Sensory Nerve Conduction

A signal without synaptic gain

A SNAP is the summed extracellular field of sensory axons depolarised by a stimulus, recorded directly over the nerve. The motor system interposes a neuromuscular junction whose endplate potentials and the ensuing muscle fibre depolarisations multiply the recorded signal into the millivolt range; the sensory system has no such amplifier. Consequently SNAP amplitudes are microvolt-scale — characteristically 5–50 µV, three orders of magnitude smaller than a CMAP — and routinely demand signal averaging of multiple time-locked sweeps to lift the response above the random background, whose amplitude falls as the square root of the number of trials. The amplitude itself, measured baseline-to-negative-peak or peak-to-peak, is a direct count of functioning large sensory axons and is the SNAP parameter most sensitive to early axonal disease.

Because no synapse intervenes between stimulus and recording, the SNAP confers a measurement the motor study cannot: a pure conduction time. The onset latency marks the arrival of the fastest fibres and converts directly into a conduction velocity over a single segment, with no NMJ or activation overhead to subtract. The peak latency, taken at the apex of the negative phase, is more reproducible and less observer-dependent and is therefore the conventional currency for normative comparison, though it corresponds to a slightly slower fibre population and cannot itself be turned into a true velocity.

Direct velocity over a single segment

Where motor conduction velocity demands two stimulation sites to cancel the distal delay, sensory conduction velocity is obtained from a single segment: CV = distance ÷ onset latency, because the onset latency is already a clean conduction time from cathode to recording electrode. This economy is the reason sensory studies localise so efficiently — one stimulus, one recording site, one velocity for that stretch of nerve. The same averaging that rescues a small SNAP from noise also sharpens the onset, but a poorly defined onset (the commonest consequence of an undersized or dispersed response) shifts the latency and biases the velocity, so amplitude adequacy and velocity accuracy are coupled.

Antidromic versus orthodromic technique

The same nerve can be studied in either direction, and the choice is a deliberate trade-off between signal size and signal purity. Antidromic recording drives the impulse opposite to its physiological direction — stimulating proximally along the nerve trunk and recording distally over the digit with ring electrodes. The recording sits close to a superficial, large-diameter cutaneous fascicle, so the antidromic SNAP is larger and more easily obtained, which is why it is the routine clinical choice. Its liability is that the stimulus also drives an orthodromic motor volley to nearby muscle; the resulting volume-conducted CMAP can overlap and obscure the SNAP, a contamination recognised by its later, larger, differently shaped waveform.

Orthodromic recording follows the physiological direction — stimulating the digit and recording proximally over the nerve trunk. The response is intrinsically smaller, because the recording electrode is farther from the generator, but it is free of motor contamination because the digital stimulus excites no muscle in the recording field. The practical rule is to default to antidromic for sensitivity and reach for orthodromic when motor overlap threatens the measurement.

This single dissociation converts the SNAP into a localiser of extraordinary value: numb limb with a normal SNAP places the lesion at or proximal to the root; numb limb with an absent SNAP places it in the plexus or peripheral nerve.It is the reason sensory studies remain indispensable even when the presenting complaint is purely motor or purely "sensory radicular," and it underwrites the electrodiagnostic separation of cervical radiculopathy from brachial plexopathy.

Axonal versus demyelinating sensory signatures

The sensory study partitions disease along the same axis as the motor study, with the parameter weightings inverted toward amplitude. An axonal sensory neuropathy is defined by reduced or absent amplitude with relatively preserved velocity and latency, distributed in the length-dependent gradient that makes distal lower-limb responses (sural, superficial peroneal) fail before upper-limb responses. A demyelinating process produces disproportionate slowing of velocity and prolongation of latency, and in acquired forms, temporal dispersion of the sensory response. Because sensory fibres are heterogeneous in diameter, sensory temporal dispersion and phase cancellation are even more pronounced than in motor nerves, so an absent SNAP can reflect either severe axon loss or severe dispersion-driven cancellation — a distinction the amplitude of accessible neighbouring nerves helps adjudicate.