Introduction. While knee arthrodesis is a salvage option for recalcitrant total knee arthroplasty (TKA) periprosthetic joint infection (PJI) it is used relatively uncommonly and contemporary data are limited. We sought to determine the reliability, durability and safety of knee arthrodesis as the definitive treatment for complex, persistently infected TKA in a modern series of patients. Methods. We retrospectively identified 41 knees treated from 2002–2016 with a deliberate, two-stage knee arthrodesis protocol (TKA resection, high-dose antibiotic spacer, targeted IV antibiotics and followed by subsequent knee arthrodesis) in patients with complex TKA PJI. Mean age was 64 years & mean BMI was 39 kg/m. 2. Mean follow-up was 4 years. The extensor mechanism was deficient in 66% of knees, and flap coverage was required in 34% of knees. The majority of patients were host grade B (56%) or C (29%), and extremity grade of 3 (71%). Twenty-nine percent had poly-microbial infections, and 49% had multi-drug resistant organisms. Fixation included intramedullary nail (61%), external fixator (24%), and dual plating (15%). Results. Two patients (5%) required amputation for persistently infected non-unions; therefore, limb salvage was accomplished in 95% of patients. After initial treatment, there were non-unions in 24% and persistent infection in 17%. Non-union was significantly correlated with persistent infection, with 50% of non-unions having persistent infection compared with just 6% of united knees (p=0.006). External fixation was a significant risk factor for non-union (70%) compared to intramedullary fixation (8%; p=0.005). Overall, twenty-seven complications occurred in 20 patients and 31% required reoperation other than external fixator removal. Intramedullary fixation led to a 90% rate of both infection control and radiographic union. Conclusion. Two-stage knee arthrodesis using a deliberate protocol (resection, high-dose abx spacer, targeted IV abx, and subsequent arthrodesis) ultimately achieved successful limb salvage in 95% of patients with complex infected TKA. One or more complications occurred in nearly half the patients and reoperation was required for 1-in-3. That substantial risk of complications is not surprising as this large contemporary series included complex, worst-case infected TKA in which: 2/3 had disrupted extensor mechanism, 1/3 required flap coverage, and the majority had poly-microbial or multi-drug resistant organisms. Summary. For contemporary patients with very complex, infected TKA a two-stage knee arthrodesis was reliable in achieving limb salvage (95%) at the cost of a high initial complication and reoperation rate. For figures, tables, or references, please contact authors directly

We retrospectively evaluated eight patients who underwent arthrodesis of the knee using cannulated screws. There were six women and two men, with a mean age of 53 years. The indications for arthrodesis were failed total knee arthroplasty, septic arthritis, tuberculosis, and recurrent persistent infection. Solid union was achieved in all patients at a mean of 6.1 months. One patient required autogenous bone graft for delayed union. One suffered skin necrosis which was treated with skin grafting. The mean limb-length discrepancy was 3.1 cm. On a visual analogue scale, the mean pain score improved from 7.9 to 3.3. According to the Knee Injury and Osteoarthritis Outcome score quality of life items, the mean score improved from 38.3 pre-operatively to 76.6 at follow-up. Cannulated screws provide a high rate of union in arthrodesis of the knee with minimal complications, patient convenience, and a simple surgical technique.