Fixation of periprosthetic or osteoporotic distal femoral fractures with locking plates: a pilot randomised controlled trial

Abstract

Introduction: We hypothesised that the use of a polyaxial locking plate design offers the same clinical benefits as a monoaxial locking plate system following distal femoral osteoporotic/periprosthetic fracture fixation.

Method: A multicentre prospective randomised pilot trial was conducted. Inclusion criteria were patients over 60 years with a displaced osteoporotic or periprosthetic distal femoral fracture. Details documented included time to union, complications, reinterventions and functional outcomes according to the Oxford knee score and EuroQol EQ-5D. Analysis of factors influencing an early fracture healing response was performed between those with clear features of radiological callus formation at three months. Statistical analysis was performed using a logistic regression model with multiple covariates assessed for each plate system (1:1 ratio) over a follow-up period of one year.

Results: Forty patients (34 females) with a mean age of 77 (60-99) were recruited. Four patients deceased within the first six months. Twenty-five patients united by the six month follow-up. Six more patients progressed to union between six and nine months. Five patients developed non-union (two patients had implant failure; one in each group) and all underwent revision surgery. Malunion was evident in two cases, one with 15° of valgus (monoaxial plate), and one with 12° of recurvatum (polyaxial plate). Between the two plate systems, statistical analysis revealed no significant differences in most of the recorded parameters. Radiological features of early bone healing were present when the surgical approach was smaller (p = 0.015), and when a greater working length of the bridging plate was present (p = 0.016).

Conclusion: Both plate systems demonstrated good union rates and limited implant related complications. Good reduction, mechanically sound construct and respect of the local fracture biology was more important than the particular plate design characteristics.

Keywords: Femoral fracture; Locking plate; Periprosthetic fracture; Polyaxial.

Fig. 1

Histogram of the age distribution of the 40 recruited patients to the study, stratified per implant type

Fig. 2

Boxplots of the Charlson comorbidity index by implant type

Fig. 3

Evolution of EQ-5D tariffs over time following surgery

Fig. 4

Type 33.C2 distal femoral fracture to a 67-year-old lady following a fall from standing height (a and b represent preoperative anteroposterior and lateral views of the right distal femur). Randomised to and fixed with a Polyaxial distal femoral plate and three free lag screws addressing the intra-articular extension of the fracture lines, (c, d, e). Uneventful fracture healing was evident to the radiological control at 6 months (f, g) and recovery of pre-injury levels of mobility, knee function, and overall health state recovery

Fig. 5

Type 33.A2 distal femoral fracture of a 92-year-old lady following a fall at her garden (a and b represent pre-operative anteroposterior and lateral views of the right distal femur). Of note, an ipsilateral cemented Thompson’s hemiarthroplasty. Randomised and treated with a LISS plating system inserted with a mini invasive technique, spanning proximally the stem of the hemiarthroplasty to neutralise an in-between implants stress riser (d, e, f). The fracture was noted to be fixed and was malunited in 15° of the valgus (c and d). The same patient underwent further surgery 9 months later whereas two of the metaphyseal locking screws were removed as they were found to be irritating the soft tissues over the medial femoral condyle (c)

Fig. 6

Individual patient trajectories (spaghetti plots) plotted against time demonstrating the progression of the recorded knee score (Oxford Knee Score (OKS)). Mean values, by implant, were calculated at each of the time points (red triangles—baseline, 1, 3, 6, 9, 12 months). The blue line represents a smoothed trend of the progress of the OKS measure for each implant type

Fig. 7

Individual patient trajectories (spaghetti plots) plotted against time demonstrating the progression of the recorded pain score as per the visual analogue scale VAS. Mean values, by implant, were calculated at each of the time points (red triangles—baseline, 1, 3, 6, 9, 12 months). The blue line represents a smoothed trend of the progress of the VAS measure for each implant type

Fig. 8

Individual patient trajectories (spaghetti plots) plotted against time demonstrating the progression of the recorded pain score as per the general health visual analogue scale HS-VAS of the EQ-5D. Mean values, by implant were calculated at each of the time points (red triangles—baseline, 1, 3, 6, 9, 12 months). The blue line represents a smoothed trend of the progress of the HS-VAS measure for each implant type