6.2 Conduction Velocity Calculation and Normal Values

Why velocity is a two-site number

Conduction velocity (CV) answers a different question from distal latency. Distal latency is a clock reading at one stimulation site. Velocity is a speed: how fast the fastest axons cover a measured nerve segment between two sites. Independent OpenExamPrep coverage of outline items III.A.2.a.3 and III.A.2.b is the formula, the unit conversion, the reasons you must not turn distal latency into a fake speed, the sensory onset-versus-peak choice, distance pitfalls, inching, and a sane attitude toward normal values.

Motor conduction velocity formula

For motor studies:

CV (m/s) = distance (mm) / (proximal onset latency − distal onset latency) (ms)

Use onset latencies at both stimulation sites, with the same recording electrodes. The distance is the nerve-path length between the two cathodes (active stimulating poles), in millimeters. Subtracting the two latencies cancels the shared distal pathway that both responses still have to travel: the distal nerve, the NMJ, and muscle-fiber depolarization under G1. What remains is time spent on the interstimulus segment of nerve.

Unit check: millimeters per millisecond equals meters per second. 1 mm/ms = 1 m/s. That is why the formula is written in mm and ms rather than in cm and seconds.

Never compute a motor velocity from distal latency alone. Distal motor latency includes NMJ transmission time and muscle depolarization time that are not nerve conduction. Dividing a wrist-to-muscle skin distance by distal latency therefore mixes nerve, synapse, and muscle into a number that is slower than true nerve CV and is not comparable to segment velocities. Distal latency is reported as latency, in milliseconds, against distal-latency reference values. Velocity is reported only for a nerve-to-nerve segment.

Worked example (handbook-style arithmetic)

A motor study has:

• Distal (wrist) onset latency = 3.4 ms
• Proximal (elbow) onset latency = 7.4 ms
• Cathode-to-cathode distance = 226 mm

Latency difference = 7.4 − 3.4 = 4.0 ms

CV = 226 / 4.0 = 56.5 m/s, which rounds to the nearest meter per second as 57 m/s.

If someone mistakenly divides 226 by the distal latency 3.4, they get about 66 m/s and have not measured nerve speed. If they divide 226 by the proximal latency 7.4, they get about 31 m/s and have punished the proximal site for distal NMJ and muscle time that do not belong to the forearm segment.

A second common arithmetic error is mixing units (centimeters in the numerator without converting). 22.6 cm / 4.0 ms is not 5.65 m/s; you must convert 22.6 cm to 226 mm, or convert 4.0 ms to 0.004 s (22.6 cm / 0.004 s = 5650 cm/s = 56.5 m/s). The mm/ms form is the least error-prone at the machine.

Sensory conduction velocity

Sensory CV uses the same subtraction idea, but the latency marks follow the protocol:

• Onset-to-onset CV estimates the fastest sensory fibers and is preferred when both onsets are clear.
• Peak-to-peak CV uses the time between SNAP peaks. Peaks are easier to mark on small SNAPs. They do not isolate the single fastest axon as cleanly as onsets, because peak time depends on waveform shape. Many sensory laboratory manuals still define velocity with peaks because that is how their reference data were built.

A distal sensory latency — onset or peak — is still not a velocity. Digital distance divided by peak latency folds in nerve timing plus whatever the mark represents; laboratories publish it as latency, then calculate velocity only when they have two stimulation sites (or a measured nerve segment with two marks). Mixed-nerve studies follow the same algebra as motor or sensory depending on which potential you marked.

Stay consistent across a case. Do not calculate one segment with onsets and the next with peaks unless the laboratory's method book explicitly defines that mix, and never compare an onset-based velocity with a peak-based normal table.

Distance measurement pitfalls

Velocity is only as honest as the millimeters. Teaching pitfalls:

1. Path, not a shortcut. Lay the tape or caliper along the estimated nerve course, including gentle curves. A straight line that chops across the palm or across a flexed joint underestimates distance and lowers the calculated CV (the numerator is too small).
2. Calipers versus tape. On short segments, a tape that bridges skin folds overestimates or underestimates depending on slack. Calipers (or a firm tape laid on skin) reduce that error. Infant and inching segments are where this matters most; the pediatric section returns to it.
3. Cathode to cathode. Measure between stimulating cathodes, not from anode to anode, and not from a random skin pen mark that drifted after you moved the stimulator.
4. Ulnar nerve across the elbow. Measure with the elbow flexed (commonly taught near 70–90 degrees, matching however the laboratory standardized its ulnar norms). In full extension the ulnar nerve is redundant and slack; skin distance underestimates true nerve length, the numerator shrinks, and CV looks slower than the nerve really is — a false-positive trap for ulnar neuropathy at the elbow.
5. Limb position consistency. Median forearm, fibular (peroneal) across the knee, and tibial studies all have a position in the laboratory's method book. Changing hip, knee, or wrist angle between the distance measurement and the stimulation changes the path.
6. Skin stretch and obesity. Soft tissue lets the tape wander. Have an assistant stabilize; do not cinch a tape until the skin tents several extra millimeters.

A 10 mm error on a 200 mm segment is 5% — about 2–3 m/s on a 50 m/s nerve. The same 10 mm error on a 50 mm inching or infant segment is 20%.

Inching (short-segment stimulation)

Inching is sequential stimulation at short, equal steps, typically 1 cm or 2 cm, across a suspected entrapment (ulnar groove / cubital tunnel, wrist, fibular head). You plot latency against distance and watch for a step increase in latency or a focal amplitude drop. Inching converts a long-segment CV that only says the elbow segment is slow into a map of where the clock jumps.

Inching does not have a single unpublished national cutoff that you should memorize as the AAET inching number. Laboratories define their own step size and what they consider a disproportionate latency jump compared with neighboring centimeters. The skill is the geometry: equal measured steps, cathode placement perpendicular to the nerve, and honest distances with the elbow or knee in the standardized position. Long-segment CV can dilute a 15 mm lesion inside a 200 mm segment; inching exists because of that dilution.

Normal values: laboratory-derived, not a fake national line

III.A.2.b asks for normal values, not for an invented national cut score. Independent OpenExamPrep teaching is:

• Each laboratory derives reference ranges from its own distances, temperatures, instruments, electrode sizes, and population.
• Do not treat any number in this study guide as the AAET normal. The credentialing outline does not replace a laboratory's reference table, and this independent resource does not publish official unpublished exam cut scores.
• Values depend on limb temperature, age, and height. Cooling slows sodium-channel kinetics and slows CV while often increasing amplitude; height and longer legs tend to associate with slightly slower lower-limb CV; later adulthood slows CV. Full physiology of those factors is a later chapter; here you only need to know that a number is meaningless without those contexts.
• Common teaching approximations for adult lower limits of motor CV are about 50 m/s in the arm and about 40 m/s in the leg. Those figures order the expected anatomy (upper-limb segments usually faster than lower-limb segments) for classroom memory. They are not official unpublished cut scores, not a substitute for the laboratory printout, and not pediatric norms.

When a tracing is 49 m/s in the forearm, the technologist's job is to check temperature, distance, onset marks, and the lab's table — not to apply a remembered slogan. Side-to-side comparison on the same date, same temperature, and same method sometimes flags a problem before an absolute value does, but asymmetry rules are also lab-specific.

If you can run 226 / (7.4 − 3.4) to 57 m/s, refuse to divide distal latency into a fake speed, measure the ulnar elbow flexed, inch when a long segment hides a focal lesion, and treat normals as lab-derived and temperature/age/height dependent, you have the velocity half of waveform evaluation.