The analysis of epidermal nerve fibre spatial distribution improves the diagnostic yield of skin biopsy

AIM: Small fibre neuropathy (SFN) diagnosis represents a challenge for neurologists. The diagnostic gold standard is intraepidermal nerve fibre (IENF) density, but in about 10-20% of patients with symptoms/signs and abnormalities on functional tests, it remains within normal range. We propose an adjunctive parameter to improve the efficiency of skin biopsy diagnosis. METHODS: We recruited 31 patients with SFN symptoms/signs, normal nerve conduction study, abnormal quantitative sensory testing and normal IENF density. We also included 31 healthy controls and 31 SFN patients with reduced IENF density as control groups. RESULTS: We measured the distance between consecutive IENFs in the three groups. Mean inter-fibre distances did not differ between patients with normal counts and healthy controls (66.7 ± 14.5 μm vs. 76.7 ± 13.4 μm; P = 0.052), while the relative standard deviation was significantly (P < 0.001) higher in patients (79.3 ± 29.9) compared to controls (51.6 ± 12.2). Using ROC analysis, we identified an inter-fibre distance of 350 µm as the measure that better differentiated patients from controls (AUC = 0.85, sensitivity: 74%, specificity: 94%). At least one such segment was also observed in all patients with reduced IENF count. CONCLUSION: Irregular spatial distribution is an SFN intrinsic feature preceding actual nerve loss. The presence of a stretch of denervated epidermis longer than 350 µm is a parameter able to increase the diagnostic efficiency of skin biopsy.