BACKGROUND: Activated N-methyl-D-aspartate (NMDA) receptor is involved in the formation of chronic neuropathic pain, and its antagonist, ketamine, exhibits effective amelioration of diabetic neuropathic pain (DNP). However, the mechanisms of NMDA receptor participation in the formation and maintenance of DNP remain poorly understood. OBJECTIVE: To evaluate the role NMDA receptor plays in DNP and effects on p38 mitogen activated protein kinase (p38 MAPK) in a rat model of DNP.DESIGN, TIME AND SETTING: A randomized, controlled, animal experiment was performed at the Human Embryonic Stem Cell Research Institute of Yunyang Medical College Affiliated Taihe Hospital between July 2005 and September 2007.MATERIALS: Streptozotocin was provided by Sigma, USA; p38 MAPK inhibitor (SB203580) was provided by Shanghai KangChen Biotech, China; NMDA receptor antagonist (MK-801) was purchased from Shanghai Yope Biotech, China.METHODS: A total of 128 healthy, Wistar rats of clean grade, aged 3 months and weighing 180- 220 g, were randomly assigned to 4 groups: control, DNP model, p38 MAPK, and NMDA receptor. Each group contained 32 rats. DNP was established in all groups except for the control group by intraperitoneal injection of streptozocin (65 mg/kg). Subsequently, 1 mg/kg SB203580 and 1 mg/kg MK-801 were injected once each week via intraperitoneal injection in the p38 MAPK and NMDA receptor groups, respectively. MAIN OUTCOME MEASURES: At the end of 2, 4, 6, and 8 weeks following streptozotocin injection, mechanical withdrawal threshold was measured in 8 animals from each group following von Frey filament stimulation. The rats were anesthetized and nerve conduction velocity of the left sciatic nerve was measured. Subsequently, the right sciatic nerve, the lumbar segment of the spinal cord, and dorsal root ganglia were removed from the L_(3-6) segment for microscopic examination. p38 MAPK expression was determined using immunohistochemistry and West blot analysis. Expression of NMDA receptor 1 mRNA in dorsal root ganglion and spinal cord neurons was detected using RT-PCR. RESULTS: Mechanical withdrawal threshold and nerve conduction velocity were significantly reduced, and p38 MAPK and NMDA receptor 1 mRNA expression in the spinal cord and dorsal root ganglia were significantly increased, in the model, p38 MAPK, and NMDA receptor groups compared with the control group at all time points (P < 0.05). At 4-8 weeks following successful DNP model establishment, SB203580 and MK-801 increased mechanical withdrawal threshold, accelerated nerve conduction velocity, and attenuated p38 MAPK expression, compared with the model group. The NMDA receptor group exhibited downregulated mRNA expression of NMDA receptor 1 compared with the model and p38 MAPK groups (P < 0.05). CONCLUSION: NMDA receptor was highly expressed in the brains of DNP rats and was involved in DNP development via activation of the p38 MAPK signal pathway.