[Objective] To compare the clinical efficacy of arthroscopic release combined with manual release versus manual release only under anesthesia for advanced joint stiffness following total knee arthroplasty (TKA). [Methods] A retrospective study was performed on 40 patients who received release for advanced knee stiffness after TKA from January 2017 to August 2021. According to doctor-patient communication, 19 patients received arthroscopic release combined with manual release (the combined group), while the other 21 patients received manual release alone. The documents regarding perioperative period, follow- up and images were compared between the two groups. [Results] All patients in both groups were operated smoothly. Although the combined group consumed significantly longer operative time [(65.4±12.4)min vs (27.7±5.3)min, P＜0.05] than the manual group, the former got significantly greater maximum knee flexion [(105.4± 12.8)° vs (97.4±11.6)°, P＜0.05] and knee extension [(0.8±1.9)° vs (3.7±2.5)°, P＜0.05] than the latter intraoperatively. All patients were fol- lowed up for a mean of (26.7±6.9) months, and there was no significant difference in the time to resume full weight-bearing activity between the two groups (P＞0.05). The VAS score for pain decreased significantly (P＜0.05), while the KSS clinical and functional scores, HSS score as well as knee extension-flexion range of motion (ROM) increased significantly in both groups over time (P＜0.05). There was no significant difference in the above indexes between the two groups before surgery (P＜0.05), but the combined group proved significantly superior to the manual group in terms of HSS score [(79.8±5.1) vs (74.2±4.3), P＜0.05], KSS clinical score [(88.6±3.4) vs (83.9±3.2), P＜0.05], KSS function score [(85.7±3.5) vs (82.1±3.3), P＜0.05], and knee ROM [(93.4±6.5)° vs (87.2±6.3)°, P＜0.05] at the latest follow-up. Radiographically, fem- orotibial angle (FTA), states of prosthesis loosening, and ectopic ossification remained unchanged over time in both groups (P＞0.05), which were not significantly different between the two groups at any corresponding time points (P＞0.05). [Conclusion] The arthroscopic release combined with manual release does effectively improve the knee range of motion with good clinical efficacy and low complications for the knee stiffness following TKA.