Aims: The involvement of NO modulation in tramadol's mechanism of action is demonstrated indirectly by the alteration of the effect with the use of nitric oxide synthase inhibitor nitroG-L-arginine methyl ester (L-NAME). We aimed to investigate the alterations in serum and spinal cord NO levels following tramadol used alone or in combination with L-NAME or specific neuronal nitric oxide synthase (NOS) inhibitor 7-nitroindazole (7-NI), beside the changes in tail immersion reponses. Methods: Seventy-seven rats were randomly allocated into 7 groups. Group 1: sham group; Group 2: no drug; Group 3: tramadol 10 mg kg-1; Group 4: L-NAME 5 mg kg-1 + tramadol 10 mg kg-1; Group 5: L-NAME 10 mg kg-1 + tramadol 10 mg kg-1; Group 6: 7-NI 50 mg kg-1 + tramadol 10 mg kg -1; Group 7: 7-NI 100 mg kg-1 + tramadol 10 mg kg -1. Apart from Group 1, all the animals were exposed to a tail immersion test before medications (baseline) and 15 min after the tramadol injection. Following the tail-immersion test, a blood sample was taken from the heart, and the lower segments of the spinal cord were extracted. Results: Tramadol increased tail immersion latency (p < 0.05). The addition of L-NAME or 7-NI respectively decreased (p < 0.05) and increased (p > 0.05) tail immersion latency. Exposure to nociceptive stimulus caused a decrease in plasma NO levels; administration of tramadol caused an approximation to the sham group values; and a decrease of the levels followed the addition of L-NAME or 7-NI. Conclusion: In conclusion, the changes in tail immersion latency and plasma and spinal cord NO levels induced by the use of tramadol alone or in combination with NOS inhibitors indicate that the modulation of NO is involved in the mechanism of the antinociceptive effect of tramadol. © 2006 VSP.