Abstract
BACKGROUND:
Femoral neck fracture is an unsolved challenge in orthopedics. The complication rate in particular is high. There remains a lack of consensus on the optimal choice of internal fixation for unstable femoral neck fracture.
OBJECTIVE:
The study aimed to develop a new headless compression supporting screw (HCSS) for the treatment of unstable foemoral neck fracture.
METHODS:
We designed a new HCSS and used a femoral neck fracture (Pauwels III fracture) model (left, fourth-generation composite, Sawbones) and three-dimensional finite element analysis to compare the biomechanical performance of HCSSs with that of cannulated compression screws (CCSs) for treatment of unstable femoral neck fracture.
RESULTS:
Maximum displacement, peak von Mises stress, peak strain, and rotation for the HCSS were smaller than those for the CCS. The stress was more widely distributed for the HCSS, whereas the stress was concentrated for the CCS.
CONCLUSIONS:
The HCSS resulted in better biomechanical stability than that from the CCS. For Pauwels III fractures the HCSS exhibits better resistance to shear forces and better support, providing a new clinical treatment.
Keywords
Introduction
Femoral neck fracture, a challenge in orthopedics, is a common injury, accounting for 3.6% of all fractures [1]. The incidence of femoral neck fracture possesses a bi-modal age distribution—in older adults it is caused by falls, resulting in damage caused by low-energy impacts, while in younger adults the damage is often caused by high-energy impacts, resulting in an unstable femoral neck fracture [2,3]. The Pauwels classification is an important criterion for the biomechanical evaluation of fracture healing; fractures with an angle greater than 50° between the fracture line and horizontal are defined as Pauwels III fracture, in which shear forces are predominant and the post-operative fracture healing prognosis is not favorable [4]. A meta-analysis describing complications of femoral neck fracture in young adults revealed an avascular necrosis rate of 14%, a re-operation rate of nearly 18%, a non-union rate of 9%, and an implant failure rate of approximately 10% [5].
At present, internal fixation remains the gold standard for the treatment of femoral neck fractures in young adults and for the treatment of non-displaced femoral neck fractures in older adults [6]. The cannulated compression screw (CCS) is most commonly used because of its unique advantages, when compared with other types of internal fixation, in the treatment of femoral neck fracture, including less soft tissue damage, less bleeding, and convenience [7]. However with increased fracture angles in unstable femoral neck fractures, the CCS has a higher rate of post-operative failure from causes such as screws retraction, femoral neck shortening, and fracture varus displacement [8]. It is still controversial which type of internal fixation can better maintain fracture stability, promote fracture healing, and avoid complications [9].
In this study, through the summary and analysis of current clinical cases, we designed a new headless compression supporting screw (HCSS) to treat femoral neck fractures, especially unstable femoral neck fractures (patent no. ZL 2019 2 23314131.7). Finite element analysis is an important stress analysis tool in biomechanical research that can be used to provide a reliable basis for the selection of new internal fixation methods for the clinical treatment of femoral neck fractures with the ultimate goal of decreasing the complication rate for unstable femoral neck fractures. In this study, we performed finite element analysis to compare the biomechanical performance of the HCSS and the CCS.
Materials and method
Construction of femur model
Because cadavers are too different from those of the skeletal cortex in young patients, whereas the biomechanical properties of synthetic composite femurs are similar to those of the femurs of young adults [10], we used fourth-generation composite femurs (left, Model 3406; Sawbones, Vashon, WA, USA) as specimens rather than using cadaveric bones. The flexural and torsional stiffness of the fourth-generation composite femur is within the range of flexural and torsional stiffness values reported for the femur in healthy adults below 80 years old [11].
Computed tomography (CT) scans (Siemens 64-slice spiral, Erlangen, Germany) of the composite femurs were taken, and the images were stored in DICOM format. These data were imported into the interactive medical imaging control system Mimics 17.0 (Materialise, Leuven, Belgium), and a model was constructed through threshold segmentation, dynamic growth, editing masks, and filling based on the different CT values of intact femoral cortical and cancellous bone tissue and was then imported into the reverse engineering software Geomagic Studio 12.0 (Geomagic, Morrisville, NC, USA) for smoothing, clipping, offset, and Boolean subtraction to produce a more detailed and accurate femoral neck model.
Construction of unstable femoral neck fracture model
The Pauwels III femoral neck fracture was modeled by 3-matic (Mimics 17.0, Materialise, Leuven, Belgium). We first created a horizontal plane and then a cutting plane across the center of the femoral neck at an angle of 50° with respect to the horizontal plane. The cut was used to simulate a Pauwels III fracture (Fig. 1).
Outputted it in STL format and then imported it into the reverse engineering software Geomagic Studio 12.0 for smoothing, clipping, offset, and Boolean subtraction to produce a more detailed and accurate entity femoral neck model. Saved output in Iges format.

Building the Pauwels type III femoral neck fracture model (fracture angle = 50°).
Traditional CCSs were selected as the control for the comparison to verify the biomechanical performance of HCSS. According to clinical fixation programming and engineering geometric data modeling, Unigraphics NX 8.0 (Siemens, Erlangen, Germany) was used to create models of 2 CCSs (CCS 1: thread diameter 7.3 mm, thread length 16 mm, thread pitch 2.5 mm, shaft diameter approximately 5.0 mm, total length 90 mm; CCS 2: thread diameter 7.3 mm, thread length 16 mm, thread pitch 2.5 mm, shaft diameter approximately 5.0 mm, total length 85 mm) and 2 HCSSs (HCSS 1: thread diameter 7.3 mm, length of head end threaded part 16 mm with 1.75 mm pitch, length of trailing end threaded part 10 mm with 1.45 mm pitch, total length 90 mm; HCSS 2: thread diameter 7.3 mm, length of head end threaded part 16 mm with 1.75 mm pitch, length of trailing end threaded part 10 mm with 1.45 mm pitch, total length 85 mm (Fig. 2, Table 1). Saved output in Iges format.
The specific dimensions of CCS and HCSS
The specific dimensions of CCS and HCSS
A parallel inverted triangle configuration of three screws was used for fixation in this study because (1) this arrangement provides cross-sectional anti-rotation stability for the tip of the fracture, which contributes to stabilization of the coxa varus angle [12]; (2) compared with an equilateral triangle arrangement, the inverted triangle arrangement avoids subtrochanteric stress concentration and effectively reduces the incidence of secondary femoral subtrochanteric fracture [3,13]; and (3) a number of studies have confirmed the relationship between the location and beneficial effects of CCSs and have found that the inverted triangle configuration exhibits better strength and stability [3,13]. Moreover, the inverted triangle configuration can provide optimal axial and torsional rigidity [3,14].

3D geometry models of the CCS (a) and HCSS (b).
The screw and femoral osteotomy models were respectively assembled in HyperMesh 12.0 (Altair) and the finite element models were generated. Three screws were arranged parallel to the axis of the femoral neck in an inverted triangle configuration (Fig. 3) [15]. The assembled model was meshed by Solid186 tetrahedron elements in the HyperMesh, and the grid convergence calculations are tested by different sizes. The number of elements and the total numbers of nodes is shown in Table 2.

Geometric modeling of internal fixation of the femoral neck fracture with CCSs (a), HCSSs (b).
After generating nodes and units, they are imported into Ansys 18.0 (ANSYS, Pittsburgh, PA, USA) for processing and analysis. The material attributes (modulus of elasticity and Poisson’s ratio) were assigned by Ansys (Table 3) [12]. The synthetic composite bone was assumed to be homogeneous and isotropic with linear elastic properties defined by the manufacturer and from previous studies [12–14]. Friction between model parts was included in constraint definitions. The interface between the surface of the threaded portion of the screw and bone was defined as a contact constraint, and the coefficient of friction between the screw shaft and the bone was set at 0.3. The fracture surface completely divided the bone; the surfaces remained in a state of complete contact, and the friction coefficient between these surfaces was set at 0.46 [16]. To simulate the single leg standing position, a force of 2100 N was applied to the center of the femoral head (approximately equal to 3 times the body weight), and the assembled model was abducted 10° and tilted posteriorly by 9° (Fig. 4) [17].
Details of the models
Details of the models
Bone and internal fixation material properties

Model constraints: fixed at the bottom and 2100 N force applied to the femoral head.

Displacements of the femur (a, b) and the screws (c, d).

Deformation of the femur for the CCS (a) and the HCSS (b).

Stress within the femur: CCS (a–d) and HCSS (e–h).

Elastic strain within the femur (a, b) and the screws (c, d).

Rotation (change in angle) after stress loading for CCS (a) and HCSS (b).
All results are shown in Table 4.
Parameters results
Parameters results
In all models, displacement occurred at the femoral head. As shown in Fig. 5a and 5c, the maximum displacements were 10.910 mm and 10.769 mm for the CCS and the HCSS, respectively. Displacements of the internal fixation occurred at the screw head, and the maximum displacements were 10.729 mm and 10.625 mm for the CCS and the HCSS, respectively (Fig. 5b and 5d); therefore, the internal fixation displacements of the HCSS were smaller, suggesting better resistance to shear force and bending force.
Vertical displacements of the femoral head center were 1.693 mm of CCS and 1.650 mm for HCSS, respectively (Fig. 6a and 6b); and that for the HCSS was less than that of the CCS, which demonstrated that the HCSS was better at resisting axial compressive stress than the CCS.
Stress distribution
Peak von Mises stresses within the femur were 138.730 MPa for the CCS and 76.136 MPa for the HCSS (Fig. 7a and 7e). The peak stress was lower for the HCSS, which demonstrates that it provides stronger support and stability to the femur, the stress distribution within the femur was relatively dispersed, except for uniform distribution along the fracture and stress transferred to the cortical bone by the tail of the screw.
Peak von Mises stresses within the screws were 378.470 MPa and 282.460 MPa for the CCS and the HCSS, respectively (Fig. 7b and 7f). The peak von Mises stress within the CCS was higher than that within the HCSS. Peak von Mises stresses within the upper two screws were greater than that within the lower screw, and stresses within the HCSS were not only distributed in the screw graft, but were also distributed in the tail of the screw, which indicates that the double-thread design effectively dispersed stresses. The maximum stress area of HCSS is larger and the stress is more dispersed.
Elastic strain
The maximum strains within the femur were 0.0170 and 0.0157 for the CCS and for the HCSS, respectively (Fig. 8a and 8b), and the maximum strains within the screws were 0.00389 and 0.00269 for the CCS and for the HCSS, respectively (Fig. 8c and 8d); therefore, under the same load, there was less deformation of the femur and of the screw itself for the HCSS.
Rotation angle
Before loading, the angle of the fracture line and the horizontal line was 50°. Rotations after loading (change in angle of the loading force) were 0.550° and 0.420° for the CCS and for the HCSS, respectively (Fig. 9); the HCSS had a smaller rotation.
Discussion
Compared with the CCS, the HCSS demonstrated good biomechanical stability by (1) providing better support-the HCSS resulted in a smaller maximum displacement of the internal fixation and lower peak von Mises stresses within the femur demonstrating better resistance to shearing and bending; (2) being better able to prevent shortening of the femoral neck-the HCSS resulted in a smaller vertical displacement of the femoral head and less maximum strain within the femur, which demonstrated better resistance to axial compression; and (3) exhibiting good ability to stabilize-the HCSS resulted in a more dispersed stress distribution within the femur and within the screw, lower maximum strain within the screws, and a smaller change in loading angle, which demonstrated better resistance to shearing, deformation, and rotation.
The femoral neck fracture is an intra-articular capsular fracture. Currently, it is generally believed that the therapeutic principles of femoral neck fracture should refer to intra-articular fracture. Treatment emphasizes accurate reduction and stable fixation [18,19]. Some studies indicate that compressive stress, tensile stress, and torsional stress are the main stresses on the femoral neck during activities, and compressive stress applied to the fracture can generate shear force [20]. Pauwels III fracture is a more vertical fracture line with obvious displacement. The greater the shear stress at the fracture, the greater the fracture instability, and it will be more difficult for the fracture to heal [4]. Because of the required consideration of stress distribution and anatomical characteristics, the treatment has attracted great attention from clinicians.
Patients under the age of 65 years without chronic disease should undergo emergency surgery for open reduction with internal fixation such as CCS, dynamic hip screws (DHS), 1/3 tubular plate with cannulated screws, or medial anatomical buttress plate with cannulated screws, but which treatment is optimal for internal fixation remains a matter of debate [21]. Because of the simplicity and small amount of trauma caused by its use, the CCS has become the primary treatment for femoral neck fracture. Moreover, use of the CCS applies a sliding compression force that makes the fracture surfaces close and maintain contact, which facilitates healing [22]. However, with increasing fracture angles, the incidence of post-operative implant failure increases. A loss of internal support for the unstable femoral neck fracture and an insufficient holding force applied by the CCS to the lateral cortical bone can easily lead to the loosening of the cannulated screws and femoral head varus after surgery, thus leading to the failure of internal fixation [23]. The sliding compression effect of the CCS (i.e., secondary compression during post-operative healing) can easily lead to shortening of the femoral neck [8]; if femoral neck shortening exceeds 5 mm, it will have a greater impact on the function of hip joint and on the patient’s activities [23,24]. In an age-independent multi-center study, Zlowodzki et al. found that 30% of non-displaced fractures and 52% of displaced fractures treated with multiple cannulated screws healed with a >10 mm shortening of the femoral neck [25]. In our opinion, the main reason for this phenomenon is that this type of fixation is not sufficient to maintain fracture stability post-operatively. To solve this problem, the dynamic hip screw (DHS) was developed. Although use of DHS for the treatment of Pauwels III fractures of the femoral neck results in greater biomechanical stability than use of CCSs, as a sliding compression device, the DHS produces more pronounced femoral neck shortening than CCS. Biomechanical studies have shown that DHS only slight increased resistance to shearing [26]. In addition, the use of a single DHS results in less ability to resist rotation compared with the use of CCS. Therefore, patients receiving DHS treatment are more likely to have femoral head rotation [27,28]. To improve the stability of internal fixation, a 1/3 tubular plate or medial anatomical buttress plate is added when using cannulated screws. However, such internal fixation usually requires two incisions, causing a large amount of trauma and a long operation time. Implantation of a steel plate may affect the blood supply of femoral head though proper screw placement can reduce damage to the local blood supply. Therefore, to improve fixation strength and stability and avoid further shortening of the femoral neck, we designed a new HCSS.
The HCSS has threads on both ends, with a smaller pitch on the tail than that on the head. Because of the difference in pitch, the distance of each end of the HCSS to the distal and proximal ends of the fracture is different when the tail end is inserted into the lateral cortical bone. It can provide pressure on the fracture end to achieve the effect of pressure. Some studies have found that, in the treatment of Pauwels type III femoral neck fractures, adding lateral locks can provide angular stabilization and increase resistance to shearing; the basic principle may be reduced micro-motion around the fracture site [29,30]. Kuan et al. also concluded that combining loop wire with three cannulated screws greatly reduced micro-motion and improved biomechanical performance [31]. In this study, a similar effect was achieved with HCSS–the tail of the three screws combined with the lateral cortical bone to form a structural complex, because of the strength and complete structure of the lateral cortical bone of the femur, which played a role in angular stabilization similar to that of the locking plate. Compared with traditional point fixation with CCS, this 3D fixation has increased shear strength and reduces screw fretting, thus making the overall structure of the implant system more stable; we found that the maximum displacement of the HCSS was smaller than that of the CCS, which reflected that the overall structure of the HCSS group was more stable.
When standing, vertical shear force is exerted on the femoral head. When stress near the fracture in the femoral neck is over-concentrated, bone destruction and bone resorption may occur. Because CCSs lack locking support, they cannot prevent excessive stress concentration near the fracture and do not prevent shortening of the femoral neck as a result of bone destruction and resorption. Our HCSS design, in which the thread of tail end is locked to the bone cortex, provides support and elevates the femoral head, reducing excessive stress concentration between along the fracture, and maintains mechanical stability after reduction, with intra-operative compression at the fracture, thereby reducing or avoiding shortening of the femoral neck during healing. The thread of tail also prevents the screw from being pulled out, and the headless design is convenient for closing and minimally invasive penetration, which reduces interference to the internal biological environment (within the fracture and in the surrounding soft tissue) and is conducive to healing.
Although stress concentration occurred below the fracture end and the joint of lateral cortical bone with screw junction with both types of internal fixation, the peak von Mises stress within the femur for the HCSS was lower than that for the CCS. The stress distribution for the HCSS was also more uniform and more dispersed, and stress was better transmitted vertically along the medial and lateral cortical bone of the femur. The HCSS was superior to CCS in resisting loosening, failure of the implant, and shearing. Similarly, elastic strain within the femur for the HCSS was less than that for the CCS, indicating less deformation under the same conditions, which proves that the HCSS resulted in better stability.
This study had some limitations. First, only the parallel inverted triangle configuration, which showed better mechanical behavior in previous CCS configuration studies, was used. Second, although many types of internal fixation implants, such as DHS and locking plates, are widely used in the treatment of unstable femoral neck fracture, only CCS and HCSS were compared. In future finite element analysis studies, we plan to include other types of internal fixation and other configurations. Finally, direct experimental validation was not performed in this study, which is a common limitation of similar simulation research; however, we intended to compare biomechanical behavior under the same loading environment and boundary conditions rather than determine precise values of response. Therefore, the lack of validation is justified. However, we indirectly verified the validity of the experiment. The high stress levels observed in the central part of CCS in this study are consistent with those observed and reported in previous studies [17,32].
Conclusion
The HCSS possesses better biomechanical stability than the CCS. For Pauwels type III fractures, the HCSS exhibits better resisting shearing and support, providing a new clinical treatment; however, further biomechanical and clinical trials are needed to verify the results reported herein.
Footnotes
Acknowledgements
The authors thank Dr. Coren Walters-Stewart for editing a draft of this article.
Author contributions
Yang Xue: Data curation, Investigation, Writing-review and editing, Software. Xiong-Fei Wang: Writing-review and editing. Fu-Long Zhao: Investigation, Software. Da-Cheng Han: Software, Formal analysis. An-Hua Long: Software, Resources. Jin Wang: Formal analysis, Software, Supervision. Xue-Fei Wang: Supervision, Project administration.
Conflict of interest
None to report.
