Background: Curcumin has been reported to have anti-inflammatory, antioxidant, antiviral, antifungal, antitumor, and antinociceptive activity when administered systemically. We investigated the analgesic efficacy of intrathecal curcumin in a rat model of inflammatory pain. Methods: Male Sprague Dawley rats were prepared for intrathecal catheterization. Pain was evoked by injection of formalin solution (5%, $50{mu}l$) into the hind paw. Curcumin doses of 62.5, 125, 250, and $500{mu}g$were delivered through an intrathecal catheter to examine the flinching responses. The $ED_{50}$ values (half-maximal effective dose) with 95% confidence intervals of curcumin for both phases of the formalin test were calculated from the dose-response lines fitted by least-squares linear regression on a log scale. Results: In rats with intrathecal administration of curcumin, the flinching responses were significantly decreased in both phases. The slope of the regression line was significantly different from zero only in phase 2, and the $ED_{50}$ value (95% confidence interval) of curcumin was $511.4{mu}g$ (23.5-1126.5). There was no apparent abnormal behavior following the administration of curcumin. Conclusions: Intrathecal administration of curcumin decreased inflammatory pain in rats, and further investigation to elucidate the precise mechanism of spinal action of curcumin is warranted.