Torsional Stability of the Femoral Neck System for Pauwels Type III Femoral Neck Fracture Fixation: A Finite-element Analysis

Abstract

Introduction:

Femoral neck fracture fixation remains challenging, particularly for Pauwels type III fractures that require robust implants to withstand vertical shear forces. The femoral neck system (FNS) offers greater stiffness and more stable load sharing than multiple screw fixation; however, it has a lower load-to-failure than plate systems. We evaluated the torsional stability of the FNS using finite-element analysis.

Methods:

We analyzed a transcervical Pauwels type III femoral neck fracture without a fracture gap, fixed with one of three constructs: multiple screw fixation, a dynamic hip screw (DHS) with an antirotation screw, or an FNS. Finite-element simulations incorporated muscle forces during the leg-swing phase, with constraints applied at the fovea capitis. Cortical-cancellous interfaces were fully bonded, except at the reduced fracture interface where relative translation was allowed.

Results:

The DHS with an antirotation screw had the highest equivalent von Mises stress and strain energy density (SED) (1130.11 MPa/1,000) and the greatest fragment displacement (0.44 mm). Multiple screw fixation produced the lowest SED (151.73 MPa/1,000), and the FNS showed similar displacement with an intermediate SED (201.96 MPa/1,000). In the FNS, peak stress localized near the lateral cortex of the proximal femur, with reduced stress at the barrel-screw interface.

Conclusions:

The FNS provides rotational stability for Pauwels type III fractures comparable with multiple screw fixation and reduces stress at the fracture site. The DHS with an antirotation screw was less stable during hip rotation. Given its greater compressive strength than multiple screw fixation, the FNS is the preferred fixation option.

Femoral neck fracture fixation remains a challenging orthopaedic problem that presents notable technical demands and has high rates of fixation failure.¹ Minimally displaced Garden I or II fractures with posterior tilt angles less than 20° on lateral radiographs have higher union rates after surgical fixation.^2,3 The Pauwels angle guides implant selection because Pauwels type III fractures, with vertical fracture lines, require more rigid implants to withstand vertical shear forces during the stance phase.⁴

A dynamic hip screw (DHS) with an antirotation screw previously supported greater loads than multiple screw fixation.⁵ Recently, the femoral neck system (FNS) has gained popularity for femoral neck fracture fixation, offering increased stiffness and rotational stability through an integrated antirotation screw.⁶ Biomechanical studies show that the FNS provides greater stiffness and more favorable load distribution than multiple screw fixation.^6,7 However, the maximum load to failure of the FNS is lower than that of the percutaneous compression plate.⁸

Most biomechanical studies have emphasized compressive loading during the stance phase,^9,10 leaving torsional stability, particularly during early postoperative recovery in the swing phase, uncertain.¹¹ Accordingly, we evaluated the rotational stability of the fixation constructs during hip rotation using finite-element analysis.

Methods

Study Approval and Model Development

This research received approval from the Institutional Review Boards of the authors’ affiliated institutions. We created three-dimensional computer-aided design (CAD) models of the femur and fixation implants. The femur model was derived from CT imaging of one of the authors (N.C.), who had no prior trauma or bony deformities. We modeled three fixation constructs: multiple solid 6.5 mm diameter AO Synthes screws, a DHS with an antirotation screw, and an FNS. We generated the finite-element model from the CAD models using preprocessing software (Patran; MSC Software). All analyses were done using commercial finite-element software (MSC Mentat; MSC Software).

CT Acquisition and Three-dimensional Reconstruction

We digitized the femur geometry using a spiral CT scanner (Philips Brilliance CT 64-slice; Philips Medical Systems). The protocol scanned the hip-to-ankle region with 0.625 mm slices and a 512 × 512 matrix. Data were stored in a Digital Imaging and Communications in Medicine (DICOM) format as a stack of two-dimensional transverse images. We reconstructed a three-dimensional femur model in three-dimensional Slicer (slicer.org) from the DICOM file. During reconstruction, we applied a Hounsfield unit threshold of +500 to +2,400 to extract the cortical outer surface and the diaphyseal intramedullary canal. For the epiphysis, we set the cortical thickness at 3 mm and connected the inner surface to the intramedullary canal to complete the cortical layer. Cancellous bone was modeled only in the epiphysis, extending 90 mm from each bone end, and finalized in VISI CAD software (Hexagon AB).

Fracture Definition

The fracture pattern was a transcervical Pauwels type III femoral neck fracture oriented at 75° to the horizontal plane, without a gap between the femoral head and the trochanteric fragment, simulating a fracture reduction.

Implant Geometry and Measurement

We evaluated three fixation constructs. These comprised 6.5 mm diameter cannulated partially threaded screws (DePuy Synthes), a DHS with an antirotation screw (DePuy Synthes), and an FNS (DePuy Synthes). The DHS included a four-hole side plate (78 mm length, 135° caput-collum-diaphyseal angle, 38 mm barrel length) and a lag screw (110 mm length, 8 mm shaft diameter, 12.7 mm thread diameter). It was secured with four 4.6 mm cortical screws. The FNS comprised a one-hole plate (130° CCD angle), a 10 mm bolt (90 mm length), a 6.4 mm antirotation screw (75 mm length), and a 5 mm locking screw. We measured all fixation dimensions using a digital microscope (Dino-Lite; AnMo Electronics) and a Vernier caliper (Digimatic Caliper; Mitutoyo). These measurements were used to reconstruct three-dimensional CAD models in VISI (Hexagon AB).

Virtual Implant Placement

Fixations were applied according to standard surgical techniques. The FNS was placed with a 130° orientation parallel to the femoral neck-shaft angle. The dynamic bolt was centered in the femoral head, with a 5 mm distance from the bolt tip to the femoral head. Multiple screw fixation used three screws in an inverted-triangle configuration, with tips positioned 10 mm from the subchondral bone. For the DHS with an antirotation screw, the lag screw was aligned parallel to the femoral neck axis, with its tip 10 mm from the subchondral bone. The lag screw was seated within the barrel, and the side plate was affixed to the lateral cortex using four 4.5 mm cortical screws. The antirotation screw was oriented parallel to the femoral neck axis and positioned just above the lag screw. After virtual insertion, the fractured femur model was subtracted from the fixation models to create complete three-dimensional CAD models for finite-element analysis (Figure 1).

Figure 1.

Diagram showing computer-aided design models of a Pauwel III femoral neck fracture with three fixation constructs. The fracture line was inclined 75° to the horizontal with no gap at the fracture site, representing an anatomic reduction. Constructs: FNS, DHS with an antirotation screw, and multiple screw fixation (inverted-triangle). DHS = dynamic hip screw, FNS = femoral neck system

Finite-element Model Generation

We used finite-element preprocessing software (Patran; MSC Software) to generate element models from the CAD designs of the fractured femur and fixation constructs. All models used four-node tetrahedral elements throughout the analysis. We performed a convergence analysis by monitoring the equivalent von Mises stress (EQV) on the fixation while varying the number of elements. The multiple screw construct comprised 419,002 elements and 101,027 nodes. The DHS with an antirotation screw contained 492,519 elements and 119,437 nodes. The FNS included 447,145 elements and 108,381 nodes.

Material Properties

All materials were modeled as homogeneous and linearly elastic. Cortical and cancellous bone was treated as anisotropic, and fixation implants were treated as isotropic. Material properties by region are summarized in Table 1.

Table 1.

Material Properties Used in the Finite-element Analysis

Materials	Elastic Modulus (MPa)	Poisson Ratio	Shear Modulus (MPa)
Cortical bone	E1 = 17,900	ν12 = 0.26	G12 = 5,710
	E2 = 18,800	ν23 = 0.31	G23 = 7,110
	E3 = 22,800	ν31 = 0.38	G31 = 6,580
Cancellous bone	E1 = 676	ν12 = 0.30	G12 = 370
	E2 = 968	ν23 = 0.30	G23 = 292
	E3 = 1,352	ν31 = 0.30	G31 = 505
Fixations (titanium)	E = 110,000	ν = 0.33	—

Bone was modeled as anisotropic (orthotropic); fixation implants were modeled as isotropic titanium.

Boundary and Contact Conditions

To simulate hip rotation, boundary conditions were based on prior biomechanical literature. Muscle forces associated with the leg-swing phase were applied to the femur,¹¹ and a constraint was applied at the fovea capitis of the femoral head (Figure 2). Cortical-cancellous interfaces were fully bonded, except at the reduced fracture interface, where relative translation was allowed. Fixation surfaces embedded in bone were bonded to the bone interfaces. Relative translation was allowed at the DHS barrel-lag screw and cortical screw-side plate interfaces, and at the FNS bolt-antirotation screw interface. Coefficients of friction were 0.36 for bone-titanium, 0.30 for titanium-titanium, and 0.46 for bone-bone.

Figure 2.

Diagram showing finite-element model and boundary conditions for simulating hip rotation. Four-node tetrahedral elements were used. Muscle forces associated with the leg-swing phase were applied to the femur; a constraint was applied at the femoral head (fovea capitis), and the distal femur was left unconstrained.

Results

Equivalent Stress Distribution in Fixation Constructs

The EQV in all fixation constructs was primarily concentrated near the fracture site. The DHS with an antirotation screw had the highest EQV, localized at the screws and the barrel. In the FNS, peak stress occurred at the proximal barrel near the lateral wall of the proximal femur, with lower stress at the barrel-screw junction. Multiple screw fixation showed the lowest EQV during hip rotation (Figure 3 and Table 2).

Figure 3.

Diagram showing EQV distribution in fixation constructs during simulated hip rotation. The DHS with an antirotation screw shows the highest EQV, the FNS concentrates peak stress near the proximal barrel/lateral cortex, and multiple screw fixation shows the lowest EQV. DHS = dynamic hip screw, EQV= equivalent von Mises stress, FNS = femoral neck system

Table 2.

Equivalent von Mises Stress in Fixation Constructs During Simulated Hip Rotation (MPa)

Fixation	EQV Stress (MPa)
FNS	959.06
DHS with an antirotation screw	1413.99
Multiple screw fixation (inverted-triangle configuration)	315.98

DHS = dynamic hip screw, EQV = equivalent von Mises stress, FNS = femoral neck system

Lower EQV indicates lower stress in the fixation under the applied rotational loading.

Fragment Displacement

Fragment displacement is a critical indicator of fracture stability. The DHS construct with an antirotation screw showed the greatest fragment displacement during hip rotation (0.44 mm). The FNS and multiple screw models showed interfragmentary displacements of 0.28 mm and 0.27 mm, respectively (Figure 4 and Table 3).

Figure 4.

Diagram showing interfragmentary displacement at the fracture site during simulated hip rotation. The DHS with an antirotation screw shows the greatest fragment displacement; the FNS and multiple screw configurations show smaller displacements. DHS = dynamic hip screw, FNS = femoral neck system

Table 3.

Interfragmentary Displacement at the Fracture Site During Simulated Hip Rotation (mm)

Fixation	Displacement (mm)
FNS	0.28
DHS with an antirotation screw	0.44
Multiple screw fixation (inverted-triangle configuration)	0.27

DHS = dynamic hip screw, FNS = femoral neck system

Displacement measured at the fracture plane of the proximal femur.

Strain Energy Density in Proximal Cancellous Bone

Strain energy density (SED) was used to evaluate lag screw attachment to cancellous bone; higher values indicate greater risk of screw loosening or cutout. The DHS construct with an antirotation screw showed the highest SED at 1130.11 MPa/1,000, which was 5 to 10 times greater than those of the other two constructs. The cannulated screws in an inverted-triangle configuration had the lowest SED during hip rotation at 151.73 MPa/1,000. The FNS produced femoral head displacement and cancellous bone SED comparable with the cannulated screw configuration, with SED of 201.96 MPa/1,000 (Figure 5 and Table 4).

Figure 5.

Diagram showing SED in proximal cancellous bone during simulated hip rotation. Higher SED indicates greater risk of screw loosening/cutout. The DHS with an antirotation screw shows the highest SED; multiple screw fixation shows the lowest; the FNS is intermediate. DHS = dynamic hip screw, FNS = femoral neck system, SED = strain energy density

Table 4.

Strain Energy Density in Proximal Cancellous Bone During Simulated Hip Rotation (MPa/1,000)

Fixation	SED (MPa/1,000)
FNS	201.96
DHS with an antirotation screw	1130.11
Multiple screw fixation (inverted-triangle configuration)	151.73

DHS = dynamic hip screw, FNS = femoral neck system, SED = strain energy density

Higher SED has been associated with greater risk of screw loosening/cutout.

Discussion

Femoral neck fracture fixation remains a challenging orthopaedic problem characterized by high rates of fixation failure.^1,12 Despite advances in fixation devices, complications—implant failure, nonunion, and osteonecrosis of the proximal fragment—remain common. Accordingly, even in minimally displaced or impacted valgus fractures, delaying weight bearing and restricting hip motion are still recommended.¹³

Biomechanical testing has typically emphasized compressive loading; most studies report fixation failures under compressive loads,^6,7,9,10 while few examine complications related to hip rotation.^11,14 In patients of advanced age with osteoporosis—a leading cause of fragility fractures—proximal femoral geometry and trabeculation change.¹⁵ The metaphyseal bone loses trabeculation, increases in diameter, and becomes marginally calcified to maintain its moment of inertia.¹⁶ These alterations reduce the bone-screw interface, thereby lowering fixation pullout strength.¹⁷

Our results show that the DHS with an antirotation screw has the greatest biomechanical disadvantages during hip rotation. The maximum EQV was 1.5 to 4.5 times greater than that of the FNS or multiple screw constructs. The SED of the DHS was also markedly higher, reaching 5 to 10 times the values observed in the other two constructs. Therefore, the DHS with an antirotation screw must withstand greater rotational stress and is more susceptible to fixation loosening.¹⁸

High Pauwels angles in femoral neck fractures increase vertical shear at the fracture site.⁸ A DHS with an antirotation screw was previously preferred for high Pauwels-angle fractures because it withstands vertical shearing forces better than multiple screw fixation.^19,20 However, single-leg support during the stance phase accounts for 40% of the gait cycle, and the same leg remains in the swing phase for another 40%. Therefore, biomechanical testing focused solely on compressive loads does not fully represent the overall stability and integrity of fixation constructs.¹¹

The FNS, recognized for its flexible biomechanical properties, is a promising option for femoral neck fracture fixation. Previous studies show that the FNS provides superior vertical compressive resistance compared with multiple screw constructs.^6,8,14 Furthermore, this study indicates that the FNS offers superior rotational resistance compared with the DHS with an antirotation screw. This advantage may facilitate early hip mobilization after surgery and increase confidence in weight-bearing and walking training during rehabilitation.^11,21 The FNS is designed to withstand both compressive forces and rotational torque, thereby providing greater security during recovery.

The strength of this study lies in its analytical methods, particularly the treatment of boundary and contact conditions. We simulated total contact force during the swing phase of the gait cycle while leaving the distal femur unconstrained,^11,21 unlike previous studies that analyzed compressive deformation under vertical loads across gait cycles. We also evaluated SED, which is associated with implant loosening in the metaphysis of the proximal femur, a notable cause of fixation failure.¹⁸ Implant loosening diminishes pullout strength and reduces the bone-screw interface, a critical concern in geriatric patients whose trabeculation deteriorates with age or osteoporosis.¹⁶ This demographic is particularly relevant to decisions about femoral neck fracture fixation, given the ongoing surgical dilemma between arthroplasty and fixation.

This study has several limitations. First, the finite-element model used a simple fracture configuration with perfect anatomic reduction, a condition rarely achieved clinically because younger patients with high-energy trauma often have comminution. Older patients typically present with valgus impaction fractures that often remain unreduced yet are considered stable. Accordingly, the analysis may deviate from real-world fixation configurations because of these differences.

Second, the finite-element model was derived from a single individual younger than 60 years, which may not represent the general population because bone quality varies by age. Third, fracture healing is dynamic and progressively enhances fixation configuration stability over time. Therefore, the results likely reflect the immediate postoperative phase because callus formation and healing will influence longer term outcomes.

Conclusions

The FNS provides excellent rotational stability for Pauwels type III femoral neck fractures, with fragment displacement and SED comparable with multiple screw fixation. Multiple screw fixation shows the lowest stress concentration at the fracture site. By contrast, the DHS with an antirotation screw is less stable during hip rotation. Taken together with its strength under compressive load, the FNS is the preferred fixation option for Pauwels type III femoral neck fractures.