Introduction
Despite their rare incidence (ranging from 8% to 26%) compared to that of other proximal
or articular hip lesions, femoral head fractures, described in 1869,[1 ] have great clinical and scientific importance due to their surgical complexity and
predisposition to the development of severe dysfunctions.[2 ] With an etiology linked to automobile accidents with high energy impact, they are
usually accompanied by posterior hip displacements (ranging from 5% to 15%).[3 ]
[4 ] The conservative approach to treatment does not present good results, so open reduction
and internal fixation is the main recommendation.[3 ]
[5 ] Surgical planning is dependent on the severity of the lesion, and it is initially
guided by the Pipkin classification (from types I to IV)[1 ]
[2 ]
[3 ]
[4 ]
[5 ]
[6 ] ([Fig. 1 ]).[7 ]
Fig. 1 The Pipkin classification. (A ) Type I: fracture of the femoral head inferior to the central fovea. (B ) Type II: fracture extended superiorly to the central fovea. (C ) Type III: any fracture of the femoral head with associated femoral neck fracture.
(D ) Type IV: any fracture of the femoral head with associated acetabular fracture.[7 ]
Type-II lesions present great controversy regarding approach and fixation models,
with little scientific research.[2 ]
[3 ]
[8 ]
[9 ]
[10 ]
[11 ] Previous studies (case reports and retrospective analyses) show favorable results
with the use of cortical screws of minifragment (measuring between 2.0 mm and 2.4 mm)[9 ] and Herbert screws.[8 ]
[10 ] On the other hand, despite their differentiated compression capacity compared to
that of other syntheses and their wide clinical use, 3-mm canulated screws have been
associated with a greater predisposition to develop osteoarthritis.[12 ]
The scientific foundation for the direction and surgical clinical planning of Pipkin
type-II fractures is composed of case reports and longitudinal studies. The lack of
mechanical tests based on validated methodologies makes it difficult to practice a
solid evidence-based medicine, a tangible fact due to its low frequency. As far as
we know, the biomechanical investigation of the treatment of Pipkin type-II fractures
through the finite-element method (FEM) has so far been neglected. The possibility
of performing complex biomechanical studies that enable the visualization of the mechanical
performance of the synthesis and fracture under analysis demonstrates the potential
and explains the use of the FEM.[13 ]
[14 ]
Thus, we aimed to evaluate the biomechanical ability of two forms of fixation of Pipkin
type-II fractures (3.5 mm cortical screw and Herbert screw), describing the vertical
fracture deviation, the maximum and minimum principal stresses, and the Von Mises
equivalent stress in the syntheses used. The present is the first FEM biomechanical
report comparing two treatments for Pipkin type-II fractures.
Materials and Methods
Dimensional Characteristics and Bolt Insertion Technique
The Herbert screw was formatted with a larger diameter of its threads, of 4.3 mm and
5.3 mm, distal and proximal respectively, and 3.3 mm in diameter in its body (threadless
area). The 3.5-mm cortical screw, on the other hand, carefully respected the dimensional
similarities for each structural part, sold by Depuy-Synthes (Raynham, MA, United
States) ([Figs. 2C ] and [D ]).
Fig. 2 Conditions and contours of the tests. (A ) Frontal view. (B ) Lateral view of a model representing a Pipkin type-II femoral head fracture and
the position during the essay. Green area at the base of the femur: fixation point.
Pink area on the femoral head: loading area. (C ) 3.5-mm cortical screw. (D ) Herbert screw.
Analysis though the Finite-Element Method (FEM)
Tomographic images of a medium-sized synthetic femur (Sawbones, Vashon Island, WA,
United States, fourth generation, model 3403-103, of 10 pounds per cubic foot (PCF)
on the left side were used for the analyses.
In the study through the FEM, the materials used are divided according to their characteristics
into ductile and non-ductile. Metallic materials, for example, the synthesis, belong
to the ductile group, and their tension is measured by Von Mises equivalent stress
test. However, the von Mises equivalent stress is not used in bone analysis, since
bones belong to the non-ductile group of materials, and the maximum and minimum principal
stresses are more suited for their evaluation.
We will now describe in detail the variables for the analysis of the von Mises equivalent
stress and of the maximum and minimum principal stresses.
Von Mises equivalent stress: the stress of materials with metallic characteristics
is measured by the Von Mises equivalent stress test, which consists of a magnitude
proportional to the distortion energy used in failure tests of ductile materials in
which the failure of the material is predicted, regardless of the stress/strain status,
which means that the traction and compression stresses are considered equal and treated
in the same way.
Maximum principal stress: for the analysis of the maximum principal stress, there
is the traction force composed of load that corresponds to pulling the solid, and
this type of forces presents positive values, so the traction force is composed of
a load that intends to stretch or extend the solid.
Minimum principal stress: for the analysis of the minimum principal stress, the compression
force is composed of load that corresponds to compressing the solid, and this type
of force is conceptually represented by negative values, just to inform the opposite
direction of its application in relation to the maximum principal stress.
Developing the Biocad
The three-dimensional (3D) virtual models of each system (bone, synthesis) were developed
using the Rhinoceros (Robert McNeel & Associates, Seattle, WA, United States) software,
version 6, and the FEM analysis was performed in the Altair SimLab (HyperWorks, Troy,
MI, United States) software using the Altair Optistruct solver.
Based on the models of synthetic bones, tomographic images of the bone were obtained
and saved following in the Digital Imaging and Communications in Medicine (DICOM)
protocol. We used the 16-channel Emotion tomograph (Siemens Healthineers, Erlangen,
Germany) with a resolution of 512 × 512 and a distance of 1.0 mm between cuts. The
DICOM file was imported to the InVesalius(Software livre do Centro de Tecnologia da
Informação Renato Archer, Campinas, SP, Brasil) software for the 3D reconstruction
of the anatomical structure. Based on a set of two-dimensional imagesd obtained through
computed tomography equipment, the software enables the development of 3D virtual
models of the regions of interest of the system imported there ([Fig. 2 ]). After the 3D reconstruction of the DICOM images, the software enables the creation
of 3D files in the format called stereo lithography or standard triangle language
(STL).
File Conversion
In the InVesalius software, all the slices were imported to obtain the STL file with
the images that would be used in the process of obtaining the 3D solid, and with this
there is the option of multiplanar generation that shows the sagittal, coronal, and
axial views, and the volume. The development of the 3D surface is based on the volume,
in which one may select the regions of interest using masks and/or filters, which
cause the file to be hidden or portrayed according to the algorithm in question, thus
generating the 3D surface.
Simulation
The FEM was used for the simulations of the stability of the different assemblies.
First, the files were imported to the Altair Simlab software, with the identification
of each part of the digital models.
Material Properties
To perform the simulations, one must know and define the properties of the materials
of each part of the digital models, which are the cortical bone, the trabecular bone,
and the steel alloy. The properties of the materials used for the simulations are
the modulus of elasticity and the Poisson coefficient ([Table 1 ]).
Table 1
Material
Properties
Modulus of Elasticity and (MPa)
Poisson coefficient (v)
Cortical bone
17
0.26
Trabecular bone
1.7
0.26
Syntheses (steel)
193
0.33
Boundary Conditions
To define the boundary conditions, a load of 6,000 N was applied to the upper region
of the femoral head in the direction of the Z axis. No load was applied to the X and
Y axis. Loading was performed with the positioning of 20° of posterior inclination
and 0° in the axial axis, maintaining the physiological 10° of anteversion of the
femoral neck. Subsequently, the movement restriction regions (fixed) were delimited,
marked in all directions of the X, Y, and Z axes, of displacement and rotation. These
restrictions are to ensure that the system has a perfect alignment without displacement
and/or rotation ([Figs. 2A ] and [B ]).
After the control of the meshes of each part, care should always be taken to preserve
the size of the element, so that there are no contact issues between different parts
(femur and synthesis) in the simulations. The element adopted for mesh formation was
tetrahedral. The number of nodes was also established.
Analysis Criteria
The displacement of the models and the specific displacement of each fragment were
analyzed through the FEM. For the analysis of the stress in the non-ductile materials
(bone and fracture), we used the variables maximum principal stress (traction) and
minimum principal stress (compression). For the ductile (metallic) materials, the
stress analyzed was the Von Mises equivalent stress.
The variables maximum and minimum principal stresses and Von Mises equivalent stress
are principles of matter presented in the form of tension. The unit of measurement
for stress is the Pascal (Pa), and the stress forces are measured in megapascals (MPA)
and gigapascals (GPa).
The results were expressed as absolute values and percentiles through following the
equation: higher value × X = lower value × 100 (simple rule of 3), and the final percentile
value equals 100–X.
Results
Description of the Vertical Fracture Displacement in the Different Fixation Models
The vertical displacements evaluated were of 1.5 mm and 0.5 mm for the fixation models
using the 3.5-mm cortical screw and Herbert screw respectively ([Fig. 3 ]).
Fig. 3 Description of the vertical displacement of the fracture with the different fixation
models. (A ) 3.5-mm cortical screw (displacement: 1.5 mm). (B ) Herbert screw (displacement: 0.5 mm).
We observed that the Herbert screw reduced the vertical displacement at a rate of
∼ 66.6% in Pipkin type-II femoral head fractures.
Distribution of Maximum (Traction) and Minimum (Compression) Principal Stresses in
Fractures in the Different Fixation Models
The values for the maximum principal stress obtained in the upper region of the femoral
neck, adjacent to the fracture, were of 9.7 KPa and 1.3 KPa for the fixation models
with the use of the 3.5-mm cortical screw and Herbert screw respectively ([Figs. 4A ] and [B ]), representing a reduction of 87% in the local tension and a better distribution
with the use of Herbert screw.
Fig. 4 Distribution of the maximum principal stress in fractures with the different fixation
models. (A ) 3.5-mm cortical screw: 9.7 KPa. (B ) Herbert screw: 1.3 KPa. Distribution of the minimum principal stress in fractures
with the different fixation models. (C ) 3.5-mm cortical screw: -8.7 KPa. (D ) Herbert screw: -9.3 KPa.
The values for the minimum principal stress obtained in the lower region of the femoral
neck, adjacent to the fracture, were of -8.7 KPa and -9.3 KPa for the fixation models
with the use of the 3.5-mm cortical screw and Herbert screw respectively ([Figs. 4C ] and [D ]), representing an increase of 6.4% in the local tension with comparable distribution
using the Herbert screw.
Distribution of the Peak Von Mises Equivalent Stress in the Different Fixation Models
The peak values for the Von Mises equivalent stress were of 7.2 GPa and 2.0 Gpa for
the fixation models with the use the 3.5-mm cortical screw and Herbert screw respectively.
The reduction observed with the Herbert screw was of approximately 72.2%. Moreover,
the synthesis models presented their largest area of tension in the fracture line,
a site that represents a greater concern in the synthesis fracture ([Fig. 5 ]).
Fig. 5 Distribution of the peak Von Mises equivalent stress in the different fixation models.
(A ) 3.5-mm cortical screw: 7.2GPa. (B ) Herbert screw: 2.0 GPa.
Discussion
Fractures of the femoral head have historically been associated with controversial
results that depend on the synthesis used. Internal fixation must ensure stability,
preferably with compression between the fracture fragment and the rest of the femoral
head.[3 ] Because it is a rare fracture, experimental or computational mechanical essays are
extremely important, for they provide data that assist in the outcomes of the patients.
The FEM has been proven to be an efficient methodology for biomechanical research
in the field of bone fractures.[14 ]
[15 ]
Thus, through the FEM, we evaluated the vertical dsplacement of the fracture, the
maximum and minimum principal stresses, and the Von Mises equivalent stress of 2 syntheses
(3.5-mm cortical screw and Herbert screw) widely used in the treatment of Pipkin type-II
fractures. As far as we know, the present is the first FEM model in Pipkin type-II
fractures, and the first study comparing treatments using biomechanical methods. Our
results show the superiority of the Herbert screw, which causes a reduction decrease
in the vertical displacement and the distribution of the maximum principal stress
and the peak of the Von Mises equivalent stress.
The search for syntheses that promote adequate internal fixation, enabling early mobilization
and thus contributing to good clinical results was the focus of previous clinical
research, which pointed in the same direction as the biomechanical results of the
present study, demonstrating the positive effects of the Herbert screw on Pipkin type-II
fractures. In 1988, Murray et al.[10 ] performed open reduction and internal fixation with the Herbert screw in an osteochondral
fracture of the femoral head. After twelve months, the results showed an excellent
hip function and no radiographic evidence of avascular necrosis. More recently, Zaizi
et al.[8 ] reported good results with the treatment of Pipkin type-II fractures using anatomical
reduction and internal fixation by means of two Herbert screws, after two years of
follow-up. Wang et al.,[11 ] in a prospective analysis of three patients treated using Herbert screws, reported
results of satisfactory hip function assessed by the modified Merle d'Aubigné score.
Despite the numerous clinical difficulties regarding the treatment of fractures in
load-bearing joints, the possibility of compression inherent to the characteristic
of the differences in the threads (distal and proximal) of the Herbert screw, and
that for this it needs to have a larger diameter in relation to the 3.5-mm screw,
we were able to experimentally confirm its biomechanical advantage.[10 ] Furthermore, our results suggest that the ability to distribute stress and decrease
fracture dislocation are relevant factors to understand the mechanical effectiveness
of the Herbert screw. One of the indirect objectives of the present study was to qualitatively
analyze the close relationship between synthesis material and bone structure using
the interfragment compression technique. The close contact of the Herbert screw due
to the lack of need for a smooth tunnel in its technique (unlike the 3.5-mm screw)
seems to be a hypothesis for its better biomechanical results, and it may also make
a difference in terms of fracture stability in the reabsorption phase of the fractured
edges. Future studies need to improve the methodology to perform a more accurate evaluation
and determination of this clinical hypothesis ([Fig. 6 ]).
Fig. 6 Qualitative analysis of the relationship between synthesis material and bone structure.
(A ) close relationship between the Herbert screw and bone structure. (B ) The need for a smooth tunnel in the interfragment compression technique with the
3.5-mm cortical screw does not yield the same degree of relationship observed between
synthesis and bone structure with the Herbert screw.
The present is the first study to use the FEM to compare different fixation methods,
analyzing complex biomechanical variables (peak Von Mises equivalent stress and compression
and traction distribution in fractures) in Pipkin type-II fractures. The present study
has certain limitations that should be highlighted. The lack of effects of muscle
and ligament on fracture stability, bone quality and possible individual differences
in terms of gender, ethnicity, age, and previous diseases were not taken into account
during the analyses. These limitations should be evaluated in future clinical manuscripts.
