Immunohistochemical measurements of nerves and neuropeptides in diabetic skin: relationship to tests of neurological function

Abstract

Image-analysis was used to measure nerves immunoreactive to the general neuronal marker protein gene product 9.5 (PGP 9.5-IR) and the neuropeptides calcitonin gene-related peptide and vasoactive intestinal polypeptide in standardised leg skin biopsies of three age-matched groups of young subjects: non-diabetic (n=14), diabetic patients with normal small fibre function ("non-neuropathic", (n=11) and diabetic patients with abnormal small fibre function ("neuropathic", n=11). Depletion of nerves and neuropeptides was most marked in the epidermis, where calcitonin gene-related peptide-immunoreactivity was more frequently absent than PGP 9.5-IR in diabetic patients. Epidermal PGP 9.5-IR nerve area and counts were reduced in neuropathic compared with normal subjects (p<0.001), as were epidermal calcitonin gene-related peptide nerve counts (p=0.003). Sweat gland PGP 9.5 and vasoactive intestinal polypeptide, which may be involved in sweat production, showed no diminution in diabetic patients (area: p=0.160, p=0.372 by ANOVA). Two diabetic patients showed elevated sweat gland PGP 9.5-IR and three had increased sweat gland vasoactive intestinal polypeptide; this may represent nerve proliferation. In local sweat tests, acetylcholine-stimulated sweat output was associated with increased immunoreactivity, while the sympathetic skin response showed inverse correlations with immunoreactivity. There were no consistent changes with other commonly-used neurophysiological tests. HbA1 correlated negatively with immunohistochemical measurements. Neuropeptide changes were seen in the absence of macro- and microvascular disease, and epidermal nerve depletion occurred in patients with normal thermal thresholds and cardiac autonomic function. Immunohistochemical measurement of cutaneous nerves in skin biopsies is a practical method for assessing peripheral small fibres in diabetes, and one which could be repeated in longitudinal studies. © 1992 Springer-Verlag.