Objective To explore the methods, fixation points, and effectiveness of staged therapy using external fixation frame in treatment of infectious nonunion near knee joint. Methods A retrospective analysis was conducted on the clinical data of 60 patients with infectious nonunion near knee joint, who underwent staged therapy using external fixation frame between June 2021 and June 2024 and were followed up. There were 48 males and 12 females with an average age of 47.9 years (range, 16-70 years). The disease duration ranged from 9 months to 20 years, with a median duration of 14 months. Among them, 21 cases of infectious nonunion located in the distal femur, 36 cases in the proximal tibia, and 3 cases in the patella; 12 cases exhibited segmental bone defects (≥4 cm), while 48 cases presented with localized bone defects (<4 cm). Osteomyelitis was classified using the Cierny-Mader system, with 3 cases classified as type Ⅰ, 6 cases as type Ⅱ, 35 cases as type Ⅲ, and 16 cases as type Ⅳ. Preoperative C-reactive protein levels ranged from 15.1 to 55.8 mg/L (mean, 36.4 mg/L). The erythrocyte sedimentation rate was 35-80 mm/1h (mean, 56.9 mm/1 h). The Hospital for Special Surgery (HSS) score for knee joint was 69.3±17.7 and the range of motion was (70.61±40.60)°. After debridement and placement of antibiotic carriers at the first-stage operation, unilateral orbital frames (n=14), combined frames (n=27), or Ilizarov frames (n=19) were used for cross joint fixation (n=9) or joint preservation fixation (n=51). After 6-8 weeks of infection control, the bone grafting or bone transport was performed at the second-stage operationbased on the type of bone defect, with internal fixation employed as an adjunct if necessary. After operation, the infection control and fracture healing were observed and the bone healing time was recorded. The knee joint function was assessed using the HSS score, and the knee joint range of motion was measured as well as the angle of motion loss. Patients were grouped according to the site of nonunion, type of external fixation frame, and fixation method. The bone healing time, change value of HSS score (difference between pre- and post-operation), and knee joint range of motion loss were compared between groups. Results All infection markers returned to the normal range within 6 weeks after the first-stage operation. All patients were followed up 12-48 months (mean, 22.0 months) after the second-stage operation. There were 5 cases of needle tract infection during the external fixation period, and 3 cases of infection recurrence after the second-stage operation, all of which were cured after symptomatic treatment. The bone healing time was 6-18 months (mean, 11.0 months). At last follow-up, the HSS score was 88.5±7.9 and the range of motion was (61.84±40.59) °, with significant differences compared to preoperative values (P<0.05); the knee joint range of motion loss was (8.77±11.07) °. The bone healing time was significant longer in the distal femur group than in the proximal tibia group (P<0.05), and the unilateral orbital frames group than in the Ilizarov frames group and the combination frames group (P<0.05). There was no significant difference in the change values of HSS score between groups (P>0.05). Conclusion During the first-stage operation, debridement is performed and antibiotic carriers are placed to control infection. External fixation frames are then precisely positioned based on the measured distance between the lesion and the joint surface, avoiding the infected wound while ensuring mechanical balance. During the second-stage operation, bone grafting options are selected according to the extent of bone defects to enhance the cure rate of infectious nonunion. Postoperative early functional exercises of the knee joint are permitted to improve joint function.