The "F" stands for foot, because these responses were first recorded from intrinsic foot muscles. On stimulation of a single motor axon, the wave of depolarization will travel distally to be recorded as part of the CMAP, or "M-wave," but also will travel proximally to its anterior horn cell (AHC). Retrograde depolarization of AHCs will result in regeneration of an action potential at the axon hillock in a small subset of neurons (~5-10%), which then travels back down the motor axon to the innervated muscle, recorded at the electrodes as the F-wave (Fig. 3). Although, for theoretical reasons, it is desirable to reverse the polarity of the stimulator, placing the anode distally and cathode proximally to avoid "anodal block," this remains more of a theoretical rather than a practical consideration. F-waves are easily generated with the cathode in the distal position. At the level of the spinal cord, no synapse is involved, and an F-wave from a single axon has the same morphology and almost identical latency each time. However, a single axon will not generate an F-wave with each successive depolarization. The probability that an F-wave will be generated from any neuron is dependent on the variable excitability of the AHC membrane that can be increased with reinforcement maneuvers (clenching teeth or making a tight fist) and decreased by sleep and anesthesia. F-waves may be absent in sleeping, sedated, or comatose persons, and should not be interpreted as a sign of peripheral nervous system disease in these patients. During NCS, supramaximal stimulation of the peripheral nerve ensures that all motor axons capable of generating an F-wave are depolarized each time. Under normal conditions, several axons generate an F-wave, and the summation of these responses is recorded. The varying probability that a single axon will produce an F-wave results in variation of the morphology and latency of the summated F-waves. In severe neuropathic conditions, when only one axon capable of generating an F-wave is surviving, it is recorded as an "all or none" response without variability. Also under neuropathic conditions, unusually high amplitude F-waves may be recorded because of reinnervation of the motor units.
At least 10 F-waves are usually collected for analysis. The minimum F-wave latency is the parameter measured in most labs, but reflects only the fastest motor axon contributing to the F-wave; this value may be spared in pathological situations and, importantly, in acquired demyelinating neuropathies. Mean or median F-wave latencies provide better information concerning the proximal sections of the nerves in diseased states. However, most EMG
laboratories do not have normal values from which to interpret abnormalities in these parameters. Other F-wave parameters that may also be useful include chronodispersion and persistence. Chronodispersion is a measure of the range of minimum latencies and is normally less than 5 ms. This range can be exceeded in demyelinating neuropathies or radiculopathies. Persistence is a measure of the frequency of obtaining an F-wave after supramaximal stimulus, and varies for different motor nerves. Persistence is 80 to 100% for most nerves, but peroneal F-waves may be difficult to elicit, even in healthy persons.
F-wave minimum latency is dependent on limb length, and nomograms are used to judge abnormality. If a height-adjusted table is not available, the appropriate F-wave latency corrected for height (±2.5 ms) can be estimated using the formula:
F-estimate (ms) = [2 x F distance (mm)/CV (m/s)] + distal latency (ms) + 1 ms where F distance is from the stimulus site to the C7 spinal process (median, ulnar) or the xiphoid process (peroneal, tibial), and CV is the conduction velocity. The addition of 1 ms allows for the central conduction delay. F-estimates can also be useful for identifying proximal pathology in the proximal segments of a nerve. Because the calculation of the estimate relies on the conduction velocity, which is obtained distally, proximal pathology will create a longer measured F-wave latency than would be anticipated by the F-estimate calculation.
Because F-waves probe the proximal portions of nerves, it seems reasonable that they should be useful in studying the plexus and nerve roots, which otherwise can only be directly stimulated using special techniques. Unfortunately, F-waves do have limitations for common pathology, such as structural radiculopathies, for several reasons. For example, the standard median and ulnar motor NCS evaluate only the C8 and T1 roots, which are not commonly damaged in structural radiculopathies, C5, C6, and C7 radiculopathies being much more common. Further, the shared root innervation (C8, T1) of abductor digiti quinti and abductor pollicis brevis muscles means that, in the setting of a completely transected C8 root, the F-wave latency may still be normal because of sparing of the T1 root. F-waves are likely to be more useful in detecting radiculopathies affecting the lower extremities, where standard peroneal and tibial motor conduction studies evaluate L5 and S1 derived neurons. Finally, because the F-response is a pure motor phenomenon, radiculopathies affecting only the sensory root cannot be detected (although this also remains a limitation of needle EMG).
Recall that the F-wave evaluates the proximal and distal segments of the nerve because it travels to the root level and then back to the distal muscle to be recorded. This implies that F-waves may be prolonged in distal entrapment neuropathies and polyneuropathies; thus, abnormal F-waves are not specific for proximal nerve pathology.
F-waves have their greatest usefulness in the setting of demyelinating radiculoneu-ropathies, especially Guillain-Barre Syndrome (GBS). Early GBS is frequently characterized by prominent involvement of the spinal roots. In these cases, the distal motor nerve segments may be electrophysiologically normal despite clinical weakness and areflexia caused by demyelination at the root level. F-wave minimum latency is commonly prolonged, and persistence can be severely reduced or absent in this setting; this suggests demyelination with conduction block in the proximal segments of the nerves. The combination of a normal distal CMAP and absent F-waves is highly specific for proximal demyelination. F-wave latencies are most prolonged 3 to 5 wk after the onset of GBS, secondary to further demyelination throughout the length of the nerve.
Was this article helpful?
This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.