Finite element analysis of sliding vs static configuration of partial threaded cannulated compression screws for Pauwels type III fracture neck of femur

Abstract

Background

Cannulated Cancellous Screws fixation is done either as a sliding or static configuration according to the number, thread length and orientation of the screws. This study aims to compare the biomechanical properties of 3-CCS as sliding configuration and BDSF & 4-CCS (Alpha configuration) as static configuration, by finite element analysis for Pauwels type III fracture neck of femur.

Methods

DICOM format images of saw-bone were used to build the 3-D geometric models, which were hypermeshed to create models of Pauwels type III FNF with an angle of 65°. CAD models of three configurations were created as per AO specifications and analysed for stress distribution of bone & implant, relative displacement between fracture planes & stiffness.

Results

The peak stress at the fracture surface was found to be low in 4-CCS, which promotes fracture healing. 4-CCS had a decreased chance of failure due to lower implant stress and fracture surface displacement, indicating improved length stability. Overall, 4-CCS had the maximum construct stiffness, indicating a stable and rigid fixation.

Conclusion

The static configuration of 4-CCS showed superior biomechanical properties in terms of low stress of bone & implant, and lower relative displacement of the fracture surface with high construct stiffness over other configurations. So, 4-CCS (Alpha configuration) should be the preferred configuration for the Pauwels type III FNF.

Abbreviation:

CCS

Cannulated Cancellous Screw

FNF

Fracture Neck Femur

AO

Arbeitsgemeinschaft für Osteosynthesefragen

BDSF

Biplanar Double Supported Fixation

DICOM

Digital Imaging and Communications in Medicine

CAD

Computer Assisted Design

MPa

Megapascal

N/mm

Newtons per millimeter

AVN

Avascular Necrosis

IIT

Indian Institute of Technology

FEA

Finite Element Analysis

1. Introduction

Cannulated Cancellous Screws (CCS), in their various configurations, are commonly used to fix fracture neck of femur (FNF). A stable configuration has been suggested by several biomechanical studies for fracture healing, regardless of the fracture morphology, in order to prevent complications such as non-union, neck shortening, implant back-out, varus collapse, or avascular necrosis¹ (AVN). Pauwels type III fracture with dominant shear and bending force at the fracture site, has the highest complication rate.²

According to literature,³ non-union following internal fixation of Pauwels type III FNF is reported to vary from 16 % to 59 %, and the occurrence of femoral head necrosis is recorded to be between 11 % and 86 %, greatly damaging the patient's prognosis. Commonly used CCS in sliding/parallel configuration has limited ability to resist bending & shear forces, resulting in high failure rate in Pauwels type III FNFs. In addition, uncontrolled dynamic axial compression leads to bone absorption at the fracture site, subsequently causing neck shortening and lateral protrusion of the implant, contributing to poor hip scores.⁴

In order to counter these problems, static configurations are created using diverging screws, as in biplane-double supported fixation (BDSF), or 4-CCS (Alpha configuration) with the promise of positive clinical results.^{5, 6, 7, 8, 9, 10} Jiang et al.⁶ in their retrospective comparative study with 4-CCS vs 3-CCS on 157 patients with Pauwels Type III FNF, found a significantly lower rate of femoral nek shortening (8.3 % vs. 28.9 %), non-union (3.3 % vs. 11.3 %), AVN (21.7 % vs. 25.8 %) with 4-CCS.

Sami et al.⁸ in his prospective observational study using BDSF technique on 27 patients including 16 Pauwels type III, found the union rate of 92.59 % with excellent functional outcome.

The anatomical, biological, and mechanical factors that affect the prognosis of these fractures are highly variable and partially outside the treating surgeon's control. Therefore, for unstable Pauwels type III FNF, it is necessary to comprehend the biomechanical characteristics of screw constructs. To establish the optimal configuration for Pauwels type III FNF, this study used finite element analysis (FEA) to examine the biomechanical properties of widely used configurations, such as the sliding configuration of 3-CCS and the static configuration of BDSF and 4-CCS.

2. Material & methods

Three dimensional models of the femoral neck fracture [Fig. 1]: The study was registered in the Institute research cell, Indian Institute of Technology, Delhi (IITD/DPGSR/MSR-468). To avoid any ethical conflict, synthetic saw-bone conforming to human anatomical proximal femur was used.

Fig. 1.

FE models of fracture neck fracture with Pauwels type III (65°) & CAD of various CCS configuration, a,d:3-CCS; b,e: BDSF; c,f: 4-CCS.

The whole length of the saw bone femur was scanned using a CT scanner. Then the data in DICOM format was transferred to a software MIMICS, and the 3-D geometric model of the femur was exported in STL format. STL files of the femur were imported into Hypermesh software (version 2019.1) to smooth the surface mesh and to create a fracture with an angle of 65°. To evaluate the accuracy of finite element (FE) models, convergence tests were performed to determine the optimum maximum element size. The average size of elements in bone and implant were 3 mm and 1 mm, respectively. The geometric model of the bone and the implant components were put into Solidworks software and assembled using Boolean techniques. The assembled 3-D meshed models of the fixation technique were imported to Abaqus for further simulation.

In the 3-CCS model,¹⁰ three CCSs (length 97.18 mm) ran parallel to the femoral neck axis in an inverted triangular shape from the lateral cortex of the proximal femur as widely apart as feasible within the femoral neck. Three parallel CCSs were added to the 4-CCS model,¹¹ as previously reported. The fourth Pauwels screw (92.18 mm) was introduced at the greater trochanter and secured within inferior part of the femoral head. In the BDSF model,¹² the distal CCS's (126.523 mm) entry point was located in the anterior one-third of the femur, approximately 5 cm distally from the lesser trochanter, with the screw tip entering the dorsal half of the femur head with an anteriorly-distally to posteriorly-proximally inclination. The entrance position of the middle CCS (111.52 mm) was about 2 cm proximal from the entry point of the distal CCS, and the proximal CCS (96.77 mm) was about 1 cm proximal from the middle CCS, in the dorsal one-third of the femur. These two parallel CCSs entering the front half of the femur head with inclination from posteriorly–distally to anteriorly–proximally.

The screw tip was placed 5 mm below the cortical bone of the femoral head. All cannulated screws with a diameter of 7.3 mm and a thread length of 16 mm were created using SolidWorks software based on clinical screw data.¹³ Because the thread was not the focus of this study, threaded part of screw was simplified by modeling it as smooth surfaces with diameters that corresponded to the planned thread size & length.

Material properties of implants and bones¹³ [Table 1]: Synthetic bone and implants were assumed to be of homogeneous, isotropic, and linearly elastic properties. All the implants were made of titanium alloy for our analysis and stainless steel for comparative analysis.

Table 1.

Material properties of implants and bones according to Li J et al.¹³.

Material	Young’ s Modulus (MPa)	Poisson’ Ratio	Tensile Strength (MPa)
Cortical Bone	17000	0.3	106
Cancellous Bone	155	0.3	6
Stainless steel	200000	0.3	766
Titanium	110000	0.30	950

Interfacial conditions between different components [Table 2]: Contact between cortical and cancellous bone was taken as fully bonded so that load can be transmitted from bone to the fixation devices. According to the contact method described in references, the bone-implant interface was considered as frictional contact, and the value of the coefficient of friction was taken as 0.3¹³. The coefficient of friction between fracture planes was taken as 0.46¹³. A tie constraint was applied between femoral head and threaded part of screws which prevented any relative motion between bone and implant while along the smooth shaft of the screws, the resistance against motions between bone and implant was optimized by increasing the friction at the interface. This differential bonding of screws to bone, allowed independent evaluation of parameters at the fracture site, enables us to investigate fixation methods without the influence of the screw-bone interface of proximal part which is fully bonded.

Table 2.

Interfacial conditions between different components according to Li et al.¹³.

Interface	Condition
Implant-Bone	Frictional Contact(μ = 0.3)
Between fracture planes	Frictional contact(μ = 0.46)
Cortical bone-Cancellous bone	Fully Bonded
Threaded part of screws-Bone	Fully Bonded

Boundary and loading conditions:. A hip reaction force of 2100 N was given to the weight-bearing area of the femoral head, which is three times the weight of a person weighing 70 kg. Load was applied with 10° abduction and angled backwards by 9° to replicate the single-leg standing posture.¹³ Using Abaqus software, bolt load of 224 N was preloaded to the center of the implant screws in order to replicate the compression effect between the fracture planes in all models. All nodes at the distal end of the femur were constrained from six degrees of freedom to zero to prevent its motion during simulation.¹⁴

2.1. Validation of our model

We selected 3-CCS to validate the models & compared our results [Table 3] with three literatures available on 3-CCS. Models by Li et al.,¹³ Huang et al.,¹⁴ Zhan et al.¹⁵ and our model presented similar trends of values after giving same properties. On nephogram [Fig. 2, Fig. 3] comparison we found, there was similar picturisation of distribution pattern between models by Li et al. and our models. Thus validating our model for further study.

Table 3.

Comparison of our model results with the works of literature for Validation of models.

Parameter	Our model	Literature13	Our model	Literature14	Our model	Literature15
MSI	328.8	363.43	319.1	336.3	216.2	196.04
MSB	105.1	116.32	156.2	138.8	93.68
MDI	3.086	7.9592	4.537	1.262	2.492	2.893
MDB	3.502	8.1479	5.73	1.342	2.697	3.149

Fig. 2.

Stress distribution of bone & implant in model by Li et al.¹³ (i,ii) and our model (iii,iv).

Fig. 3.

Displacement distribution of bone & implant in model by Li et al.¹³ (a,b) and our model (c,d).

2.2. Outcome measures

In the finite element analysis, Von Mises stress was assessed on the bone, fracture surface, and implant, and displacement was measured between fracture surfaces. Model stability was determined by calculating construct stiffness. The higher the stiffness value, the greater the model stability. These parameters were compared for the three fixation methods (3-CCS, BDSF, and 4-CCS).

3. Result

3.1. Von Mises stress distribution of femur bone [Fig. 4]

Fig. 4.

Von-Mises stress distribution of femur bone.

Among titanium models(Ti), the Maximum von Mises stress for the femur bone was 90.64 MPa (Mega Pascal) and 79.69 MPa for 3-CCS and BDSF, respectively, located very close to the calcar area. In the 4-CCS, the maximum von Mises stress was 95.45 MPa, located near the intersection of the off-axis screw and cortical bone. These results imply that bone failure occurs in the calcar region near the fracture in the 3-CCS & BDSF configuration, whereas in 4-CCS, bone fails at the intersection of the off-axis screw & lateral cortical bone under the given load.

The maximum von Mises stress among the stainless steel model was 74.64 MPa, 59.90 MPa, and 89.84 MPa, at the same location as in the case of the titanium model for 3-CCS, BDSF, and 4-CCS, respectively.

3.2. Maximum stress on fracture surface [Fig. 5]

Fig. 5.

Maximum stress on Fracture surface.

High stress induces considerable micromotion, that potentially hinders frature healing and results in non-union of fracture. The maximum value of von Mises stress on the fracture surface occured at the lower portion in each configuration. The value [Bar Graph 1] thus obtained was low in 4-CCS model, suggesting lower failure risk of bone at fracture site and favouring fracture healing. This low stress was explained by presence of off-axis screw which significantly reduced the stress value from 87.19 MPa (3-CCS, Ti) to 60.92 MPa (4-CCS, Ti). Similarly stress value was significantly reduced with BDSF configuration (79.69 MPa). Lower value was seen for stainless steel models. [71.83 MPa, 54.83 MPa & 51.70 MPa for 3-CCS, BDSF & 4-CCS, respectively]

Bar Graph 1.

Maximum stress on fracture surface.

3.3. Von Mises stress distribution of implant [Fig. 6]

Fig. 6.

Von-Mises stress distribution of implant.

In the titanium model, the implant's maximum von Mises stress was found to be 322.5 MPa, 262.8 MPa, and 153.6 MPa in 3-CCS, BDSF, and 4-CCS, respectively, positioned in the center of the inferior screw in all.

In the case of the stainless steel model, the maximum von Mises stress of the implant was 332.9 MPa, 298.7 MPa, & 184.6 MPa in 3-CCS, BDSF, & 4-CCS, respectively, located at the same positions.

The value [Bar Graph 2] implied lower deformation of 4-CCS model with titanium implant compared to a stainless steel implant. The stress nephogram[Fig. 6] showed that the addition of the off-axis screws did not change the stress distribution in 3-CCS but reduced the value significantly on inferior screw (Ti: 3-CCS: 322.5 MPa; 4-CCS: 153.6 MPa).

Bar Graph 2.

Maximum stress in implant.

3.4. Maximum interfragmentary displacement of nodes at fracture gap [Fig. 7, Bar Graph 3]

Fig. 7.

Maximum interfragmentary displacement of nodes at fracture gap.

Bar Graph 3.

Maximum relative displacement of fracture plane.

The pattern and rate of healing in fractures are significantly influenced by the magnitude of inter-fragmentary displacement. We estimated the fracture gap between two nodes (1 from each fracture plane) after the hip reaction force. The relative displacement at the fracture gap in the titanium models 3-CCS, BDSF, and 4-CCS was 1.006 mm, 0.637 mm, and 0.35 mm, respectively. The corresponding measurements for stainless steel models were 0.653 mm, 0.452 mm, and 0.259 mm. Overall, the 4-CCS with stainless steel implant showed minimum displacement.

3.5. Stiffness calculations [Bar Graph 4]

Bar Graph 4.

Stiffness of various models.

Rigid fixation is required to allow for countering various deforming forces and allow early postoperative mobility, which would reduce the risk of complications. Stiffness being direct measure of stability, was calculated by dividing the force applied by the maximum displacement.¹⁶

For the titanium models, the maximum stiffness was 2087.47 N/mm, 3296.70 N/mm, & 6000 N/mm, respectively, in 3-CCS, BDSF, & 4-CCS. The maximum stiffness in stainless steel models of 3-CCS, BDSF, & 4-CCS was 3215.92 N/mm, 4646.01 N/mm, & 8108.10 N/mm, respectively.

Because stainless steel has a high Young's modulus, the 4-CCS configuration with the stainless steel implant had the maximum stiffness of all the models. In contrast, titanium implants had a relatively low stiffness.

4. Discussion

The orthopaedic surgeons have always been fascinated by the challenge of treating fractures in the neck of the femur, often referred as ‘the unsolved fracture’.¹⁷ The outcome of these fractures can vary widely and depends on anatomical, biological, and mechanical factors that are not entirely within the control of the treating surgeon.

FNF caused by high-energy trauma are frequently seen in young individuals, and the primary surgical approach for preserving the patient's hip joint is closed reduction and internal fixation. Vertical FNFs, on the other hand, are unstable due to shearing forces, and fracture fragments tend to shift after internal fixation, resulting in a significant risk of sequelae. According to literature,³ non-union following internal fixation of Pauwels type III FNF is reported to vary from 16 % to 59 %, and the occurrence of femoral head necrosis is recorded to be between 11 % and 86 %, greatly damaging the patient's prognosis.

Rigid fixation, controlled dynamic axial compression, and minimal invasiveness are essential requirements for an effective approach in treating FNFs.¹⁸ The use of CCS in sliding configuration, is widely utilized method due to its effective dynamic axial compression. However, its limited ability to withstand the shear force of Pauwels type III FNFs results in two negative consequences. First, the shear force causes considerable micromotion at the fracture site, hindering fracture healing and leading to non-union and AVN.¹⁹ Second, uncontrolled dynamic axial compression leads to bone absorption at the fracture site, subsequently causing neck shortening and lateral protrusion of the implant, contributing to poor hip scores.⁴

To address these negative effects, research has suggested using a static configuration of CCS instead of 3-CCS as sliding configuration. This configuration can be achieved by placing screws in a diverging plane, as seen in BDSF, or by using 4-CCS (Alpha configuration).^{5, 6, 7, 8, 9, 10}

In a biomechanical analysis⁵ of 4-CCS, it was proven that the crossed screw prevents the sliding effect and secondary displacement caused by shear force. Therefore, the crossed screw is initially utilized to compress the fracture site. This crossed fixation offers superior interfragmentary stability when compared to sliding screws. The interlocking arrangement with other screws delivers controlled dynamic axial compression force, thus achieving a better equilibrium between rigid fixation and dynamic compression.

Comparably, Filipov's BDSF¹² uses a unique arrangement of three CCS with a distal screw that is positioned at a greater angle with the femoral diaphysis, touching the femoral calcar and posterior cortex. A biplanar support is formed that can withstand axial compressive stress as well as torsional and shearing stresses. The study concluded that BDSF can improve continuous stability while a patient is engaged in different activities. Additionally, BDSF showed positive results in clinical settings.⁹

Our study explored the biomechanical performance of commonly used configurations of CCS for Pauwels III FNF using finite element analysis. Understanding the stability of the internal fixation model was provided by examining the displacement and stress of the proximal femur and the internal fixation device following force application. We contrasted the static configuration of BDSF & 4-CCS (Alpha configuration) with the sliding configuration of 3-CCS fixations.

The results of the study show that titanium implants generally fared lower than stainless steel implants. This is because stainless steel has a higher Young's modulus than titanium implants, which gives it greater resilience²⁰

The maximum Von Mises stress represents a region of high load concentration within the bone. In the calcar region of the 3-CCS and BDSF, stress was situated close to the fracture line; in the 4-CCS, however, the maximum stress was found where the lateral cortex and off-axis screw met. When compared to the 4-CCS configuration, high stress values near the fracture in 3-CCS and BDSF may cause an increased mechanical load on the bone, which could impede fracture healing. This result is consistent with biomechanical and clinical research that favours a high non-union rate in the 3-CCS configuration.²¹ In the 4-CCS configuration, the addition of an off-axis screw causes the high-stress area to move from the calcar region to the lateral cortex, away from the fracture site. Other studies have also reported potential benefits of this configuration for fracture healing.²²

When an implant has a lower stress value under a given load, the likelihood of breakage or failure is reduced. Among all models, the 4-CCS configuration was found to have the lowest maximum stress. In the 3-CCS and 4-CCS, the maximum stress in the implant was found in the upper part of the inferior screw's middle portion, whereas in the BDSF model, it was found in the posterior-inferior screw, indicating that screw would be the one to bear the maximum load with the greatest likelihood of bending deformation. Over the fracture line, the stress on the implant was evenly distributed along the screw. Out of the three configurations, the 4-CCS configuration offers a lower maximum stress value and a better distribution of stress in internal fixation.

One criterion used to assess the stability of fixation techniques was displacement. By measuring the relative displacement between two nodes across the fracture gap, we were able to determine the displacement at the fracture gap under the given load. Our findings were consistent with those of other studies in which maximum displacement in 3-CCS configurations was noted.^{13, 14, 15} This result demonstrates that sliding 3-CCS fixation is ineffective in providing sufficient stability for high shear force in a Pauwels type 3 fracture and correlates with a high rate of varus collapse or non-union. In a similar vein, the study suggests that configuration with BDSF and 4-CCS offers superior stability over 3-CCS with lower displacement in bone, implant, and fracture gap. The 4-CCS's off-axis screw provides better fracture stability than BDSF and resists displacement, as demonstrated in the study.

Construct stiffness of models was also calculated to counteract deforming forces and resist cyclic loads until failure. We discovered that the 4-CCS configuration had the highest stiffness for a Pauwels type III fracture out of all models.

5. Conclusion

This study infers the importance of static configuration of CCS for FNF and conclude that 4-CCS should be chosen as it provides more stable fixation for type III fracture. It provides insight into decision making of best configuration of CCS for Pauwels type III FNF.

5.1. Limitation

Due to the absence of biological variability and viscoelasticity in synthetic bone, the results may not be applicable to the osteoporotic femur. Because the study's goal was unrelated to threads, it was modeled as a smooth surface with a tie constraint between bone and screw threads. We replicated hip reaction force in the absence of diverse muscle forces, which dynamically vary the reaction force throughout the joint during different physical activities. Because FEA did not capture the healing process mechanism, all results reflect only on initial stability. The omission of these components, although simplifying the real-life settings, is important in illustrating the static stability of fixation. Clinical studies should be done to validate these findings.

CRediT authorship contribution statement

Amit Kumar: Conceptualization, Writing – original draft. Somesh Kumar: Software, Data curation, Methodology. Ajeet Kumar: Conceptualization, Project administration. Kumar Keshav: Resources, Writing – review & editing. Anurag Baghel: Software, Validation. Pulak Sharma: Writing – review & editing.

Informed consent

As no human subject was involved in the study, consent was not required.

Ethical approval

Ethical approval was not required.

Funding

This research did not receive any specific funding.

Declaration of competing interest

The authors declare that they have no relevant financial or non-financial interests to report.