The present study was conducted to evaluate the local antinociceptive actions of fluoxetine, a selective serotonin reuptake inhibitor, and the possible involvement of the l-arginine/NO/cGMP/K channel pathway in this effect using the formalin test in rats. To elucidate the underlying mechanisms, animals were pre-treated with l-NAME, aminoguanidine, methylene blue, glibenclamide, l-arginine, sodium nitroprusside, or diazoxide. Local ipsilateral, but not contralateral, administration of fluoxetine (10-300 μg/paw) dose-dependently suppressed flinching number during both early and late phases of the test, and this was comparable with morphine also given peripherally. Pre-treatment with l-NAME, aminoguanidine, methylene blue, or glibenclamide dose-dependently prevented fluoxetine (100 μg/paw)-induced antinociception in the late phase. In contrast, administration of l-arginine, sodium nitroprusside, and diazoxide significantly enhanced the antinociception caused by fluoxetine in the late phase of the test. However, these treatments had no significant effect on the antinociceptive response of fluoxetine in the early phase of the formalin test. Our data demonstrate that local peripheral antinociception of fluoxetine during the late phase of the formalin test could be due to activation of l-arginine/NO/cGMP/K channel pathway. The peripheral action of fluoxetine raises the possibility that topical application of this drug (e.g., as a cream, ointment, or jelly) may be a useful method for relieving the inflammatory pain states.