Management of periprosthetic femoral fractures following total knee arthroplasties using locking plates or intramedullary nailing. Comparative study of 567 cases

INTRODUCTION:Proximal femur fractures in elderly patients pose a common problem that requires surgical management. The main objective of this study was to compare the results of treatment by locked plate (LP) and by anterograde (AIN) or retrograde (RIN) intramedullary nailing. The secondary objective was to analyze the complications related to each group. The hypothesis of this study was that the radiological and clinical results of each group were comparable. MATERIALS AND METHODS:The data of 438 patients from 8 centers were studied retrospectively with 365 LP and 69 cases of intramedullary nailing (49 AIN and 20 RIN). The groups were comparable in terms of mean age (75±2 years), mean preoperative Parker score (5.3±0.5) and mean KOOS score (74±2). Most patients lived at home before their fracture (n=375 (86%)). At follow-up, the joint range of motion, the KOOS and Parker scores, living premises, the time required to regain full weight-bearing, the time required for consolidation as well as the reduction alignment: frontal, sagittal, horizontal. RESULTS:The average follow-up was: 79±2 weeks (65 to 82). The functional and radiological results were comparable, although nailing allowed a faster return to weight-bearing (1 ± 2 weeks for nailing versus 7±2 weeks for LP, p=0.048) and a better post-operative Parker score (5.5±0.5 for nailing versus 4.7±0.2 for LP, p=0.045). Rotational malalignment in femoral nailing was often in the direction of internal rotation, compared to the LP (p=0.045). Complications were comparable with the exception of mortality which was more frequent for LP (22.5% in the LP group versus 13% in the nailing group (p=0.045). Complications were categorized as: mechanical complications 6.8% of LP, 5.7% of nailing (p>0.05)), infections (11% of LP, 5.8% of nailing (p>0.05)), surgical revisions (2% of LP, 10% of nailing (p>0.05)). DISCUSSION AND CONCLUSION:The hypothesis was refuted. These results show that nailing conferred a faster recovery of weight-bearing, a better Parker score, and is associated with a lower mortality rate. The morbidity rate is comparable between the two techniques. Indeed, when the indication for osteosynthesis by LP or by nailing is made, and both choices are technically justifiable, it would be prudent to favor the technique which carries the least risk, which in this case is nailing. This conclusion must, however, be taken in to consideration with regard to the reduced number of nailing. A larger series of nailing, particularly retrograde, would be necessary to be able to confirm this last difference between the 2 groups. LEVEL OF EVIDENCE:IV; Comparative retrospective study.