Motor Conduction Studies

Motor conduction studies are technically less demanding than sensory and mixed nerve studies; thus; they usually are performed first. Performing the motor studies first also has other major advantages. It is not uncommon for the sensory responses to be very low in amplitude or absent in many neuropathies. Performing the motor studies first allows one to know where the nerve runs, where it should be stimulated, and how much current is needed, and also gives some information about whether the nerve is normal or abnormal. On the other hand, if the sensory study is done before the motor study, one might spend a lot of unnecessary time stimulating and trying to record a sensory response which is not present. For example, imagine a patient with a moderately severe median neuropathy at the wrist who is sent for an EDX evaluation. If the median motor study is performed first, the correct stimulation site can be confirmed, the amount of current needed to stimulate the median nerve will be known, and one will also know that the median nerve is abnormal, before doing the median sensory study. Then, when performing the median sensory study, one is confident of where to stimulate the nerve and how much current is needed. In this case, if no sensory response is present, one can have a high degree of certainty that the response is truly absent, and move along to the next nerve to be studied. However, if the sensory conduction study is done first, and is absent, it will not be as obvious if the absent response is due to a technical problem, or is truly absent. One can waste a lot of time unnecessarily trying to figure this out. Do the motor conduction study first; your study will be more efficient, and the patient will tolerate the study much better.

Motor responses typically are in the range of several millivolts (mV), as opposed to sensory and mixed nerve responses, which are in the microvolt (µV) range. Thus, motor responses are less affected by electrical noise and other technical factors. For motor conduction studies, the gain usually is set at 2 to 5 mV per division. Recording electrodes are placed over the muscle of interest. In general, the belly–tendon montage is used. The active recording electrode (also known as G1) is placed on the center of the muscle belly (over the motor endplate), and the reference electrode (also known as G2) is placed distally, over the tendon to the muscle (Figure 3–1). The designations G1 and G2 remain in the EMG vernacular, referring to a time when electrodes were attached to grids (hence the G) of an oscilloscope. The stimulator then is placed over the nerve that supplies the muscle, with the cathode placed closest to the recording electrode. It is helpful to remember “black to black,” indicating that the black electrode of the stimulator (the cathode) should be facing the black recording electrode (the active recording electrode). For motor studies, the duration of the electrical pulse usually is set to 200 ms. Most normal nerves require a current in the range of 20 to 50 mA to achieve supramaximal stimulation. As current is slowly increased from a baseline 0 mA, usually by 5 to 10 mA increments, more of the underlying nerve fibers are brought to action potential, and subsequently more muscle fiber action potentials are generated. The recorded potential, known as the compound muscle action potential (CMAP), represents the summation of all underlying individual muscle fiber action potentials. When the current is increased to the point that the CMAP no longer increases in size, one presumes that all nerve fibers have been excited and that supramaximal stimulation has been achieved. The current is then increased by another 20% to ensure supramaximal stimulation.

FIGURE 3–1 Motor conduction study setup.

Median motor study, recording the abductor pollicis brevis muscle, stimulating the median nerve at the wrist. In motor studies, the “belly–tendon” method is used for recording. The active recording electrode (G1) is placed on the center of the muscle, with the reference electrode (G2) placed distally over the tendon.

The CMAP is a biphasic potential with an initial negativity, or upward deflection from the baseline, if the recording electrodes have been properly placed with G1 over the motor endplate. For each stimulation site, the latency, amplitude, duration, and area of the CMAP are measured (Figure 3–2). A motor conduction velocity can be calculated after two sites, one distal and one proximal, have been stimulated.

FIGURE 3–2 Compound muscle action potential (CMAP).

The CMAP represents the summation of all the underlying muscle fiber action potentials. With recording electrodes properly placed, the CMAP is a biphasic potential with an initial negative deflection. Latency is the time from the stimulus to the initial negative deflection from baseline. Amplitude is most commonly measured from baseline to negative peak but also can be measured from peak to peak. Duration is measured from the initial deflection from baseline to the first baseline crossing (i.e., negative peak duration). In addition, negative CMAP area (i.e., the area above the baseline) is calculated by most modern computerized electromyographic machines. Latency reflects only the fastest conducting motor fibers. All fibers contribute to amplitude and area. Duration is primarily a measure of synchrony.

Latency

The latency is the time from the stimulus to the initial CMAP deflection from baseline. Latency represents three separate processes: (1) the nerve conduction time from the stimulus site to the neuromuscular junction (NMJ), (2) the time delay across the NMJ, and (3) the depolarization time across the muscle. Latency measurements usually are made in milliseconds (ms), and reflect only the fastest conducting motor fibers.

Amplitude

CMAP amplitude is most commonly measured from baseline to the negative peak and less commonly from the first negative peak to the next positive peak. CMAP amplitude reflects the number of muscle fibers that depolarize. Although low CMAP amplitudes most often result from loss of axons (as in a typical axonal neuropathy), other causes of a low CMAP amplitude include conduction block from demyelination located between the stimulation site and the recorded muscle, as well as some NMJ disorders and myopathies.

Area

CMAP area also is conventionally measured as the area above the baseline to the negative peak. Although the area cannot be determined manually, the calculation is readily performed by most modern computerized EMG machines. Negative peak CMAP area is another measure reflecting the number of muscle fibers that depolarize. Differences in CMAP area between distal and proximal stimulation sites take on special significance in the determination of conduction block from a demyelinating lesion (see section on Conduction Block).

Duration

CMAP duration usually is measured from the initial deflection from baseline to the first baseline crossing (i.e., negative peak duration), but it also can be measured from the initial to the terminal deflection back to baseline. The former is preferred as a measure of CMAP duration because when CMAP duration is measured from the initial to terminal deflection back to baseline, the terminal CMAP returns to baseline very slowly and can be difficult to mark precisely. Duration is primarily a measure of synchrony (i.e., the extent to which each of the individual muscle fibers fire at the same time). Duration characteristically increases in conditions that result in slowing of some motor fibers but not others (e.g., in a demyelinating lesion).

Conduction Velocity

Motor conduction velocity is a measure of the speed of the fastest conducting motor axons in the nerve being studied, which is calculated by dividing the distance traveled by the nerve conduction time. However, motor conduction velocity cannot be calculated by performing a single stimulation. The distal motor latency is more than simply a conduction time along the motor axon; it includes not only (A) the conduction time along the distal motor axon to the NMJ, but also (B) the NMJ transmission time and (C) the muscle depolarization time. Therefore, to calculate a true motor conduction velocity, without including NMJ transmission and muscle depolarization times, two stimulation sites must be used, one distal and one proximal.

When the nerve is stimulated proximally, the resulting CMAP area, amplitude, and duration are, in general, similar to those of the distal stimulation waveform. The only major difference between CMAPs produced by proximal and distal stimulations is the latency. The proximal latency is longer than the distal latency, reflecting the longer time and distance needed for the action potential to travel. The proximal motor latency reflects four separate times, as opposed to the three components reflected in the distal motor latency measurement. In addition to (A) the nerve conduction time between the distal site and the NMJ, (B) the NMJ transmission time, and (C) the muscle depolarization time, the proximal motor latency also includes (D) the nerve conduction time between the proximal and distal stimulation sites (Figure 3–3). Therefore, if the distal motor latency (containing components A + B + C) is subtracted from the proximal motor latency (containing components A + B + C + D), the first three components will cancel out. This leaves only component D, the nerve conduction time between the proximal and distal stimulation sites, without the distal nerve conduction, NMJ transmission, and muscle depolarization times. The distance between these two sites can be approximated by measuring the surface distance with a tape measure. A conduction velocity then can be calculated along this segment: (distance between the proximal and distal stimulation sites) divided by (proximal latency − distal latency). Conduction velocities usually are measured in meters per second (m/s).

FIGURE 3–3 Motor conduction velocity (CV) calculation.

Top: Median motor study, recording abductor pollicis brevis, stimulating wrist and elbow. DL, distal motor latency; PL, proximal motor latency. The only difference between distal and proximal stimulations is the latency, with PL being longer than DL. Bottom: DL represents three separate times: the nerve conduction time from the distal stimulation site to the neuromuscular junction (NMJ) (A), the NMJ transmission time (B), and the muscle depolarization time (C). Accordingly, DL cannot be used alone to calculate a motor conduction velocity. Two stimulations are necessary. PL includes the nerve conduction time from the distal stimulation site to the neuromuscular junction (A), the NMJ transmission time (B), and the muscle depolarization time (C), as well as the nerve conduction time between the proximal and distal stimulation sites (D). If DL (A + B + C) is subtracted from PL (A + B + C + D), only the nerve conduction time between the distal and proximal stimulation sites (D) remains. The distance between those two sites can be measured, and a conduction velocity can be calculated (distance/time). Conduction velocity reflects only the fastest conducting fibers in the nerve being studied.

It is essential to note that both latency and conduction velocity reflect only the fastest conducting fibers in the nerve being studied. By definition, conduction along these fibers arrives first and thus it is these fibers that are the ones measured. The many other slower conducting fibers participate in the CMAP area and amplitude but are not reflected in either the latency or conduction velocity measurements.

Sensory Conduction Studies

In contrast to motor conduction studies, in which the CMAP reflects conduction along motor nerve, NMJ, and muscle fibers, in sensory conduction studies only nerve fibers are assessed. Because most sensory responses are very small (usually in the range of 1 to 50 µV), technical factors and electrical noise assume more importance. For sensory conduction studies, the gain usually is set at 10 to 20 µV per division. A pair of recording electrodes (G1 and G2) are placed in line over the nerve being studied, at an interelectrode distance of 2.5 to 4 cm, with the active electrode (G1) placed closest to the stimulator. Recording ring electrodes are conventionally used to test the sensory nerves in the fingers (Figure 3–4). For sensory studies, an electrical pulse of either 100 or 200 ms in duration is used, and most normal sensory nerves require a current in the range of 5 to 30 mA to achieve supramaximal stimulation. This is less current than what is usually required for motor conduction studies. Thus, sensory fibers usually have a lower threshold to stimulation than do motor fibers. This can easily be demonstrated on yourself; when slowly increasing the stimulus intensity, you will feel the paresthesias (sensory) before you feel or see the muscle start to twitch (motor). As in motor studies, the current is slowly increased from a baseline of 0 mA, usually in 3 to 5 mA increments, until the recorded sensory potential is maximized. This potential, the sensory nerve action potential (SNAP), is a compound potential that represents the summation of all the individual sensory fiber action potentials. SNAPs usually are biphasic or triphasic potentials. For each stimulation site, the onset latency, peak latency, duration, and amplitude are measured (Figure 3–5). Unlike motor studies, a sensory conduction velocity can be calculated with one stimulation site alone, by taking the measured distance between the stimulator and active recording electrode and dividing by the onset latency. No NMJ or muscle time needs to be subtracted out by using two stimulation sites.

FIGURE 3–4 Sensory conduction study setup.

Median sensory study, antidromic technique. Ring electrodes are placed over the index finger, 3 to 4 cm apart. The active recording electrode (G1) is placed more proximally, closest to the stimulator. Although the entire median nerve is stimulated at the wrist, only the cutaneous sensory fibers are recorded over the finger.

FIGURE 3–5 Sensory nerve action potential (SNAP).

The SNAP represents the summation of all the underlying sensory fiber action potentials. The SNAP usually is biphasic or triphasic in configuration. Onset latency is measured from the stimulus to the initial negative deflection for biphasic SNAPs (as in the waveform here) or to the initial positive peak for triphasic SNAPs. Onset latency represents nerve conduction time from the stimulus site to the recording electrodes for the largest cutaneous sensory fibers in the nerve being studied. Peak latency is measured at the midpoint of the first negative peak. Amplitude most commonly is measured from baseline to negative peak but also can be measured from peak to peak. Duration is measured from the initial deflection from baseline to the first baseline crossing (i.e., negative peak duration). Only one stimulation site is required to calculate a sensory conduction velocity, as sensory onset latency represents only nerve conduction time.

Onset Latency

The onset latency is the time from the stimulus to the initial negative deflection from baseline for biphasic SNAPs or to the initial positive peak for triphasic SNAPs. Sensory onset latency represents nerve conduction time from the stimulus site to the recording electrodes for the largest cutaneous sensory fibers in the nerve being studied.

Peak Latency

The peak latency is measured at the midpoint of the first negative peak. Although the population of sensory fibers represented by the peak latency is not known (in contrast to the onset latency, which represents the fastest conducting fibers in the nerve being studied), measurement of peak latency has several advantages. The peak latency can be ascertained in a straightforward manner; there is practically no interindividual variation in its determination. In contrast, the onset latency can be obscured by noise or by the stimulus artifact, making it difficult to determine precisely. In addition, for some potentials, especially small ones, it may be difficult to determine the precise point of deflection from baseline (Figure 3–6). These problems do not occur in marking the peak latency. Normal values exist for peak latencies for the most commonly performed sensory studies stimulated at a standard distance. Note that the peak latency cannot be used to calculate a conduction velocity.

FIGURE 3–6 Sensory nerve action potential (SNAP) onset and peak latencies.

Onset and peak latency measurements each have their own advantages and disadvantages. Onset latency represents the fastest conducting fibers and can be used to calculate a conduction velocity. However, for many potentials, especially small ones, it is difficult to precisely place the latency marker on the initial deflection from baseline (blue arrows: possible onset latencies). Marking the peak latency is straightforward, with nearly no inter-examiner variation. However, the population of fibers represented by peak latency is unknown; it cannot be used to calculate a conduction velocity.

Amplitude

The SNAP amplitude is most commonly measured from baseline to negative peak, but it can also be measured from the first negative peak to the next positive peak. The SNAP amplitude reflects the sum of all the individual sensory fibers that depolarize. Low SNAP amplitudes indicate a definite disorder of peripheral nerve.

Duration

FIGURE 3–7 Compound muscle action potential (CMAP) and sensory nerve action potential (SNAP) comparison.

CMAPs (top) and SNAPs (bottom) both are compound potentials but are quite different in terms of size and duration. CMAP amplitude usually is measured in millivolts, whereas SNAPs are small potentials measured in the microvolt range (note different gains between the traces). CMAP negative peak duration usually is 5 to 6 ms, whereas SNAP negative peak duration is much shorter, typically 1 to 2 ms. When both sensory and motor fibers are stimulated (such as when performing antidromic sensory or mixed studies), these differences (especially duration) usually allow an unknown potential to be recognized as either a nerve or muscle potential.

Conduction Velocity

Unlike the calculation of a motor conduction velocity, which requires two stimulation sites, sensory conduction velocity can be determined with one stimulation, simply by dividing the distance traveled by the onset latency. Essentially, distal conduction velocity and onset latency are the same measurement; they differ only by a multiplication factor (i.e., the distance). Sensory conduction velocity represents the speed of the fastest, myelinated cutaneous sensory fibers in the nerve being studied.

Sensory conduction velocity along proximal segments of nerve can be determined by performing proximal stimulation and calculating the conduction velocity between proximal and distal sites, in a manner similar to the calculation for motor conduction velocity: (distance between the proximal and distal stimulation sites) divided by (proximal latency − distal latency). However, proximal sensory studies result in smaller amplitude potentials and often are more difficult to perform, even in normal subjects, because of the normal processes of phase cancellation and temporal dispersion (see later). Note that one can also determine the sensory conduction velocity from the proximal site to the recording electrode by simply dividing the total distance traveled by the proximal onset latency.

Special Considerations in Sensory Conduction Studies: Antidromic versus Orthodromic Recording

When a nerve is depolarized, conduction occurs equally well in both directions away from the stimulation site. Consequently, sensory conduction studies may be performed using either antidromic (stimulating toward the sensory receptor) or orthodromic (stimulating away from the sensory receptor) techniques. For instance, when studying median sensory fibers to the index finger, one can stimulate the median nerve at the wrist and record the potential with ring electrodes over the index finger (antidromic study). Conversely, the same ring electrodes can be used for stimulation, and the potential recorded over the median nerve at the wrist (orthodromic study). Latencies and conduction velocities should be identical with either method (Figure 3–8), although the amplitude generally is higher in antidromically conducted potentials.

FIGURE 3–8 Antidromic and orthodromic sensory studies.

Median sensory nerve action potential (SNAPs). Top trace: Antidromic study, stimulating wrist, recording index finger. Bottom trace: Orthodromic study, stimulating index finger, recording wrist. Latencies and conduction velocities are identical for both. The antidromic method has the advantage of a higher-amplitude SNAP but is followed by a large volume-conducted motor potential. If the SNAP is absent in an antidromic study, care must be taken not to confuse the volume-conducted motor potential as the sensory potential. Note the difference in duration between SNAP and CMAP, which helps discriminate between the SNAP and the volume-conducted motor potential that follows.

In general, the antidromic technique is superior for several reasons, but each method has its advantages and disadvantages. Most important, the amplitude is higher with antidromic than with orthodromic recordings, which makes it easier to identify the potential. The SNAP amplitude is directly proportional to the proximity of the recording electrode to the underlying nerve. For most antidromically conducted potentials, the recording electrodes are closer to the nerve. For example, in the antidromically conducted median sensory response, the recording ring electrodes are placed on the finger, very close to the underlying digital nerves just beneath the skin from which the potential is recorded. When the montage is reversed for orthodromic recording, there is more tissue (e.g., the transverse carpal ligament and other connective tissues) at the wrist separating the nerve from the recording electrodes. This results in attenuation of the recorded sensory response, resulting in a much lower amplitude. The higher SNAP amplitude obtained with antidromic recordings is the major advantage of using this method. The antidromic technique is especially helpful when recording very small potentials, which often occur in pathologic conditions. Furthermore, because the antidromic potential generally is larger than the orthodromic potential, it is less subject to noise or other artifacts.

However, the antidromic method has some disadvantages (Figure 3–9). Since the entire nerve is often stimulated, including the motor fibers, this frequently results in the SNAP being followed by a volume-conducted motor potential. It usually is not difficult to differentiate between the two, because the SNAP latency typically occurs earlier than the volume-conducted motor potential. However, problems occur if the two potentials have a similar latency or, more importantly, if the sensory potential is absent. When the latter occurs, one can mistake the first component of the volume conducted motor potential for the SNAP, where none truly exists. It is in this situation that measuring the duration of the potential can be helpful in distinguishing a sensory from a motor potential. If one is still not sure, performing an orthodromic study will settle the issue, as no volume conducted motor response will occur with an orthodromic study. In this case, the antidromic and orthodromic potentials should have the same onset latency.