Peripheral Nerve Testing and Electromyography

Appendix D

George C. Farnbach


Just as bones, joints, tendons, muscle, and nerve must work in concert to produce normal canine gait, so must the orthopaedic surgeon, peripheral neurologist, and neurosurgeon work together to affect reasonable approaches to problems in the canine musculoskeletal system. Often the trauma that produces fractures of long bones also produces trauma to peripheral nerves. Not infrequently, front quarter trauma leads to peripheral nerve damage-brachial root avulsion-without fractures in the affected limb, but orthopaedic intervention (e.g., tendon transplant) is required to produce a favorable outcome. Furthermore, it is sometimes difficult to diagnose subtle or painful gait abnormalities in the dog without a complete evaluation of orthopaedic, neural, and muscular systems. Accordingly, the orthopaedic surgeon should be aware of the techniques, advantages, and limitations of peripheral nerve testing and electromyography (EMG).

Peripheral Nerve Testing


The peripheral spinal nerve is formed by the confluence of the dorsal and the ventral spinal roots near the spinal column. For both the forelimbs and hindlimbs the spinal nerves generally participate in the formation of a nerve plexus from which the nominate peripheral nerves emerge. At this point the peripheral nerve is a "mixed" nerve, that is, it contains many types of myelinated and unmyelinated axons that are part of both sensory and motor systems. As the nerves proceed to the periphery, they often form branches that are mainly sensory in nature. (A good example of this is the lateral cutaneous radial nerve, which is branched from the radial trunk. ) In contrast to these sensory cutaneous branches, the nerves that proceed to supply specific muscles remain mixed nerves, since they contain a large proportion of fibers that carry sensory information from receptors within the muscle and tendons, as well as the axons of the motorneurons (Fig. D- 1). Because of this anatomical difference, there are fundamental differences between the techniques used to study sensory nerve conduction velocities (SNCV) and motor nerve conduction velocities (MNCV).

FIG. D-1 Schematic diagram of a peripheral nerve with sensory and motor branches Stimulation or recording at site A involves both sensory and motor axons, at site C sensory axons only, and at B both sensory and motor axons, since the motor nerve has many sensory axons.


The techniques for determining SNCV and MNCV can best be understood with reference to Figure D-1, which shows three possible electrode sites that might be used for either stimulating or recording from peripheral nerve. Note that the site marked A is on a main trunk, and that if it were used for stimulation, both sensory and motor axons would be excited by stimulation. At the site marked C, stimulation would excite mainly sensory fibers, since it is near a sensory branch of the main nerve. Stimulation at site B would excite both sensory and motor fibers in spite of the fact that it is close to a motor nerve, since there are many sensory fibers in even the most terminal of motor nerves.


In order to record mainly sensory fibers and thereby estimate SNCV, stimulating a real nerve at a site that corresponds to C and recording at site A would produce a recorded response from mainly sensory axons. To estimate conduction velocity over the region of the nerve from C to A, one could simply divide the distance along the nerve from C to A by the time that it took from the onset of stimulation for the response to appear at A. This is exactly the way in which SNCV is estimated.


Since there is no point along the nerve at which stimulation could be expected to excite only motor nerves, the procedure for estimation of MNCV is a little more complex. In order to ensure that the response being recorded is the result only of motor axon stimulation, the recording is done by monitoring the electrical response of muscle fibers. This is accomplished by placing electrodes near the belly of a muscle innervated by the nerve being studied. In such a situation, the latency from stimulus (site A) to recorded response includes more time than that required for the action potential to go from stimulus site to nerve terminus. It also includes the delay at the neuromuscular junction and any small conductive delays along the muscle. To circumvent this problem, the nerve can be stimulated at a second site (B) and a second latency recorded. This second latency will be shorter than the first because, while it contains the same neuromuscular and muscular delays, the time required for nerve conduction is less owing to the fact that the stimulus is closer to the muscle. In fact, the difference between the two latencies is the amount of time that the action potential took to get from site A to site B after the first stimulus. It is this time-the difference in latencies- that can be used to calculate the MNCV by dividing it into the distance along the nerve between sites A and B. This approach to MNCV estimation is depicted graphically in Figure D-2.

One drawback to the methods for measurement of nerve conduction velocity is that they evaluate only the section of nerve between electrode sites. Another method of evaluating peripheral nerve takes advantage of what is called the "F wave"-a small, delayed response that can be recorded from certain muscles after stimulation of their motor nerve. The F wave is thought to be produced by a small percent of stimulated motor nerve cell bodies that initiate a second action potential in response to the first one produced by stimulation, perhaps via internuncial cells. Thus, the action potential is "reflected off" the cell body located in the ventral gray matter of the spinal cord and travels the length of the axon to produce a second, delayed motor responses Since these F waves involve action potentials that twice travel the span of axon proximal to the stimulation site, they are theoretically valuable in the assessment of the proximal portion of the peripheral nerve, including the ventral root. The use of F waves has not yet found much popularity in veterinary medicine. However, we have found that they are easily observed in normal animals and have noted their absence or marked reduction in dogs with proximal nerve and root diseased

FIG. D-2 Schematic diagram of motor nerve conduction velocity (MNCV) estimation. The nerve is stimulated at two sites, and recording is done from the muscle.

Neuromuscular Junctions

The integrity of neuromuscular junctions can be evaluated by repetitive stimulation of the motor nerve. Generally, diseases that affect the neuromuscular junction, such as tick paralysis, myasthenia gravis, and botulism, are systemic diseases, and therefore any muscle and its motor nerve may be used. It is usually best to use the extensor digitorum brevis (EDB) and the fibular nerve, since stimulation of the nerve at its more peripheral site causes contraction only of the small EDB and thus minimizes motion problems. In order to determine a starting point or "baseline," the magnitude of the muscle action potential is recorded while stimulating the nerve at rates of once per second or less. In normal dogs the EDB produces an amplitude of about 10 mV to 12 mV. In dogs with junctional disease the baseline amplitude is generally decreased. It can also be decreased as a result of either neural or muscular disease. Neuromuscular junction disease manifests itself in changes in amplitude of the muscle potential evoked by repetitive stimulation of the nerve. In myasthenic dogs (postsynaptic lesions) the amplitude decreases by 20% or more at stimulation rates of 5 to 10 stimuli per second. This decrement can be prevented with anticholinesterase drugs (e.g., edrophonium chloride). In clinical reports of humans with either tick paralysis or botulism (presynaptic lesions), the evoked muscle potential increases significantly with repetitive stimulations The increment or decrement upon repetitive stimulation is thought to be a good indicator of whether the disease is postsynaptic or presynaptic in nature.


Signal averaging techniques are not absolutely necessary for the evaluation of SNCV. What is most important is the careful attention to electrode placement in order to minimize noise and artifacts and to ensure a good response. The two nerves that can be studied readily are the superficial radial and the superficial fibular. The locations for electrode placement on these two nerves are shown in Figure D-3. The exact location is less important than the arrangement of sets of electrodes. The recording electrodes (two differential and one ground) must be placed in such a way as to minimize common mode noise and stimulus artifacts. In general, the closer together the two differential electrodes are, the less noise they will record. Electrodes should be placed parallel to each other 1.5 cm apart along the course of the nerve and perpendicular to the presumed course of the nerve. For the sake of uniformity, the recording electrodes should always be the same distance apart. Usually, placing the ground electrode at the level of recording electrode closest to the set of stimulating electrodes but away from the nerve gives the best signal with the smallest stimulus artifact. In fact, the placement of the ground electrode is a matter of trial and error. The stimulating electrodes should be placed as far as possible from the recording electrodes and only 1 cm apart in order to minimize stimulus artifact. The anode should be away from the recording electrodes relative to the cathode. Stimulus strength should be just large enough to produce the maximum recorded response.

In measuring the SNCV, the gain on the oscilloscope should be set to achieve a value of about 20 FV/div. The sweep should be triggered by the stimulus and set around 1 msec/div. The recorded amplitudes vary between approximately 20 uV and 120 FV (peak-to-peak), and the duration of the compound sensory action potential is less than 2 msec. An example taken from a single trial is shown in Figure D-4. In this example, the sweep begins with the onset of the stimulus, hence the latency is measured by the length of the sweep before the response is seen. In this case the latency is 1.7 msec and the interelectrode distance is 92 mm, giving a SNCV of 54 m/sec.

In MNCV studies, the recorded response from the muscle is 100 to 1000 times greater than that recorded from the sensory nerve; consequently, the signal-to-noise considerations are almost negligible. However, care still must be taken in electrode placement for the sake of reproducibility in the test results. Stimulating electrodes are placed along the nerve as in the sensory tests. Recording electrodes are placed somewhat differently. The electrode that is attached to the active (noninverting) input on the amplifiers is inserted into the skin just over the belly of the muscle that is to be used. The inactive electrode is placed into the skin over the electrically inactive area, such as the tendon of insertion of the muscle. The ground electrode is also placed in an inactive area but near to the active electrode. Stimulus strength is just supramaximal as in SNCV estimation.

The calculation of a MNCV from an actual patient is shown in Figure D-5. Again, latencies are calculated from the length of the line to the left of the response. Here the proximal latency was 1.3 msec and the distal latency was 3.5 msec. The distance between stimulating sites was 112 mm, giving an estimate of the MNCV as 51 m/sec. In this example the nerve was the peroneal nerve, which was stimulated at the level of the fibular head (proximal) and at the level of the tarsus (distal).

FIG. D-3 Sites used for stimulation and recording in sensory nerve conduction velocity estimation from fibular and radial nerves.

FIG. D-4 Compound action potential recorded from the radial sensory nerve. The peak-to-peak amplitude is 30 FV, and latency is 1.7 ms.

FIG. D-5 Two responses recorded from the EDB muscle in the dog while stimulating the fibular nerve at the hock (A) and stifle (B). Note the difference in latencies and the similarity in response shapes. Latency A is 1.3 ms and B is 3.5 ms. The distance between stimulation sites was 112 mm. MNCV = 51 m/s.

The response was recorded from the EDB muscle. Note that the responses recorded from stimulation at the two levels were very similar. If the responses to the two different stimuli vary greatly, it may mean that different axons or even different muscles are involved in the response. This would invalidate all the assumptions involved in subtracting the two latencies to produce the conduction time and would therefore lead to invalid results.


Several factors must be considered in interpreting the results of nerve conduction studies. First, from the nature of the latency calculations it is clear that the techniques actually measure the fastest conducting fibers in either sensory or motor nerves (latency is calculated to the beginning of the response). Therefore, if only some of the axons in the nerve are involved, perfectly normal results for MNCV and SNCV may be obtained in the face of actual nerve pathology. Second, as mentioned above, the studies estimate nerve conduction velocity only over the areas between electrode sites, and since total nerve integrity is necessary to function, abnormal areas may be missed, again yielding values in the face of disease. Finally, such factors as age and temperature of the animal have important effects on nerve conduction and must not be overlooked. Young animals (under about 6 months) have markedly slower conduction velocities than adult dogs; decreasing the temperature of a nerve will also slow the velocity considerably. Consequently, when comparing the results of nerve studies from a clinical case with known values for normal dogs (values are given in Table D-1), considerable discretion should be exercised.

TABLE D-1. Normal Values for Nerve Conduction Velocity for Selected Nerves in the Dog

The usual adjunct to peripheral nerve testing is needle EMG. The purpose of this study is to find and characterize abnormalities of the neuromuscular system as they are manifested in the muscles themselves. If there is demonstrable peripheral nerve damage, EMG studies can help to determine the extent and severity of the nerve damage. In the absence of peripheral neuropathy, EMG may help to localize the lesion to one primarily of muscle.


Equipment required for EMG studies includes electronic amplifiers and display devices. Generally both visual and auditory senses are used by the observer to evaluate the muscular electrical activity. The amplifiers consist of high-impedance, high-gain devices that are capable of gains well over 1,000 and nominal input resistances of at least 1,000,000 Q. The outputs of these amplifiers are used to drive the display devices, which generally are an oscilloscope and an audio amplifier. The frequency characteristics of the combined amplifiers and display devices are critical to the correct result. For needle EMG the upper frequency cutoff should be at least 10 kHz. The lower frequency should be between 2 Hz and 20 Hz. Pre-packed equipment with all of these components and characteristics is available from a number of manufacturers.


There are three basic types of EMG needle electrodes: simple needle electrodes, single concentric electrodes, and double concentric (bipolar) electrodes. The simple needle electrode consists of a single metal needle insulated a l over except at the very tip. With this electrode a separate needle is needed as a ground or reference electrode. For the ground or reference electrode, a bare needle is suitable if not preferred. Metal electrodes tend to average all the signals over the length of their surface, and since most of the volume of a muscle is electrically silent, large areas of bare recording surface will be very close to a silent ground. For this reason also, large bare needles are not good for use as recording electrodes.

Single concentric electrodes consist of an outer hollow metal needle through which has been inserted a wire that is bare only at the tip. The outer needle is bare and serves as a ground or reference electrode. This needle type has the advantage that the recording electrode is shielded by the outer needle from electromagnetic noise and that the ground is kept as uniformly close to the recording electrode as possible. The double concentric (bipolar) electrode is similar to the single concentric electrode except that there are two recording wires in the center of the outer shell. In this case the voltage difference between the two inner wires is recorded and the outer needle serves as the ground electrode.

These three types of electrodes, because of their differing configurations, can be expected to produce different voltage records even when they are observing similar intramuscular signals. These differences are illustrated diagrammatically in Figure D-6. Between the simple electrode and the single coaxial electrode little difference is noted except that the simple electrode tends to slow down the observed response because of the large distance between it and its reference electrode. Between these two types of electrodes and the double coaxial electrode, however, there is a marked difference in the observed voltage record, since the double electrode compares two active electrodes (one is electronically subtracted from the other). A mirror image across zero voltage is often seen in the voltage record. With this electrode equal amounts of positive and negative voltage swings are to be expected. Thus, what is described as a "positive sharp wave" (see below) for single electrodes could be seen as either a positive wave or a negative wave with the double coaxial electrode. Because of these differences produced by electrode configurations, care must be exercised when comparing the voltage records from different EMG laboratories. For the purposes of this chapter, all wave-form descriptions are based on the single coaxial needle electrode.

FIG. D-6 Three types of EMG needles and records that could be used to record the same extracellular voltage changes associated with an action potential.


In the performance of needle EMG studies of clinical cases, patient cooperation is critical. Since the study requires multiple insertions and frequent movement of the needle electrode, some form of chemical restraint is generally necessary. Some electromyographers prefer general anesthesia for EMG of cats and dogs, whereas others use sedatives rather than anesthetics. Although sedatives can be effective, they do not abolish reflex activity and are therefore less useful than general anesthesia. The animal that is under anesthesia (or sedation) may be placed in any position that is convenient for the electromyographer. Of course, all needles should be sterile, but no other preparation of the animal is necessary unless there is gross uncleanliness.

To begin the EMG study, the recording needle is inserted through the skin over the muscle that is to be studied. Care must be exercised at this point lest the needle be bent and damaged as pressure is applied to penetrate the skin. Should the needle be bent, it may be damaged irreparably. Once through the skin, the needle may be inserted into the belly of the muscle. Because this is done in normal and in abnormal but excitable muscle, a large-amplitude discharge will be recorded on the oscilloscope screen (cathode ray tube [CRT]). This discharge is produced by movement of the needle and by local action potentials in the muscle initiated by physical excitation of the muscle fibers by the recording needle. Once the motion of the needle has stopped, all electrical activity recorded should also cease if the animal is under general anesthesia. Sedated animals tend to react slightly, and local action potentials may continue briefly. This electrical activity seen in association with needle movement is called "insertional activity." Because it occurs right at the tip of the moving needle and involves most of the surrounding tissue, it is recorded with a large amplitude-a few millivolts (Fig. D-7). If the muscle into which the needle is inserted is severely diseased or incapable of electrical activity, the insertional activity will be greatly reduced but not necessarily absent, since it also contains motion artifacts, potentials produced by rubbing the electrodes across anything.

FIG. D-7 Photographic record of insertional activity. The EMG needle was inserted midway during the sweep of the oscilloscope beam

Once the insertional activity has ceased and before the needle is moved again, one should watch and listen for any background electrical activity. In normal muscle in a fully anesthetized animal, this period should be one of essential electrical silence, In the lightly anesthetized animal or in the sedated one, occasional isolated activity might be observed. Activity during this period is called "resting activity," since it is observed in resting muscle. In the fully anesthetized normal animal this activity is nonexistent because there is nothing to stimulate the muscle fibers to contract. In the lightly anesthetized or sedated animal the occasional activity is thought to be recorded from those motor units that are maintaining muscular tone. It is the analysis of the activity observed during this resting period that is most important in veterinary EMG work, for it is in this period that abnormal and normal muscle are most likely to distinguish themselves (see below). Once the analysis of insertional and resting activity has been performed in one spot in a muscle, the needle may be withdrawn from the muscle (but not from the skin if possible) and reinserted into another site until the area has been adequately sampled. When sampling muscles in this way, it is important to remember that the coaxial needle records only from a small area near its tip, and while one area may be normal an immediately adjacent one can manifest an important abnormality. Once the area of muscle immediately under the site of skin penetration has been adequately sampled, the needle may be withdrawn from the skin and inserted in the next site. This approach to sampling saves wear and tear on both the needle and on patient skin.


The interpretation of abnormal electrical activity observed and heard during the EMG study has often been overdone. When an electromyographer gets used to seeing a certain wave form again and again, he generally gives the form a name, associates it with some form of muscular or neural pathology, and then goes one step too far by creating a story about the physiological genesis of the wave form.' The first two stages are valuable and necessary in the interpretation of EMG results. Indeed, they provide its clinical utility. The third step is unnecessary and misleading and should be avoided as much as possible.

There are several features of denervated and diseased muscle that serve to alter its electrophysiological characteristics, and these alterations are reflected in the patterns of activity seen in the EMG study, particularly in the resting activity. The altered electrical activity seen in abnormal muscle, except when associated with myotonia, is generally very much smaller than the large amplitude insertional activity and therefore observable only during the resting phase, although it probably is continuous. Several types of wave forms have been characterized and associated (variously) with neural and muscular diseases.

The abnormal wave forms include the fibrillation potential, the positive sharp wave, the motor-unit potential (also considered normal), the myotonic potential, and bizarre high-frequency discharges. These wave forms are categorized in terms of their amplitude, duration, multiplicity of phase, and sign of their major component. Clinical examples of these various wave forms are reproduced in Figure D-8. Note that both time and voltage scales are different in each section. The most commonly observable abnormality is the fibrillation potential. This wave form is small, less than about 200 FV, uniphasic (goes mainly in one direction), and less than about 7 msec in duration. When many fibrillation potentials occur together, they are generally randomly spaced. This small random signal produces a crackling in the audio amplifier, which is often likened to the sound of frying food. The positive sharp wave is characterized by a large positive (downward) deflection-several hundred microvolts- and a duration of several times that of the fibrillation potential. The positive sharp wave is generally considered a uniphasic potential, but it often has a small slow negative component that follows the positive spike. When double coaxial electrodes are used, "negative sharp waves" (mirror images of positive sharp waves) are also seen. There are a variety of theories about the genesis of these signals'; like the fibrillation potentials, they are seen in association with either primary muscle pathology or that which is secondary to denervation.

FIG. D-8 Types of activity seen during the resting phase: (A) mild fibrillation potentials; (B) positive sharp waves; (C) a motor unit; (D) bizarre high-frequency discharge.

One type of electrical activity that was mentioned above in connection with muscular tone but that may be seen in other circumstances as well is termed the motorunit potential. The functional element of the muscular system is the motor unit, which is made of a single motor neuron and all the muscle fibers to which its axon is connected. Depending upon the muscle type, the muscle fibers innervated by a single axon may number several hundred. Excitation of a single axon under such conditions will cause all of the muscle fibers to initiate an action potential at almost, but not quite, the same time. The variability in onset of these action potentials is compounded further by slight variability in conduction velocity of each activated muscle fiber. Thus, to an observer (or an electrode) near the muscle fibers of an active motor unit, the resultant electrical activity will seem to be a sudden brief shower of action potentials racing by. The extracellular current will be a complex mixture of currents running in one direction, then another, and back again, as the beginning and ending of each action potential pass nearby but at slightly different times and distances. Such activity is recorded by the electrode with large-amplitude and many-phase (direction) changes on the oscilloscope screen. Further, since motor nerves fire with a regular frequency, the activity recorded from the motor unit will be regularly repetitive. (A word of caution: While this may be a plausible description of the genesis of the motor-unit potential, it must not be assumed that all large-amplitude, multiphasic potentials are motor units, since a variety of electrical events might produce similar potentials.)

Two striking electrical phenomena that are seen occasionally by the electromyographer are the myotonic discharge and the bizarre high-frequency discharge. The individual components of these two discharges are quite similar; however, the duration and modulation characteristics (envelope) of the two are different. The repetitive units of both of these signals are similar to what has just been described as the motor-unit potential, namely, large amplitude, multiphasic signals. The units are seen repetitively but with a much higher frequency than is normally seen with motor-unit activity-up to as many as 200 times per second. For both types of activity the frequency is variable with time. In the myotonic discharge the frequency and amplitude tend to diminish over a period of 4 to 5 seconds but may then increase again for the same period. This waning and waxing of amplitude and frequency produces the auditory effect similar to that of the dive-bomber from World War II. This auditory component is unmistakable. What is referred to as the bizarre high-frequency discharge is similar to the myotonic discharge but tends to be shorter in duration and generally does not gain in amplitude and frequency after diminishing in these characteristics. It may be a brief period of remarkable activity lasting a second or two and then ending very abruptly. The myotonic discharge has been described only in association with myopathy not due to denervation. In the dog it may be seen in conjunction with primary muscle disease or disease secondary to high levels of either endogenous or exogenous corticosteroids. It can be seen in conjunction with inflammatory muscle disease-polymyositis. The bizarre high-frequency discharges, on the other hand, may be seen with denervation myopathy.

From the above descriptions of the abnormal events that can be observed by needle EMG and the loose association of these events with denervation and muscle disease, it should be obvious that EMG does not often lead directly to diagnosis. Rather, in conjunction with history, signalment, and other clinical signs, the information gained by EMG may help to confirm or deny diagnostic impression. EMG is not the final word. It is a study that can provide insight into the distribution and extent of muscle involvement in an animal's disease.

It should be obvious that the techniques of peripheral nerve testing are beneficial mainly in the evaluation of cases of suspected peripheral nerve or muscle disease. In general, these cases have in common one cardinal sign-loss or depression of spinal (segmental) reflexes. This loss may be generalized or it may be well confined to a single area, but peripheral disease almost always means hypoactivity in local reflexes. The loss of activity can be due to disease in either the sensory or the effector branch, or both, of the reflex arc. The electrophysiological techniques help to define and quantify the effects of peripheral disease. The observations one would expect to make in the three main classes of peripheral diseases are listed in Table D-2.

TABLE D-2. Electrophysiological Signs of Peripheral Nerve/Muscle System Disease According to Location

If a peripheral nerve is transected surgically, it will not allow conduction of action potentials across the transection. However, for several days following the transection, stimulation of the distal portion of the nerve will excite action potentials that will be conducted with almost normal velocity for up to 5 days after transection. Thus, in a clinical case of peripheral nerve disease, regardless of the severity of signs, it may take some time before electrical evidence of disease is observable. However, if the portion of nerve that is directly affected by disease or transection is located at a point at which conduction across the site can be tested, the fault will be observable immediately. If the location is close to the spinal cord and thus inaccessible to standard testing, F wave studies may prove useful. For a period of about 7 days after the formation of the lesion (depending upon the distance from the muscle) there may be no EMG evidence of denervation. After this time fibrillation and the presence of positive sharp waves during the resting phase will signal the distribution and perhaps the extent of denervation in involved muscles. If the lesion is limited to sensory nerves alone (e.g., at the dorsal root), no evidence of denervation in muscle and no abnormalities of MNCV will be observed. Slowing and loss of conduction in the distal portion of the sensory fibers will appear only after a few days delay, if at all.

In the case of primary muscle disease, such as myositis, it is not known whether there are significant delays in the development of electrical evidence after the onset of the disease. Certainly, by the time animals are presented to the clinic, there is no problem in finding EMG signs of muscle pathology. Unfortunately, there are no absolute criteria for distinguishing primary muscle pathology from secondary muscle pathology using the EMG.

In the cases of postsynaptic neuromuscular junction disease that we have seen in the dog (canine myasthenia), the only electrophysiological evidence of disease has come from monitoring the magnitude of the muscle action potential during repetitive stimulation. Presumably, this sign develops as the disease becomes clinically significant (produces weakness). we have not seen any abnormal activity in the EMG studies, nor have we appreciated any slowing in MNCV in these cases. In cases of presynaptic junctional disease, one could presumably appreciate some abnormal EMG activity if the damage to the nerve terminals were sufficiently severe. However, diagnosis would have to come from repetitive nerve stimulation and ancillary tests.


The combination of needle EMG and peripheral nerve conduction velocity measurement can provide significant information about the functional status of the various components of the peripheral neural and muscular systems. In general the studies should be done in concert with each other since they are complementary. Motor nerve lesions produce electrophysiological manifestations both in the peripheral nerve per se and in the muscles that they innervate. Simple observation of EMG changes in muscle characteristic of denervation is inadequate evidence of denervation, since these changes may also be produced by primary muscle disease. Conversely, when primary myopathy is suspected, peripheral nerve disease should also be considered and evaluated by MNCV studies. None of the evidence garnered by these techniques is concrete. Like all the other tests in clinical work, the results of these measurements must be evaluated carefully in fight of the whole animal, its history, its physical and neurologic examinations, and other laboratory findings. Final definitive diagnosis eventually requires biopsy of the diseased part.



1. Farnbach GC: Electrodiagnostic testing (Letter to the editor). J Am Vet Med Assoc 174:1152, 1979

2. Lachman T. Shahaini BT, Young RR: Late responses as aids to diagnosis in peripheral neuropathy. J Neurosurg Psychiatr43:156, 1980

3. Northington JW, Brown MF, Farnbach GC et al: Acute idiopathic polyneuropathy in the dog. J Am Vet Med Assoc 179:375, 1981

4. O'Leary JL, Landau WM, Brooks JE: Electroencephalography and electromyography. In Baker AB, Baker LH feds): Clinical Neurology. San Francisco, Harper & Row, 1977