The reasons humans develop diabetic neuropathy are not known. It seems likely that, as in animals, the polyol-myoinositol-aldose reductase controlled pathways are relevant at an early stage and are related to hyperglycemia. Perhaps these metabolic factors influence the structural and dynamic aspects of the microvasculature which are so clearly abnormal in established diabetic neuropathy. A demonstration that these metabolic factors in some way affect the vasculature would be important, for that would allow some logic in administering inhibitors to prevent vessel changes of such a pathological nature. At the moment, the case for the long-term administration of aldose reductase inhibitors to prevent nerve damage is not proven, and the decision to administer them will cause some difficulty to physicians. Unfortunately, the situation is more complex than this simple aldose reductase-microvascular hypothesis, for consideration must be given to the known glycation of nerve proteins, the involvement of fatty acid metabolism within the vasculature, and the undoubted role of growth factors.