Curriculum/Pillar 4 · Neuromuscular Disease Mechanisms

Demyelinating Neuropathies

Module 14 of 36·30 min readLIVE MODEL

Learning objectives

1Explain how segmental demyelination slows or blocks conduction
2Distinguish uniform (inherited) from multifocal (acquired) demyelination
3Relate the safety factor to block and dispersion

Myelin exists to make conduction fast and cheap. By insulating the internode and confining excitable membrane to the nodes of Ranvier, it converts the slow, continuous, metabolically expensive propagation of an unmyelinated fibre into the rapid, saltatory leap of a myelinated one. Demyelination dismantles that engineering. The signals it produces — disproportionate slowing, conduction block, temporal dispersion — are not a quantitative dimming of the axonal picture but a qualitatively different electrodiagnostic syndrome, and each feature can be derived from the cable biophysics of a node that can no longer reach threshold in time.

LIVE MODELMedian motor study — switch to the demyelinating preset

Drag calipers A and B onto the two onsets to measure conduction velocity across the forearm segment.

Median motor study · record APB

Wrist → elbow distance23 cm

Distal latency

3.4ms

CMAP amplitude

9.0mV

Measured CV (A→B)

50m/s

Prox/dist amp

95%

Normal. Normal distal latency, conduction velocity, and amplitude with negligible drop on proximal stimulation.

Select the Demyel. case in the model. It encodes the full acquired-demyelinating signature: markedly slow forearm velocity (≈30 m/s), a prolonged distal latency, temporal dispersion of the proximal response, and a greater than 50% fall in CMAP amplitude from wrist to elbow — conduction block. Drag calipers A and B onto the two onsets and watch the measured velocity drop into the demyelinating range, then compare against the Axonal case, where amplitude collapses but velocity is nearly preserved.

Segmental and paranodal demyelination

Demyelinating injury is segmental: it strips myelin from individual internodes while leaving the axon cylinder intact, rather than degenerating the fibre along its length. The earliest and most functionally consequential lesion is often paranodal — disruption of the specialised axoglial junctions (the septate-like contacts mediated by neurofascin-155, contactin, and Caspr) that flank the node and seal the periaxonal space. Detachment of the paranodal loops widens the node, short-circuits nodal current laterally under the former myelin, and disperses the dense cluster of voltage-gated sodium channels that normally concentrates at the node. Because the axon survives, the dominant deficits are of conduction, not of fibre number, and remyelination — with characteristically short, thin internodes — can restore function over weeks once the immune attack abates.

The biophysics of slowing

In a healthy myelinated fibre the internode behaves as a near-ideal cable: high transverse membrane resistance and low membrane capacitance mean that almost the entire nodal current flows longitudinally down the axoplasm to the next node, charging it to threshold with minimal loss and minimal delay. Demyelination inverts both parameters. Stripping myelin raises the internodal capacitance — the thin or absent insulating sheath now stores far more charge per unit length — and lowers the transverse resistance, opening leak pathways across the denuded membrane. The local circuit current generated at an active node must therefore first charge this enlarged capacitance and is simultaneously bled away through the lowered resistance, so the downstream node is charged toward threshold more slowly. The internodal conduction time lengthens, and across many affected internodes this accumulates into the disproportionate slowing — and the prolonged distal latency — that is the hallmark of demyelination. Slowing of this magnitude (velocities well below 70–75% of the lower limit of normal) cannot be produced by axon loss, because no surviving population of large fibres conducts that slowly.

The safety factor and conduction block

The safety factor is the ratio of the current available at a node to the current required to depolarize the next node to threshold; in healthy nerve it is roughly 5 to 7, an enormous margin. As demyelination raises capacitive load and resistive leak, that margin erodes. So long as the safety factor remains above 1, conduction continues but is slowed. The instant it falls below 1, the depolarizing current dissipates into the leaky internode before it can bring the next node to threshold, the impulse fails to propagate, and conduction block supervenes. Block is therefore the electrophysiological threshold phenomenon of demyelination — a binary failure superimposed on the graded slowing — and it is the single finding most specific for an acquired demyelinating process, because axon loss can never produce a focal drop in amplitude across a segment with the distal response preserved.

Temporal dispersion and phase cancellation

Demyelination does not slow every fibre equally. Differential slowing across the population spreads the arrival times of individual fibre action potentials at the recording electrode, so the summated CMAP becomes broadened in duration and reduced in amplitude — temporal dispersion. Crucially, the amplitude reduction here is not purely a loss of conducting fibres but a consequence of phase cancellation: as the desynchronised potentials spread out, the negative phase of a faster fibre's potential increasingly overlaps the positive phase of a slower one, and the two partially annul on summation. This is why temporal dispersion can mimic conduction block by lowering the proximal amplitude. The discriminator is the accompanying duration: pathological dispersion broadens the proximal CMAP markedly (conventionally a duration increase beyond about 30%), whereas true block drops the area and amplitude without comparable broadening. Distinguishing the two is the crux of demyelinating NCS interpretation, and the model above is built to make the contrast palpable.

Uniform versus multifocal demyelination

The spatial distribution of demyelination, not merely its presence, separates the two great categories of demyelinating neuropathy and is mechanistically decisive. Uniform demyelination is the signature of the inherited disorders — classically CMT1, in which a genetic defect of a myelin protein (the PMP22 duplication of CMT1A being the prototype) affects every internode of every fibre to a similar degree. Because all fibres are slowed equally and synchronously, there is no conduction block and no temporal dispersion; the CMAP is small but its shape is preserved, and conduction velocities are diffusely, homogeneously, often symmetrically slowed (frequently below 38 m/s in the arms). The slowing is uniform along the nerve and uniform from nerve to nerve. Multifocal demyelination is the signature of the acquired immune disorders — GBS/AIDP, CIDP, and MMN — in which the attack is patchy, falling on some internodes and sparing others. This non-uniformity is precisely what generates conduction block and temporal dispersion at the sites of focal demyelination, scattered asymmetrically along and between nerves. The presence of block or dispersion thus argues strongly for an acquired, potentially treatable process; their absence in the face of uniform slowing points to a hereditary one.

Clinical Pearl

Block and dispersion are the fingerprints of acquired demyelination, and their absence is as informative as their presence. Uniform, symmetric slowing with preserved CMAP morphology and no block is the picture of an inherited demyelinating neuropathy (CMT1); patchy slowing with conduction block and temporal dispersion is the picture of an acquired immune neuropathy (CIDP, GBS, MMN). Because the acquired forms are treatable and the inherited forms are not, this single distinction frequently determines whether a patient receives immunotherapy.

Common Pitfall

The most consequential error is conflating temporal dispersion with conduction block, because true block — when persistent and not at a common entrapment site — is a strong, specific marker of acquired demyelination, whereas marked dispersion can lower the proximal amplitude through phase cancellation alone. Before calling block, confirm that the proximal-to-distal amplitude (and area) drop is not accompanied by the substantial duration broadening that defines dispersion, and ensure the drop is not an artifact of supramaximal stimulation failure, anomalous innervation (the Martin–Gruber anastomosis can fabricate an apparent forearm drop), or an over-long segment.

Key points

Demyelination is segmental and often paranodal, sparing the axon — so deficits are of conduction (slowing, block, dispersion), not of fibre number.
Slowing is biophysical: lost myelin raises internodal capacitance and lowers transverse resistance, so nodal current charges the next node slowly — internodal conduction time rises and latency prolongs.
Conduction block occurs when the safety factor falls below 1: current dissipates into the leaky internode before the next node reaches threshold — the most specific NCS sign of acquired demyelination.
Temporal dispersion arises from differential fibre slowing plus phase cancellation (overlap of faster negative and slower positive phases), broadening the CMAP and lowering its amplitude.
Uniform demyelination (inherited, CMT1: PMP22) affects every internode equally → diffuse symmetric slowing with NO block or dispersion and preserved CMAP shape.
Multifocal demyelination (acquired: GBS/AIDP, CIDP, MMN) is patchy → conduction block and temporal dispersion — the treatable pattern.