P. K. Thomas

Acute painful diabetic neuropathy precipitated by strict glycaemic control.

SummaryA case of acute painful diabetic neuropathy that followed the establishment of strict glycaemic control using continuous subcutaneous insulin infusion is described. Sural nerve biopsy shortly after the onset of the acute painful syndrome showed no evidence of active nerve fibre degeneration; instead, the appearances were those of a chronic neuropathy with prominent regenerative activity. The suggestion is made that adequate diabetic control promoted regeneration and that the pain may have been related to the ectopic generation of impulses in regenerating axon sprouts.

Diabetic neuropathy: abnormalities of Schwann cell and perineurial basal laminae. Implications for diabetic vasculopathy.

During Wallerian degeneration, the Schwann cell basal laminal ensheathment around myelinated nerve fibres remains after the removal of myelin and axonal debris, forming a corrugated tube within which Schwann cell proliferation takes place. In nerve biopsies from patients with diabetic neuropathy, such residual basal laminal tubes tend to be circular rather than corrugated and appear to be more persistent during regeneration; this suggests increased rigidity and durability. These changes could be the result of increased cross–linkage of type IV collagen or alterations to other components of the basal lamina. A similar mechanism may be responsible for the thickening of perineurial basal laminae and the reduplication of basal laminae around endoneurial capillaries in diabetic patients; such reduplication may lead to reduced compliance of the vessel walls and impaired vascular perfusion.

Morphometry of endoneurial capillaries in diabetic sensory and autonomic neuropathy

SummaryNerve biopsies were obtained from 27 patients with diabetic neuropathy. All had a symmetric distal sensory and autonomic neuropathy or a purely sensory neuropathy. Mean age was 39.8 years (range 23–57 years). Two patients had Type 2 (non-insulin-dependent) diabetes mellitus and the remainder Type 1 (insulin-dependent) diabetes. Morphometric observations on endoneurial capillaries were compared with results from organ donor control cases and from patients with type 1 hereditary motor and sensory neuropathy. The area of the lumen of the capillaries did not differ between the three groups. The area occupied by the capillary endothelial cells in transverse section and the number of endothelial cell nuclei were increased both in the patients with diabetic neuropathy and hereditary motor and sensory neuropathy, as was the thickness of the surrounding basal laminal zone. ‘Closure’ of endoneurial capillaries in diabetic neuropathy, reported in another study, was not confirmed. Capillary density and nearest-neighbour distances were similar in the diabetic and organ donor control cases. Capillary density was reduced in the patients with hereditary motor and sensory neuropathy, this being related to increased fascicular area consequent upon the presence of hypertrophic changes. The presence of thickening of the pericapillary basal laminal zone and endothelial cell hyperplasia both in diabetic and hereditary motor and sensory neuropathy, the latter being a neuropathy in which a vascular basis can be discounted, makes it difficult to use such changes as an argument favouring a vascular cause for diabetic neuropathy. There were differences in the basal laminal zone between the diabetic and hereditary motor and sensory neuropathy cases suggesting that the reduplicated basal lamina was more persistent in the diabetic patients.

Abnormalities of the axonal cytoskeleton in giant axonal neuropathy.

SummaryIntermediate filaments accumulate abnormally in a variety of cell types in individuals with human inherited giant axonal neuropathy (GAN). A characteristic feature of this disorder is the occurrence of focal axonal enlargements filled with accumulations of neurofilaments. The minimum separations between neurofilaments in sural nerve axons of a patient with GAN were 12–30 nm compared with 24–60 nm in controls. The normal sidearm protrusions which cross-bridge adjacent filaments were rare in GAN. Average minimum neurofilament diameter was 12.4 nm in GAN compared with 10.1 nm in controls. Many axons were devoid of neurofilaments and contained an increased density of microtubules, many of which did not run longitudinally. This disorganization of microtubule alignment may reflect the lack of an associated neurofilament lattice. It is concluded that GAN involves abnormalities of neurofilament cross-linkage to one another and to adjacent microtubules. Mechanisms are discussed which could account for this inherited disorder of intermediate filament organization affecting various cell types.

Pattern of myelinated fibre loss in the sural nerve in neuropathy related to Type 1 (insulin-dependent) diabetes

SummarySural nerve biopsies were obtained from 17 diabetic patients with neuropathy. All patients except three had both a symmetric distal sensory and autonomic polyneuropathy related to Type 1 (insulin-dependent) diabetes mellitus; 3 patients had a purely sensory polyneuropathy. Mean age was 34.5 years (range 18–53 years). The biopsies were compared with specimens from an age-matched control series. Myelinated fibre loss in the diabetic nerves was found to be nonuniform. Although patchy fibre loss has been considered to favour a vascular basis, an identical pattern of nonuniform loss was observed in a series of sural nerve biopsies from patients with Type I hereditary motor and sensory neuropathy, a subgroup within the spectrum of peroneal muscular atrophy, mainly of autosomal dominant inheritance, and a condition in which a vascular causation can be discounted. Possible reasons for nonuniform fibre loss other than vascular disease are discussed.

Effects of glucocorticoids on experimental allergic neuritis

Corticosteroids were administered to rats and guinea pigs with experimental allergic neuritis, from the time of inoculation with antigen or from the onset of signs of disease. No statistically significant effects were observed in guinea pigs. In rats, to which large doses of corticosteroids were administered, disease severity was slightly but significantly reduced in both groups and recovery was more rapid in the animals treated from the time of induction of disease. These results were comparable with those obtained in trials of corticosteroids in acute inflammatory polyneuropathy in man, which have also not demonstrated any striking effects.

Glucose and leucine uptake by rat dorsal root ganglia is not insulin sensitive

Dorsal root ganglia from rats were incubated with 3-O-methyl-[14C]glucose, and [3H]leucine in the presence or absence of insulin in order to determine whether insulin influences the uptake of glucose and amino acids by the cells of the ganglion. No effect was detected. A significant proportion (38%) of the uptake of [3H]leucine was shown to be inhibited by ouabain and therefore energy dependent, utilizing Na+K(+)-ATPase. The activity of this enzyme is known to be impaired in dorsal root ganglia in diabetic rats, as is the uptake of amino acids; these phenomena are therefore unlikely to be due to a direct effect of reduced circulating insulin levels. The relevance of these findings to theories as to the causation of diabetic neuropathy is discussed.

Reduced susceptibility of T cell-deficient rats to induction of experimental allergic neuritis

Lewis rats were made deficient in T cells by adult thymectomy and lethal irradiation, and then reconstituted with T cell-free bone marrow. Their ability to develop experimental allergic neuritis (EAN) was compared with normal rats. The majority of T cell-deficient rats remained clinically and histologically unaffected, whereas all but one of the normal rats developed severe EAN. Those T cell-deficient animals which succumbed to EAN were found to have a significantly higher percentage of residual blood T lymphocytes than those which did not. Full susceptibility to EAN was restored by an inoculum of whole thoracic duct lymphocytes (TDL) from normal animals but not by TDL depleted of T cells. The results therefore provide direct confirmation that T cells are a requirement for the development of EAN.

Ultrastructural observations on myelinated fibres in experimental diabetes: Effect of the aldose reductase inhibitor ponalrestat given alone or in conjunction with insulin therapy

Six groups of rats were studied over a 12-week period: onset and end controls, untreated diabetics, ponalrestat-treated diabetics, insulin-treated diabetics, and diabetics treated with ponalrestat and insulin. The concentrations of glucose, sorbitol and fructose significantly increased and that of myo-inositol significantly decreased in the sciatic nerve of untreated diabetic animals. Ponalrestat administration completely normalized sorbitol levels and partially corrected fructose and myo-inositol concentrations without altering nerve glucose levels. The biochemical abnormalities were also corrected in both the insulin-treated and insulin and ponalrestat-treated diabetic animals. Myelinated fibre cross-sectional areas and axonal areas were significantly less in the tibial nerve of diabetic animals as compared with age-matched controls. Insulin treatment partially corrected the reduction in fibre and axonal area but teased fibre preparations showed an excess of axonal degeneration as compared with controls, untreated diabetics and ponalrestat-treated diabetics. Ponalrestat given alone or in conjunction with insulin therapy did not correct the reduction in fibre or axonal area and single isolated fibres from diabetic animals treated with ponalrestat and insulin showed a marked excess of axonal degeneration, probably related to hypoglycaemia. The study fails to reveal any significant beneficial effect of aldose reductase inhibition on the structural abnormalities in peripheral nerve in experimental diabetes.

Freeze-fracture observations on normal and abnormal human perineurial tight junctions: alterations in diabetic polyneuropathy

SummaryPerineurial cells in the human sural nerve possess tight junctions which in freeze-fracture replicas are seen to be composed of networks of branching and anastomosing P face strands and E face grooves. Isolated circular tight junctions (maculae occludentes) may represent attachment devices between adjacent perineurial lamellae. At the overlapping margins of the cells, a beltlike tight junction (zonula occludens) encircles the cells and is believed to comprise a paracellular diffusion barrier. As the permeability of the perineurium has been found to be altered in diabetic polyneuropathy, the zonulae occludentes have been studied. In freeze-fracture replicas from cases of diabetic polyneuropathy a mixed population of structurally normal and abnormal junctions was observed. In some, the strands were abnormally curved with reduced numbers of intersections, the intervening plasma membrane displaying prominent P face concavities and E face convexities. At other sites, the junctions were severely disorganized and represented by fragmented and isolated strands with few intersections and numerous free ends. These abnormalities resemble changes that have been produced experimentally in epithelial tight junctions by osmotic damage. The possibility is considered that similar mechanisms could result in the alterations of the perineurial tight junctions in diabetic polyneuropathy and account for its impaired permeability barrier properties.