Adverse Events in the Treatment of Periprosthetic Fractures Around the Knee - a Clinical and Radiological Outcome Analysis

Background The incidence of periprosthetic fractures associated with total knee arthroplasty (PpFxK) has been reported to be 0.3-5.5%. 40% of all cases are related to revision TKA. The most common localisation is the distal femur. Classification is performed according to Rorabeck (RB). RB I-II fractures are usually treated with locked plating and retrograde intramedullary nailing, whereas RB III fractures are an indication for revision arthroplasty using a hinged endoprosthesis. PpFxK of the patella can be classified according to Goldberg and PpFxK of the proximal tibia can be grouped as in Felix. Interprosthetic fractures can be regarded as a special type of PpFx. Due to the increasing numbers of TKA being performed, increasing numbers of adverse events in arthroplasty can be expected. Adverse events in the treatment of PpFxK occur in up to 41% of patients according to the literature and revision is needed in approximately 29% of all cases. Risk factors are age, osteoporosis, infection, malalignment, osteolysis/loosening of the implant and status post revision. Patients A clinical and radiographic follow-up was performed with 50 patients (14 men, 36 women) treated for PpFxK of the femur, tibia and patella between 2011 and 2015 at the department of arthroplasty at a level 1 trauma center in Europe. Results The follow-up of all patients was 68%, with an average of 19.114.6 (1-49) months between PpFxK and clinical follow-up. 16% of the patients were allocated for further treatment or revision surgery from other hospitals. The patients median age was 78.08.8 (55-94) years. Most patients were affected by several orthopaedic and internal medical comorbidities. PpFxK classified as RB II were the most common fractures (60%, n=30). PpFxK usually occurred 5.04.8 (0-20) years after index TKA (primary or revision TKA), mostly in patients with CR-retaining endoprosthesis, whereas PpFxK according to Felix occurred significantly earlier and mostly in hinged TKAs. Patients achieved on average a mean Oxford Knee Score of 31.19.9 (14-46) points. The functional Knee Society Score (KSS) was 52.624.4 (20-100) and the mean KSS was 58.7 +/- 26.8 (0-99) points (n=25). Radiographic evaluation of the RB I-II patients showed frontal and sagittal malalignment in 20.6% of all cases after reduction and plate fixation. The overall rate of surgical adverse events was 50%; 44% of all RB patients needed revision surgery. Adverse events comprised non-union, failure of osteosynthesis, infection, wound healing disorders and re-fractures in the RB II and the Felix subgroup. Conclusion PpFxK are severe injuries and are associated with a high rate of adverse events related to treatment. Patients often have a complex background and a history of revision surgery or periprosthetic joint infection. The treatment of PpFxK should therefore take place at a centre with expertise in traumatology as well as in revision arthroplasty. Preoperative infection diagnostic testing as well as adequate imaging (X-rays and CT) are essential. We furthermore advise early evaluation of revision arthroplasty, especially in elderly patients suffering from PpFxK with insufficient bone quality around the TKA and closeness between fracture and TKA. In the case of plate fixation, it is important to give attention to correct reduction - to prevent non-union, loosening of the implant and failure of the osteosynthesis - as well as to consider double plating.