Cutaneous innervation in Guillain-Barre syndrome: pathology and clinical correlations

Abstract
Guillain–Barré syndrome (GBS) is traditionally considered to be a large‐fibre neuropathy. However, the presence of hypo‐aesthesia, dysaesthesia and dysautonomia in GBS patients raises the possibility that small‐diameter sensory and autonomic nerves may also be affected. To investigate small‐fibre neuropathy in GBS, we performed a skin biopsy from the distal leg of 20 patients with the demyelinating form of GBS. Skin sections were immunohistochemically stained with antiserum against protein gene product 9.5 (PGP 9.5), a ubiquitin C‐terminal hydrolase. Cutaneous innervation was evaluated by measuring epidermal nerve density (END), and END was further correlated with various clinical and electrophysiological parameters. In GBS patients, END values were much lower than in age‐ and gender‐matched control subjects (5.03 ± 1.18 versus 10.16 ± 0.87 fibres/mm, P < 0.001). Eleven patients (55%) had reduced epidermal innervation with pathological evidence of active nerve degeneration in the dermis: fragmentation of subepidermal nerve plexuses and a beaded appearance of dermal nerves. GBS patients had significantly elevated thermal thresholds with higher warm threshold temperatures (44.54 ± 1.04 versus 39.00 ± 0.35°C, P < 0.001) and lower cold threshold temperatures (25.57 ± 1.11 versus 29.05 ± 0.21°C, P = 0.032). Reduced END values were associated with an elevated warm threshold (P = 0.027), ventilatory distress (P = 0.037) and dysautonomia (P = 0.001). END values were negatively correlated with disability grade on a scale of 1–6 (slope –0.134 ± 0.038, P = 0.0018). Patients with reduced END values tended to have a slower recovery than those with normal END values (P = 0.013, median time 12 versus 2 weeks). Patho logically, sudomotor innervation of the skin was reduced in five of 17 (29.4%) GBS patients in whom sweat glands could be recognized. These findings suggest that small‐fibre sensory and autonomic neuropathies exist in a significant proportion of GBS patients, and that END values are correlated with functional disabilities. In summary, GBS should be considered a global neuropathy instead of a pure large‐fibre neuropathy.