Rats developed tactile allodynia within days of the onset of diabetes and which persisted for up to 8 weeks. Allodynia was prevented by insulin therapy that maintained normoglycemia while established allodynia was reversed by insulin therapy and normoglycemia of days but not hours duration. Tactile allodynia persisted in diabetic rats that received enough insulin to maintain normal body and foot weights but remained hyperglycemic, whereas this therapy was sufficient to correct other nerve disorders in diabetic rats, including deficits of sensory and motor nerve conduction velocity, nerve blood flow and hyperalgesia during the formalin test. Treating diabetic rats with the aldose reductase inhibitor ICI 222155 did not prevent tactile allodynia. Tactile allodynia was of similar magnitude in diabetic rats and nerve injured control rats and diabetes did not alter the magnitude or time course of nerve injury-induced allodynia. Systemic lidocaine treatment alleviated tactile allodynia in nerve injured control rats and both sham-operated and nerve injured diabetic rats. The streptozotocin-diabetic rat develops tactile allodynia that appears to be related to prolonged periods of insulin deficiency or hyperglycemia and which is amenable to treatment with lidocaine. The model may be of use in investigating the efficacy of other potential therapeutic agents for treating painful diabetic neuropathy.