The antinociceptive effect of alpha-trinositol was examined in rats using the formalin test following systemic, spinal and local subcutaneous administration. Injection of formalin into the paw evoked two phases (phase 1: 0-9 min; phase 2: 10-60 min) of flinching behavior of the injected paw. Intrathecal administration of alpha-trinositol resulted in a dose-dependent suppression of the first (ED(50): 8 mu g) and second (ED(50): 9 mu g) phase of formalin-evoked behavioral response. Similarly, intraperitoneal delivery showed a dose-dependent reduction of the first (ED(50): 83 mg/kg) and second (ED(50): 56 mg/kg) phase of the formalin test. Subcutaneous injection of 100 mu g, but not 10 mu g, alpha-trinositol into the rat paw together with the formalin solution, had no effect on the first phase, but reduced by 20% the second phase of behavior. These data show that alpha-trinositol produces a suppression of acute and prolonged nociceptive behaviors with a central mechanism of action, although some peripheral component may contribute to the reduction of the late phase following systemic administration.