Symposium: Fractures périprothétiques autour de la hanche et du genou

Abstract

The growing incidence of periprosthetic fractures is mainly due to the increase in the number of prosthesis implantations. The goal of this SOFCOT symposium was to establish survey data of such fractures over a continuous 15-year period. Two multicentric studies were conducted with the participation of 13 public hospitals in France. The first involved a retrospective period from 1990 to 2003 and the second prospective data collected in 2004. Analysis criteria focused on patient, fracture type according to the Vancouver classification modified by SOFCOT for hip injuries, and an original classification for knee injuries. Surgical procedures, peri-operative status, bone union were the basic factors used to assess the fracture condition at 6 month follow-up. At last follow-up, specific attention was focused on the implant fixation and revision procedures if needed. Hip periprosthetic fractures were noted in 580 retrospective cases (mean age 77 years, mean followup 4 years) and in 115 prospective cases (mean age 78.9 years, and minimal follow-up 6 months). Retrospective study fractures were type A in 8.3%, type B in 60.7%, type C in 31%. Treatment was internal fixation in twothirds and prosthesis revision in one-thrid. At last followup, 81% of the patients returned to their pre-injury home living, but with poor functional outcome. The initial fixation remained intact in 63% of the implants. In the prospective study, there were 20.9% type A, 26.1% type B1, 21.7% type B2, 7% type B3, and 24.3% type C. Surgical treatment was peformed in 83.5% of fractures, as internal fixation in 64.2%, and revision in 35.8%. Mortality at 6 months was 9.8%, with no treatment effect. Peri-operative complications were 8.7%, with no significative treatment effect. Comparison between fixation and revision showed no difference for mortality, blood disorders and morbidity. There were 68 knee periprosthetic fractures in the retrospective study (mean age 73 years, 56 women, mean follow-up of 3.8 years). There were 11 type A fractures, 23 type B, and 28 type C. Sixty-three fractures were treated surgically, 55 with internal fixation, all fused at follow-up. Revision (n=8) gave good results for longitudinal axis and motion. For the prospective study, there were 28 fractures (24 women, mean age 79.7 years). There were 3 type A fractures, 13 types B, and 12 types C. Femur fractures were more frequent (22). Implant loosening was unfrequent (n=5). Internal fixation was performed in 18 cases using plates or retrograde nail, revision of the implant in 6 cases. Complications were frequent (n=12). Bone healing was obtained in 20 cases. Hip periprosthetic fracture is a life threatening condition; some negative factors can be hightlighted. Patient-related factors include: chronic hospitalization, ASA score above 2, low Katz scale; fracture and treatment-related factors include: B2 and C fractures, internal fixation. For B3 fractures, implant revision is advisable. Internal fixation of B2 fractures should be limited to frail patients with very low functional requirements. For B1 fractures, it can be helpful to determine the remaining fixation index (FRI): if FRI is lower than 2/3, outcome of implant fixation is doubtful; such fractures are close to B2 fractures. Type of surgical treatment is related more to Vancouver fracture type than the patient’s medical status. Revision is needed when loosening or osteolysis (B2 or B3) is present. Knee periprosthetic fractures, of the femur or the tibia, which are stable and non-displaced can be treated with a cast. Displaced type A fractures require implant revision. Internal fixation is needed for B1 fractures. The choice between retrograde nailing and locked plating depends on the volume of the remaining epiphyseal bone between fracture line and the implant. In B3 fractures, revision with intra-medullary extension rods is advisable. In C1 fractures, internal fixation is suitable. Comminutive fractures are best treated with intramedullary nailing.