To (1) robustly characterize the clinical spectrum of small fiber neuropathy (SFN) using quantitative, data-driven phenotyping to capture its full multidimensional symptom burden, and (2) identify reproducible, biologically meaningful patient subgroups that can inform mechanistic study and clinical treatment. 

SFN is a progressive, disabling disorder affecting small peripheral nerve fibers. While SFN is classically defined by distal neuropathic pain and sensory loss, many patients experience additional symptoms - including myalgias, fatigue, subjective weakness, and neuropathic itch.  However, the prevalence and intensity of these symptoms remains poorly defined. Current treatments—focused primarily on pain—are often ineffective, possibly reflecting the clinical heterogeneity of the disorder.

Demographic, clinical, and laboratory data were analyzed from patients with skin biopsy-confirmed SFN (n = 203) and healthy controls (n = 30). Patients rated symptom intensity (0–10) across sensory, motor, and fatigue domains. Unsupervised clustering was applied to identify data-driven subgroups. Cluster reproducibility was quantified using the Adjusted Rand Index (ARI).

The clustering model demonstrated high stability (ARI = 0.80). Fatigue—not neuropathic pain—was the most prevalent and severe symptom in SFN. Unsupervised analysis identified three reproducible clinical phenotypes. The algesic group (~20%) exhibited severe neuropathic pain with high co-occurrence of fatigue and myalgias. The myalgic group (~60%) demonstrated predominant fatigue and muscle-related symptoms that exceeded pain in both prevalence and severity. The pauci-symptomatic group (~20%) reported few symptoms of mild-to-moderate intensity. Clusters differed significantly in total symptom burden, disease duration, intraepidermal nerve fiber density, sex distribution, and body mass index.

This study challenges the pain-centric paradigm of SFN revealing fatigue and muscle symptoms as dominant and meaningful dimensions of disease. The identification of robust, data-driven phenotypes provides a foundation for mechanistic stratification, biomarker discovery, and development of precision therapies in SFN and related neuropathies.