OBJECTIVE To evaluate causes and clinical management of postoperative pain after total knee arthroplasty (TKA). METHODS From January 2004 to June 2009, 41 patients (44 knees) with knee pain post TKA were treated. There were 9 male and 32 female patients aging from 51 to 84 years with a mean of 63.5 years. The diagnosis followed to Brown diagnostic system. One case of extraarticular pain was complex regional pain syndrome type 1 (CRPS-1) and underwent conservative treatment, the remaining 5 cases were treated by surgery. Three cases of joint instability, 1 case of patellar baja, 2 cases of soft tissue impingement caused by overhang of the prosthesis, 1 case of popliteal tendon impact underwent conservative treatment, the other 27 cases underwent surgical intervention. The patients were followed up and the Knee Society Score (KSS) knee score, pain visual analog scale (VAS) score were recorded. RESULTS Forty-one cases were followed up for 1 to 6 years. At the last time of follow-up, the 5 cases received surgical treatment to extra-articular pain showed VAS score as 2.5 ± 0.2, KSS clinical and functional score as 92.8 ± 2.6 and 89.0 ± 3.4. There was significantly difference compared with preoperative (P < 0.05). One case of CRPS-1 performed conservative treatment, the therapy was effective. In the infected 12 cases of intra-articular pain, 1 case received amputation, 3 cases received antibiotic bone cement insert, 8 cases received two stage revision. All infections were cured, and VAS score was 3.8 ± 0.2, KSS clinical score was 88.3 ± 4.6, function score was 85.0 ± 4.6 postoperatively, with significantly difference compared with preoperative (P < 0.05). In the 8 cases received conservative treatment in non-infected group, at the last time of follow-up, VAS score was 4.5 ± 0.4, KSS clinical and functional score was 85.4 ± 4.2 and 84.2 ± 2.3, with significantly difference compared with preoperative (P < 0.05). Fifteen cases underwent surgical treatment, at the last time of follow-up, VAS score was 3.4 ± 0.1, KSS clinical and functional score was 86.6 ± 5.4 and 87.1 ± 2.4, with significantly difference compared with preoperative (P < 0.05). CONCLUSIONS Patients with knee pain post TKA need systematic assessment to identify the causes. Appropriate treatment due to the positive diagnosis generally lead to satisfactory results, surgical intervention with indefinite causes is strictly prohibited.