The EDX is divided into two main parts: the nerve conduction studies (NCS) and the NEE. Both portions are necessary for a complete study.

NCS are performed by stimulating either a motor or sensory nerve with an external electrical signal that propagates along the nerve and is recorded as an electrical response by surface electrodes in a different location. Analysis of the data reveals the speed of the fastest fibers, the rate of conduction, and the number of viable axons. A minimum protocol that includes both motor and sensory NCS should be performed on every patient; however, the protocols vary according to the diagnosis and should be individualized to ensure that a particular referral question is answered and that possible confounding diagnoses are excluded.

To perform a motor NCS, the recording electrode is placed over the belly of the muscle and a reference electrode is usually placed distally over the muscle’s tendon. The motor nerve is then stimulated, resulting in a waveform called a compound muscle action potential (CMAP). Each CMAP is assessed for amplitude, onset latency, and conduction velocity. The amplitude represents the number of functioning motor nerve axons and is expressed in millivolts. The latency is the time between the stimulation of the nerve and the muscle response, which is seen at the “takeoff” of the CMAP waveform and is expressed in milliseconds. The conduction velocity, measured in m/s, is the speed at which the electrical signal is propagated along the nerve with normal values greater than or equal to 50 m/s in the upper extremity.

Nerve compression caused by cervical spine disease typically does not result in significant abnormalities of the motor nerve conduction responses unless there has been moderately severe axon loss, which is manifested as a reduced CMAP. However, both the latency and conduction velocity remain unaffected in most cases, except in very severe axon loss lesions where the fastest fibers have been affected.

Routine sensory NCS of the upper extremity are performed by placing a recording electrode on a distal portion of the limb (typically one of the digits) and stimulating more proximally along the sensory component of the nerve in an antidromic fashion. This results in a waveform called the sensory nerve action potential (SNAP). The SNAP amplitude is much smaller than a CMAP as the response is directly recorded over the nerves rather than a muscle and is measured in microvolts. In addition to the amplitude, the peak latency is the other component of the SNAP that is of importance. The peak latency is the time from stimulation to the negative peak of the curve and is reported in milliseconds. As the dorsal root ganglion (DRG) lies outside of the intraspinal canal, cervical radiculopathies do not affect the sensory NCS unless there is an infiltrative or extensive mass lesion. In most cases, the presence of an abnormal sensory response is indicative of a more distal lesion such as a plexopathy or mononeuropathy.

An additional component of the NCS is the assessment of late responses, which is comprised of the H response and the F wave. As the H response is not usually performed in the upper extremity, it is not discussed further. In contrast, the F wave is routinely recorded from motor nerves in both the upper and lower extremity. Theoretically, it is helpful in evaluating the more proximal portion of the nerve as it represents the time required for a motor nerve impulse to travel from the distal portion of the limb at the site of stimulation up to the anterior horn cell in the spinal cord and then back down the motor nerve to the muscle where the recording electrode is attached. In reality, however, it is typically of limited benefit in the setting of a cervical radiculopathy as it often remains normal even in unequivocal cases of nerve root compression.

The NEE is the single most important part of the EDX when evaluating for a cervical radiculopathy. In this portion of the study, a small needle electrode is inserted into different muscles of the arm and cervical paraspinals and records the electrical activity generated by the muscle at rest and on activation. The NEE is divided into two phases: resting and activation.

At rest, the response of healthy muscle should approximate electrical silence except for the initial insertional activity, which reflects the very minute muscle fiber damage that occurs with actual needle insertion. In most normal muscles, this phenomenon is only momentary and quickly dissipates. The absence of insertional activity points toward a fibrotic or atrophied muscle, while the presence of more prolonged insertional activity, especially in the form of positive sharp waves, is suggestive of hyperacute denervation.

After the needle is inserted, the examiner looks for the presence of spontaneous activity in the form of fibrillation potentials and positive sharp waves, which are the most sensitive indicator of active or acute denervation and may be present before the onset of clinical weakness (1). Nerve injury results in spontaneous excitation of single muscle fibers, which are manifested as action potentials that fire in a regular pattern. The morphology is typically a biphasic spike but can also be in the form of a positive sharp wave if the needle electrode induced damage to the already abnormal muscle fiber. The presence of fibrillation potentials is objective evidence of denervation that cannot be suppressed or produced by the patient; finding them in a myotome distribution is the primary method of detection of a cervical radiculopathy (2). However, the average time that it takes for fibrillation potentials to appear in a denervated muscle is 21 days, after which, they may persist for up to 24 months (2). For this reason, it is advised that one waits at least 21 days after injury or onset of symptoms before undergoing an EDX.