| Objectives: Although the incidence of subtrochanteric femoral fractures is not high,the treatment of this fracture is still important and difficult for orthopedic surgeons because of the high concentration of stress on the subtrochanteric region of the femur.Appropriate internal fixator can promote the healing of fracture,shorten the patient's bed time,reduce medical expenses,reduce the disability and mortality rate.The purpose of this study was to compare the biomechanical properties of proximal femoral nail antirotation(PFNA),proximal femoral locking plate(PFLP)and less invasive stabilization system fixing four types of subtrochanteric fractures by means of biomechanical lab test and three-dimensional finite element analysis,providing some references for clinical work.Methods: 30 artificial femur models(Sawbones;Model 3403,Pacific Research Labs Inc.,Vashon,WA)were randomly divided into three groups fixed with PFNA,PFLP and inverted LISS respectively.Four strain gauges were attached to the medial and lateral femur at 2.8 cm and 5.2 cm below the apex of the less trochanter(that is,2 mm above and 2 mm below the fracture model area).Each femur was sequentially modeled as a subtrochanteric fracture.Model I: subtrochanteric transverse fracture(Seinsheimer type I),model II: reduction of subtrochanteric medial sphenoid fragment(Seinsheimer type IIIA),model III: removal of medial sphenoid fragment,and model IV: subtrochanteric comminuted fracture(Seinsheimer type IV).After Model I was made,the implants were fixed according to the manufacturer's standard.Bone cement was used to immobilize the distal part of the femur in the one leg standing position.In each fracture model,the implant-femur construct was subjected to an axial compression test of 0 ? 1400 N.Each fracture model was constructed after the previous fracture model was tested.After the axial compression of model IV,the horizontal torsional test of 0 ? 5Nm was carried out.After that,the constructs of each group were randomly divided into two subgroups,one for axial compression failure test and the other for cyclic axial compression fatigue test.Siemens 64-slice CT was used to scan the artificial femur model.The scanning method was voltage: 120 k V,current: 200 m A,scanning slice thickness: 1mm,slice spacing: 1mm,and the image was stored as DICOM format.DICOM file was imported into Mimics 17.0 software,and 3D femoral model was generated by segmentation and smoothing of the original image,and the exported file was input into software Magics 21.0 to generate the solid model.Import the entity model into 3D design software Solidworks 2016.This software was used to construct LISS,PFLP and PFNA models.The fracture models were cut and assembled with implant.The assembled model was imported into the software Ansys Workbench 16.0 in IGES format,to divide the units and grids,and set the boundary conditions and material attribute assignment.The fracture model was simulated by solid 92,a three-dimensional tetrahedral solid unit with ten nodes.Glue contact was used between the femur and screws.Friction contact was used between femur and implant.The friction coefficient was set at 0.30.The axial compression load,horizontal torsional load,axial compression failure load and cyclic axial compression fatigue load were applied to the complete three-dimensional finite element model,which was the same as the lab mechanical test.Results: In model I to III,there was no statistical difference in axial stiffness of the three implants.In model IV,the axial stiffness of PFNA was 352.71 ± 19.05 N/mm,PFLP was 134.39 ± 12.51 N/mm,and LISS was 89.13 ± 16.24 N/mm,with statistical differences among the three(P < 0.0001).The medial strain of the femur fixed by PFLP was significantly higher than that of the other two groups(P < 0.038).There was no significant difference in horizontal torsional stiffness among the three groups: PFNA was 1.97 ± 0.37 Nm/degree,PFLP was 1.71 ± 0.55 Nm/degree,and LISS was 2.11 ± 0.49 Nm/degree(P = 0.187).PFNA had the highest failure load(3387.26 ± 316.60 N),followed by PFLP(2483.28 ± 169.28 N),and LISS(1917.72 ± 194.81 N).Pairwise comparison was statistically significant(PFNA vs.PFLP,P < 0.0001;PFNA vs.LISS,P < 0.0001;PFLP vs.LISS,P = 0.007).There was no damage to PFNA after fatigue test,while the fatigue life of PFLP was 11091 ± 1533 times and LISS was 13809 ± 993 times.There was a statistical difference between the latter two(P = 0.01).In the three-dimensional finite element analysis(3D FEA),for model I,the stress of PFLP and LISS is concentrated at the connection between the locking screw and the steel plate,and the maximum stress occurs at the connection between the distal locking screw and the steel plate.The stress concentration and maximum stress of PFNA are located at the junction of main nail.For model IV,the maximum stress of PFLP and LISS was at the plate screw hole at the fracture level,and the maximum stress of PFNA was above the proximal screw hole.The femoral head sinking displacement of the PFLP group and the LISS group was similar in fracture model I to III,while in fracture model IV,the head sinking displacement of the LISS group was 4.53 mm higher than that of the PFLP group.The head sinking displacement of the PFNA group was slightly higher than that of the PFLP group and the LISS group in fracture model I to III,while the head sinking displacement of the PFNA group was significantly lower than that of the other two groups in fracture model IV.Under torsional loading,the torsional angle was 1.62° in the PFNA group,2.32° in the PFLP group,and 2.98° in the LISS group.When loaded to 3576 N,the femur of the PFNA group showed mesh elimination,located at the proximal side of the proximal screw hole.When loaded to 2331 N,the femoral head sinking displacement of PFLP group reached 20 mm.When loaded to 1757 N,the femoral head sinking displacement of the LISS group reached 20 mm.The minimum fatigue life of PFNA is 209,940 times,PFLP is 10,500 times and LISS is 13,498 times.Conclusions: The FEA method has certain application value for the simulation of experimental biomechanical research,which confirms the results of experimental biomechanical research.For model I fracture,all three internal fixators can be used,but it should be noted that LISS and PFLP are more stable than PFNA fixation which has compressive stress concentration at the medial fracture line.LISS and PFNA can be selected for model II fractures.LISS may be better than PFNA.While PFLP should be selected cautiously,as it is more likely to cause varus deformity and fixation failure.If PFLP is selected,postoperative weight bearing should be delayed.In addition,through FEA,it was found that the peak stress of the PFNA group's ring-binding steel wire in model II was close to the yield strength of stainless steel,while the peak stress of the PFLP and LISS's ring-binding steel wire exceeded the yield strength.Therefore,it was suggested to use double-strand steel wire,titanium lanyard or binding belt to fix the posteromedial buttress in clinical practice to increase the fixation strength.For model III fractures,it is suggested that the posteromedial buttress should be restored as far as possible,and stable fixation should be performed.Meanwhile,implant should be selected according to model II to reduce the incidence of varus deformity,implant failure and fracture nonunion.If the subtrochanteric posteromedial buttress could not be reduced and fixed,PFNA fixation is recommended.PFNA fixation is recommended for fracture model IV. |
PDF Full Text Request