We have developed an animal model of diabetic sympathetic autonomic neuropathy which is characterized by neuroaxonal dystrophy (NAD), an ultrastructurally distinctive axonopathy, in chronic streptozotocin (STZ)-diabetic rats. Diabetes-induced alterations in the sorbitol pathway occur in sympathetic ganglia and therapeutic agents which inhibit aldose reductase or sorbitol dehydrogenase improve or exacerbate, respectively, diabetes-induced NAD. The sorbitol dehydrogenase inhibitor SDI-711 (CP-470711, Pfizer) is approximately 50-fold more potent than the structurally related compound SDI-158 (CP 166,572) used in our earlier studies. Treatment with SDI-711 (5 mg/kg/day) for 3 months increased ganglionic sorbitol (26-40 fold) and decreased fructose content (20-75%) in control and diabetic rats compared to untreated animals. SDI-711 treatment of diabetic rats produced a 2.5- and 4-5-fold increase in NAD in the SMG and ileal mesenteric nerves, respectively, in comparison to untreated diabetics. Although SDI-711 treatment of non-diabetic control rat ganglia increased ganglionic sorbitol 40-fold (a value 8-fold higher than untreated diabetics), the frequency of NAD remained at control levels. Levels of ganglionic sorbitol pathway intermediates in STZ-treated rats (a model of type 1 diabetes) and Zucker Diabetic Fatty rats (ZDF, a genetic model of type 2 diabetes) were comparable, although STZ-diabetic rats develop NAD and ZDF-diabetic rats do not. SDI failed to increase diabetes-related ganglionic NGF above levels seen in untreated diabetics. Initiation of Sorbinil treatment for the last 4 months of a 9 month course of diabetes, substantially reversed the frequency of established NAD in the diabetic rat SMG without affecting the metabolic severity of diabetes. These findings indicate that sorbitol pathway-linked metabolic alterations play an important role in the development of NAD, but sorbitol pathway activity, not absolute levels of sorbitol or fructose per se, may be most critical to its pathogenesis.