Abstract. The purpose of the present study was to explore the efficacy of arthroscopic loose body removal for knee osteoarthritis (KOA). A total of 23 patients with KOA were enrolled and randomly received conservative treatment (conservative group; n=10) or loose body removal surgery (surgery group; n=13). The serum levels of disease activity indices, including hypersensitive C-reactive protein (hs-CRP), erythrocyte sedimentation rate (ESR) and synovial inflammatory factors [interleukin (IL)-1 and IL-6] were detected prior to surgery, and at 4 days, 2 or 4 weeks after surgery. All patients were followed up for 2 years and the cure rate was estimated. No significant difference was identified in pre-operative plasma levels of hs-CRP and ESR as well as the synovial concentration of IL-1 and IL-6 between the two groups (all P>0.05). At 2 and 4 weeks after treatment, the levels of these parameters in the surgery group were significantly lower than those in the conservative group (all P<0.05), although the maximum value of these parameters was higher in the surgery group than in the conservative group at 4 days after surgery. The cure rate for KOA in the surgery group was significantly higher than that in the conservative group. In conclusion, the results demonstrated that arthroscopic loose body removal is a more effective treatment than conservative therapy for KOA.