Human health evidence

 Quick Answer Copied

 Long Answer

Evidence supporting the clusters

1. Length-dependent distal axonopathy. A “stocking–glove” pattern is anatomically compatible with distal axonal injury, where long axons may be especially vulnerable to metabolic, mitochondrial, microvascular, or toxic stress. Direct human evidence in the supplied record comes from thalidomide: seven exposed patients developed sensory-predominant distal axonal neuropathy, and symptom, examination, and neuropathy-score severity correlated with cumulative dose (reported r values 0.59–0.83); biopsies showed axonal/Wallerian-like degeneration with little inflammation. This supports toxic axonopathy, but the study was small, uncontrolled, and confounded by underlying disease and co-medications. Human dose-related toxic axonopathy

2. Non-length-dependent, ganglionopathy-like SFN. Proximal as well as distal intraepidermal nerve-fiber loss, including a potentially informative leg:thigh IENFD ratio, supports injury centered in small dorsal-root-ganglion neurons rather than a purely dying-back distal axon. In pooled case series of 265 patients, autoimmune disease was reported in 21%, glucose dysmetabolism in 8.6%, and 53.9% were classified as idiopathic; SCN9A, SCN10A, and SCN11A variants were also associated with this phenotype. The review explicitly describes these criteria as provisional, and its narrative synthesis is vulnerable to heterogeneous definitions and publication bias. Ganglionopathy-like non-length-dependent SFN

3. Pain-gain and mixed loss/gain phenotypes. Quantitative sensory testing separates reduced detection or pain thresholds (“loss”) from hyperalgesia, allodynia, or enhanced temporal summation (“gain”). In the reported cohort of more than 1,200 patients with neuropathic pain syndromes, approximately half showed both loss and gain, about 26% loss alone, and about 20% gain alone. These profiles may reflect combinations of peripheral nociceptor injury/regeneration, abnormal ion-channel excitability, ectopic activity, and central nociceptive amplification; however, QST is psychophysical and cannot by itself localize abnormalities to peripheral nerve, dorsal-root ganglion, spinal cord, or brain. Loss/gain sensory phenotyping

4. Immune–neural amplification and autonomic-predominant expression. In systemic autoimmune disease, the supplied review describes peripheral neuroimmune signaling, autonomic dysregulation, glial activation, and central sensitization as interacting processes that can amplify pain and autonomic symptoms beyond the amount of structural fiber loss. This is a mechanistic synthesis rather than new primary evidence; animal-to-human translation and standardized biomarkers remain limited. Peripheral–central neuroimmune crosstalk

How the mechanisms may map onto phenotype

• Distal sensory/autonomic loss: distal axonal degeneration or impaired regeneration; metabolic, toxic, or protein-deposition injury are plausible etiologic categories, but the supplied evidence directly establishes toxicity most clearly for thalidomide.
• Patchy, proximal, or non-length-dependent symptoms: a DRG-centered process, immune-mediated neuronal injury, or inherited sodium-channel dysfunction is more compatible than a simple length-dependent dying-back neuropathy.
• Pain-dominant gain: surviving or regenerating nociceptors may become hyperexcitable, while neuroimmune signaling and central sensitization can increase pain responses even when fiber-density loss is modest.
• Mixed phenotype: axonal loss and pathological excitability can coexist; therefore, preserved fiber density does not necessarily exclude functional small-fiber disease, and reduced density does not establish the mechanism of pain.

Important uncertainty

These are mechanistic interpretations, not validated biological subtypes. The supplied literature does not provide a unified longitudinal cohort linking molecular cause, skin biopsy, autonomic testing, QST, genetics, and treatment-independent outcomes. The conclusion would be weakened if large prospective studies found that topography, QST loss/gain profiles, and proximal/distal IENFD patterns did not reproducibly separate etiologies or predict distinct molecular signatures. Confidence: moderate for the existence of clinically recognizable phenotypes; low-to-moderate for assigning each phenotype to a single pathophysiological mechanism.