Summary: Complex proximal femur fractures are challenging. The treatment goals of anatomic alignment with stable fixation to allow early functional rehabilitation are difficult to achieve. The Proximal Femoral Locking Compression Plate (PF-LCP) is viewed as a viable alternative for stable fixation of complex proximal femur fractures. The aim of this study is to review the short-term surgical results of the PF-LCP.
Methods: We reviewed the charts and radiographs of 30 consecutive patients treated at a single Level-I Trauma Center from February, 2006 to March, 2009. Two patients died shortly after surgery, leaving 28 patients with adequate follow-up for review. Quality of reduction was critically examined using the Garden's alignment index. All fractures were classified according to the AO system. There were 22 males and 8 females. Average age was 48.4. There were 29 closed injuries and 1 open Gustilo Type IIIA injury. The mechanism of injury was high energy in 19 cases and low energy in 11 cases. Twenty-seven of thirty (90%) patients were treated at an average of 0.8 days after injury. Three patients underwent revision to the PF-LCP from other implants, two for nonunions and one for early fixation failure. The mean follow-up period was 39.9 weeks.
Results: Displaced femoral neck fractures (31-B) were present in 12/30 (40%) patients, intertrochanteric fractures (31-A) were present in 18/30 (60%), and subtrochanteric fractures (32-A1.1, 32-A2.1, 32-A3.1) were present in 6/30 (20%). Nonunion occurred in 5/12 (41.7%) of femoral neck fractures, 0/18 (0%) intertrochanteric fractures, and 2/6 (33.3%) of subtrochanteric fractures. One of the subtrochanteric nonunions was associated with the open fracture, which had been associated with significant bone loss. All femoral neck fractures were found to be satisfactorily reduced according to Garden's alignment criteria. All femoral neck nonunions progressed to varus collapse with failure of fixation. Residual posteromedial fracture gap following fixation due to fracture comminution was found in 3/5 (60%) of the nonunion group and 1/7 (14.3%) of the healed group. This was not significant. There were a total of 6 fixation failures among the 5 femoral neck nonunions. The lengths of the proximal screws were found to be significantly different in the healed and unhealed groups. The mean screw lengths for the cephalic portion in the nonunion versus healed groups was 77 mm versus 93.6 mm for the 95° screws (p=0.02), 89 mm versus 102.9 mm for the 120° screws (p=0.03) and 74 mm versus 101.4 mm for the 135° screws (p=0.008). The average total screw length was 240.0 mm (SD 20.9) versus 297.9 mm (SD 18.5) (p=0.001).
Conclusion: The PF-LCP was found to have a high nonunion rate in complex pertrochanteric variants with ipsilateral femoral neck fractures. Short screw lengths may be a contributing factor. Weaknesses of this study include the retrospective design, small sample size, short length of follow-up, and heterogenous patient population and fracture patterns.