product of distance over time, the elbow-to-wrist conduction velocity is given by subtracting one value from the other, as illustrated by the case example.

Figure 12.4 Calculation of motor nerve conduction velocity. The same stimulator is used twice: S1 means 1st stimulus and S2 is 2nd stimulus. The double arrows represent the two length measurements. The time baseline is not included. At bottom is an example yielding a normal conduction velocity.

Second choice

It may be considered wise to perform a confirmatory MNCV on another nerve. The ulnar is the standard second choice: S1 is performed over the nerve at the wrist just lateral to flexor carpi ulnaris, S2 where the nerve emerges from behind the medial epicondyle. The active recorder is applied over the hypothenar muscles at the medial margin of the palm.

Sensory nerve conduction

For studies of sensory nerve conduction velocity (SNCV), the median is again the nerve of choice (Figure 12.5). Again it is large myelinated nerve fibers that will be stimulated, and the site and manner of stimulation at elbow and wrist will be the same. On this occasion, however, we are selectively recording antidromic stimulation of cutaneous sensory fibers – specifically, of the digital branches of the median nerve to the skin of the index finger, which is wearing an active recorder in the form of a ring.

Figure 12.5 Calculation of sensory nerve conduction velocity (SNCV). Digital branches of the median nerve are represented. The basic principles for the calculation are the same as for MNCV studies. For the asterisks see main text.

The prime function of the myelinated nerve fibers to be sampled by the ring recorder are those supplying the highly sensitive and discriminatory skin of the finger pad, described in Chapter 11. The largest, serving Meisssner and Pacinian corpuscles and Merkel cell–neurite complexes, are known to normally conduct at a speed of 60–100 m/s and the finest, serving mechanical nociceptors, at 10–30 m/s. This tenfold variation is in marked contrast to that of the relatively uniform fiber size of the stem axons supplying the small motor units of the abductor muscle and conducting at 45–55 m/s. One consequence is that, when stimulating sensory nerves at increasing distances from the recording site, a change in the waveform shape is normally noted. In the figure, the asterisks are intended to highlight the difference in the shape of the waveforms of the two compound sensory nerve action potentials (CSNAPs). Two factors are involved:

1 Physiologic temporal dispersion. As runners in a race become progressively separated over distance, the fastest impulse conductors take the lead and the slowest trail behind, with consequent elongation of the CSNAP profile over the longer test distance, caused by this temporal dispersion (scattering over time).

2 Phase cancellation. The later waveform is also flatter. This is explained in part by the phenomenon whereby positive and negative phases of adjacent waveforms tend to cancel each other out. It should be emphasized that this phase cancellation is not a physiological event: the waves themselves are not affected by the ‘eavesdropper’ wrapped around the finger close to the finishing line. An additional factor is the diminishing amount of phase summation being recorded with increasing separation of action potentials.

Sensory nerve conduction velocity (SNCV)

The basic modes of operation and calculation are the same as shown for the MNCV study. A case example is included in Figure 12.5 which demonstrates phase cancellation from temporal dispersion.

Second choice

Again, the ulnar nerve is the standard second choice. Ulnar nerve stimulation is performed at wrist and elbow as before, with a ring recorder is slipped onto the little finger.

Nerve conduction in the lower limb

The lower limb muscle most frequently sampled for motor nerve conduction velocity is extensor digitorum brevis on the dorsum of the foot (Figure 12.6). The deep peroneal nerve is stimulated first in front of the ankle and then at the level of the neck of the fibula. At times, tibialis anterior is also sampled; in this case the nerve is stimulated first at the neck of the fibula and then at the lateral edge of the popliteal fossa next to the biceps femoris tendon.