Objective : In the present study, the effect of electroacupuncture (EA) applied to several acupoints and non-acupoint in CFA-induced knee arthritis was examined. Methods : A common source of persistent pain in humans is arthritis. Arthritis was induced by injection of CFA $125;{mu}l$ into knee joint cavity under enflurane anesthesia. The rat subsequently showed swelling of the ankle and a reduced stepping force of the affected limb for the next several days. The reduced stepping force of the limb was presumably due to a painful knee. EA was applied to either of $LR_2,;LI_4$, or non-acupoint on the contralateral forelimb for 30 min under gaseous anesthesia. After the termination of EA, behavioral tests measuring stepping force were periodically conducted during the next 4 h. Results : EA applied to $LR_2$ point produced a significant improvement of stepping force of the affected foot lasting for at least 2 h. However, neigher $LI_4$ point nor non-point produced any significant increase of weight bearing force. The improvement of stepping pressure was interpreted as an analgesic effect. The analgesic effect was specific to the acupuncture point since the analgesic effect on CFA-induced knee arthritic pain model could not be mimicked by EA applied to a point, $LI_4$ or non-acupoint. In addition, both NO production and iNOS protein expression increased by arthritis were suppressed by EA applied to $LR_2$ point. Conclusion : These data suggest that EA produces a potent analgesic effect in the rat model of CFA-induced knee arthritis. This analgesic effect is produced by applying EA to an acupoint at opposite side from the painful area in a stimulus point-specific way.