Dislocation of the femoral component is one of the most frequent complications of
total hip arthroplasty (THA). There are many factors contributing to the development
of dislocations, which can be grouped into three main categories: patient-dependent
factors, implant-dependent factors, and factors associated with operative technique.[1 ] The best option in the “treatment” of a THA dislocation is to prevent the development
of primary instability.[2 ]
[3 ] However, there are cases in which even with the proper installation of the endoprosthesis
components, the dislocation results from weakness or a defect in the capsule–ligament
apparatus; for example, it can develop during repeated revision operations, or the
consequences of severe injuries of the proximal femur can lead to the excision of
massively expanded scar tissue. The dislocation or dysplastic changes can also occur
after operations which require limb lengthening and so on.
The issue of the restoration of the capsule–ligament apparatus is particularly relevant
in people with obesity, which is clinically characterized by the deposition of fat
in different parts of the body, including the thighs and anterior abdominal wall.
Thus, due to excessive deposition of adipose tissue, during the squat, a mechanical
disturbance occurs and an additional fixation point is created, around which additional
force arises, which can be significant and in some cases reach 20% of body weight.[4 ] It affects the kinematics of movement with full flexion and increases the risk of
dislocation of the endoprosthesis head from the acetabulum cup ([Fig. 1 ]).
Fig. 1 Development of a hip joint endoprosthesis dislocation in obese patients. (A ) A sagittal plane at the time of maximum flexion in a patient not suffering from
obesity. (B ) Unfavorable kinematics in a patient with obesity, with the occurrence of an additional
point of fixation followed by a torque of the head of the prosthesis, leading to a
posterior hip dislocation (Elkins Jacob Matthias, University of Iowa, 2013).
In the case of posterolateral approaches, careful restoration of the posterior structures
of the capsule and external rotators (tendons of the piriformis muscle) with nonabsorbable
sutures is one of the main conditions, though it is not always applicable. Many authors
argue that the careful restoration of soft tissue structures or, at least, the preservation
of these structures using modified approaches significantly reduces the incidence
of dislocations associated with posterolateral surgical approach. Therefore, it was
estimated that with the approach without the restoration of soft tissues, the risk
of dislocation is 8.21 times higher than with the same approach but with the restoration
of soft tissues.[5 ] In addition, the dependence of the hip endoprosthesis stability on the capsule thickness
has been proven. A joint capsule with a thickness of 1 mm weathers two times less
stress, leading to dislocation than a joint capsule with a thickness of 3.5 mm, and
three times less with a maximum capsule thickness of 6 mm.[6 ] It should be also noted that suturing even with significant damage to the capsule
by a longitudinal incision successfully restores stability to approximately 10% of
the baseline. Thus, the proven importance of the restoration and strengthening of
the structures of the capsule–ligament apparatus is in line with methods of preventing
the development of dislocation of the hip endoprosthesis, such as careful preoperative
planning,[7 ] correct installation of components, and patient management in the postoperative
period.
Currently, many methods have been developed to strengthen and restore the posterior
structures of the hip joint capsule using auto- and allomaterials, which differ both
in the method of fixation and in the characteristics of the materials themselves.
Alongside their advantages, these methods are characterized by several disadvantages
such as high manufacturing cost and the need for special skills of the operating surgeon.
We have proposed a method of restoration and strengthening of the posterior structures
of the hip joint capsule with polypropylene mesh (PPM) implants.[8 ] The essence of the proposed method lies in the fact that having all the components
of the hip joint endoprosthesis installed, PPM is applied in the projection of the
capsule defect on its outer surface while its edges are hemmed to the free edges of
the capsule, along its entire perimeter, over the entire thickness of the capsule,
thus forming a mechanical “patch” over the defect and acting as a plateau for the
formation of a durable fibrous scar. This method does not require significant financial
costs and/or special skills of the surgeon and demonstrates convincing results proved
by histological studies on laboratory animals (the study was conducted in accordance
with the Law of Ukraine on “Scientific and Scientific and Technical Activities” and
the Council of Europe's European Convention for the Protection of the Vertebrate Animals
used for Experimental and Other Scientific Purposes [Strasbourg, 1985]); bioinertness
and the safety data are based on this experiments. Moreover, during the experimental
study, physical and mechanical properties (using Young's modulus and Poisson's ratio)
of PPM with soft tissues integrated into it were collected. A computer model was built
and further calculations were performed with regard to the obtained information.
The aim of the study is to assess the strength and stiffness of the closure of the
hip joint capsule defect with PPM based on an analysis of the stress–strain condition
of the capsule models sutured by various methods.
Materials and Methods
The study was performed using a software package “ANSYS” 18.2 based on the finite-element
method. For the rational use of computer resources, a computational model was built,
which consisted of a capsule model and a model of endoprosthesis head. The bones that
form the hip joint were not modeled, and their presence was taken into account by
applying appropriate boundary conditions; moving the edges of the capsule in all directions
was forbidden.
The model of the capsule in its size corresponded to the anatomical size of the hip
joint capsule of an adult. Considering that the capsule shape follows the contours
of the femoral head and neck and has the shape of a cylindrical sleeve attached to
the edges of the acetabulum and intertrochanteric line ([Fig. 2A ]), it was modeled as a hollow cylinder with the following dimensions: length of 12 cm
with wall thickness of 3 mm ([Fig. 2B ]).[9 ] Regarding the diameter of the cylinder, the following should be noted. Considering
the fact that during the movements, the surface of the endoprosthesis head impacts
the capsule, to comply with the conditions of interaction between these elements,
the diameter of the head and the internal diameter of the cylinder had the same size,
which was 36 mm.[10 ] Taking into account the specified wall thickness of the capsule, the outer diameter
of the cylinder was 42 mm. To reduce the number of finite elements of the model, only
1/2 cylinder was considered ([Fig. 2C ]).
Fig. 2 (A ) Hip joint capsule. (B,C ) The “capsule” element of the computer model of the “capsule–head of the hip joint
endoprosthesis” system.
In regard to the head model, we note that its rigidity is much higher than that of
the capsule. In addition, the endoprosthesis is not a direct object of investigation.
Therefore, to reduce the total number of finite elements in the model, the head of
the hip joint endoprosthesis was modeled as a hollow sphere, the outer diameter of
which, as mentioned previously, is 36 mm in size and 34 mm in internal diameter, that
is, the head thickness is 1 mm ([Fig. 3 ]). To apply a load to the head of the endoprosthesis, which is later transferred
to the capsule, a rectangular element is attached to the sphere, which has a cross-section
of 1 × 1 cm.
Fig. 3 (A ) Location of the endoprosthesis head in the acetabular component. (B–D ) The “endoprosthesis head” element of the computer model of the “capsule–head of
the hip joint endoprosthesis” system.
The interaction between the sphere and the inner surface of the cylinder was performed
by creating a contact pair using software tools.
Two ways of capsule suturing were investigated in this study: interrupted stitches
and suturing with PPM, which covered the defect of the capsule and fixed to it along
the entire perimeter of the mesh through the entire capsular-ligament apparatus. Consequently,
two computational models were built. While the geometry of the two was the same, they
differed in the method of closing the dissected capsule. An additional control model
with the identical dimensions was constructed, yet the incision was not sutured here.
Capsulotomy was modeled as a cut of 0 thickness, along the basic cylinder, that is,
along the capsule model. The length of the incision is 8 cm. The incision was located
symmetrically along the height of the cylinder; therefore, the shift from the upper
and lower bases of the model is 2 cm.
The locking elements (thread and mesh) were also modeled in accordance with their
actual dimensions ([Fig. 4A, B ]). The thread diameter is 0.5 mm, the thread diameter in the mesh is 0.5 mm, and
the size of the cells is 2 × 2 mm.
Fig. 4 (A ) the element “interrupted stitches.” (B,C ) The element “polypropylene mesh” of the computer model of the “capsule–head of the
hip joint endoprosthesis” system. (D ) The polypropylene mesh used in the experiment.
The physical and mechanical properties of the model elements are as follows. For the
capsule–ligament apparatus, Young's modulus of elasticity is 150 MPa, and Poisson's
ratio is 0.25. The properties of the endoprosthesis head were chosen because of high
rigidity compared with the capsule rigidity, which amounted to 2 × 105 and 0.25 MPa,
respectively. For the thread and the mesh, the elastic properties were assumed to
be the same and corresponding: Young's modulus value was 17.2 MPa, the Poisson ratio
was 0.25. However, it should be noted that a preliminary calculation showed that modeling
the mesh with its real dimensions (cell structure) creates a large number of additional
elements in the model (lines and surfaces). Each of these elements was assigned a
number. This, in turn, leads to a significant consumption of computer resources. Therefore,
the mesh model was replaced with a fragment of a cylindrical surface ([Fig. 4C ]), the overall dimensions of which corresponded to the mesh, and the thickness coincided
with the thread diameter. This replacement required recalculation of the elastic modulus,
which amounted to 1.72 MPa. Note that the mesh modeling in the form of a solid surface
is legitimate since the difference in the structure of these objects exists only (the
presence or absence of cells). At the macro level, by defining a new modulus of elasticity
for a continuous surface, there will be no difference in the behavior of mesh models
(cell or solid structure) when it is loaded. Consequently, the impact on the capsule
from the side of the mesh model in the form of a solid surface will be the same as
in the simulation of its cell structure.
Closure of the defect was simulated by stitching and the mesh. The sutures were placed
with a step of 1 cm at a distance of 1 cm from the edges of the cut, and therefore
the total number of stitches was seven ([Fig. 5B ]). The shifts from the axis of the incision were also 1 cm each. The sizing of the
mesh corresponded to the sizes of the cut, that is, 8 cm along the cut line, and with
the intent of 2.5 cm from its axis ([Fig. 5C ]). The mesh and the surface of the capsule were connected by cylindrical elements
that simulated sewing. The dimensions and properties of the cylindrical elements corresponded
to the sizes and properties of the thread used to sew the mesh to the capsule and
were applied with a step of 1 cm along the perimeter of the mesh.
Fig. 5 (A ) Polypropylene mesh implanted to close the capsular defect. (B ) Computer model of the “capsule–head of the hip joint endoprosthesis” with the closure
of the defect of the capsule with an anchor stitch. (C ) Polypropylene mesh. (D ) Computer model of the “capsule–head of the hip joint endoprosthesis” system without
the closure of the defect.
Another contact pair was created to connect the locking elements and the capsule.
To study the strength and stiffness of the dissection fixation in various ways, two
types of loads were applied to a rectangular element of the head model: static, in
the form of a fixed force, and kinematic, in the form of a fixed displacement. In
all cases, the vector of the load application was directed along the normal to the
cutting line. Given that the load on the center of the incision is considered as the
most dangerous effect, the study was performed under the assumption that the head
is located symmetrically with respect to the incision line both in length and axis.
The amount of movement of the endoprosthesis head in the direction of the incision
was chosen from the assumption that one-fourth of the head diameter makes 9 mm. The
magnitude of the static force was 15 kg. It is noteworthy that since under all the
same conditions the models differ only in the methods of fixation, the magnitude of
the loads can be chosen arbitrarily to compare their effectiveness.
The capsule model was fixed along the entire plane of the upper and lower bases; movement
in all directions which imitated the attachment of the capsule to the bone surface
was prohibited. To this, the corresponding boundary conditions were superimposed on
the edges located on the side of the rejected part of the cylinder, which ensured
the immobility of the indicated edges of the model in the direction of the load vector.
Computing of finite elements was performed by the generator of grids of the software
complex. The element type selected was solid . The size of the finite element was set on the lines of the objects and varied from
0.25 to 1.0 mm. The created contact pair, “capsule–head of the hip joint endoprosthesis,”
suggested the absence of friction.
Results and Discussion
Images of the distribution of the stress–strain state in the “head–capsule” system
were calculated. To assess the effectiveness of the method of the closure of the capsule
with regard to stiffness as the main characteristics, the values of the opening of
the incision, as well as stresses arising in the joint capsule, were defined. The
stress in the head is considered as an additional characteristic. The results are
shown in the respective tables ([Tables 1 ]–[3 ]).
Note that at the first stage of the study, a kinematic calculation was performed,
which is aimed at studying the rigidity of fixation. Since the specified displacement
of the head will be the same for all models, the opening of the cut may also be the
same or similar in values for different models. An indicator of stiffness will be
the magnitudes of the stresses arising in the model since the more rigid model is
more resistant to the applied loads, which manifests itself in an increase in stresses.
Calculations have shown that for a given loading pattern, the largest displacements
occur in the center of the section, that is, in the place of the impact of the head
on the capsule. In this case, the maximum amount of movement does not occur on the
cutting line but is displaced from it in the circumferential direction, which is associated
with the deformation of the capsule. Therefore, this value cannot be the main indicator
of disclosure. Thus, to estimate the magnitude of the disclosure, displacements of
points located on the incision line inside (δ
ís ) and outside (δ
os ) of the capsule were used.
[Table 1 ] demonstrates the way in which the method of suturing the incision influences the
values of maximum displacements (δ
max ) occurring in the capsule, which are directed toward the opening, that is, around
the circumference of the cylinder. It also presents the movements of points located
on the surface of the section in the same direction (δ
is , δ
os ), that is, deviations of the edges of the incision from its line.
Table 1
The values of the displacements and stresses in the system “capsule–head of the hip
joint endoprosthesis” with the kinematic calculation
Fixation method
Displacement, mm
Stress σMiz , MPa
δ
max
δ
is
δ
os
Capsule
Head
Model without closing capsulotomy (control model)
6.97
5.79
6.97
20.9
27.4
Model of the system with the closure of the defect with sutures
6.35
5.37
6.04
26.9
54.5
Model of the system with the closure of the defect with polypropylene mesh
5.85
4.94
5.72
33.3
87.4
As can be seen in [Table 1 ], the values of movements at the edges of the incision inside and outside the capsule
are different. Moreover, the amount of movement outside is greater than that on the
inner surface, that is, there is a reversal of the edges of the cut. The reversal
pattern is shown in [Fig. 6 ].
Fig. 6 The location of the points of the capsule with the largest displacements.
Analysis of the results given in [Table 1 ] showed the following. The smallest displacements in the direction of opening were
obtained from the model of fixation of the incision with PPM and amounted to 5.85 mm.
In case the incision was fixed with a suture, this value was 6.35 mm, which was to
be 0.5 mm or 8.5% higher. The largest displacements were obtained for the control
model (without fixation), which were equal to 6.97 mm, and were larger than the models
with fixation with threads and a mesh of 1.12 mm or 19.1% and 0.62 mm or 9.8%, respectively.
Regarding the deviation values, it can be noted that in the model of fixation with
the mesh, it turned out to be also the smallest both on the inside and on the outside.
The difference in these values was 1.56 mm. In the fixation model, the thread deviation
was higher by 8.7% from the inside and by 5.6% from the outside, and the difference
was 1.34 mm. In the control model, deviations were greatest and exceeded these indicators
by 17.2% from the inside and by 21.9% from the outside for the model with a mesh and
by 7.8% from the inside and 15.4% from the outside for the model with a thread. The
difference in deviations for the control model was 2.36 mm.
Given that δ is the deviation of the points of the capsule from the axis of the incision, full
disclosure is determined from the following correlation:
Δ = 2*δ .
Thus, the full values of disclosures are given in [Table 2 ].
Table 2
The size of the opening of the incision depending on the model of fixation in the
kinematic calculation
Fixation method
Displacement, mm
Δmax
Δis
Δos
Model without closing capsulotomy (control model)
13.94
11.58
13.94
Model of the system with the closure of the defect with sutures
12.7
10.74
12.08
Model of the system with the closure of the defect with polypropylene mesh
11.7
9.88
11.44
To visualize the calculations, [Fig. 7 ] shows patterns of the distribution of displacements in the model in the circumferential
direction. The species are given in full ([Fig. 7A, B ]), as well as in the longitudinal ([Fig. 7C ]) and transverse ([Fig. 7D ]) sections in relation to the cut line. Only the capsule and the head without the
fixing element are shown. Note that the distribution pattern of these movements is
the same regardless of the fixation model; therefore, in [Fig. 7 ], for the type of the control model is shown (without fixation of the incision).
Fig. 7 The distribution of deformations in the “capsule–head of the hip joint endoprosthesis”
model.
To assess the stress in the “capsule–head of the hip joint endoprosthesis” system,
the values of Mises equivalent stresses both in the capsule and in the head were defined.
[Table 1 ] demonstrates that the greatest stress in the capsule develops in the model with
mesh fixation on the line of intersection of the transverse plane of the model symmetry
and the longitudinal edge of the model of the capsule from the inside ([Fig. 8A ]). The magnitude of these stresses was 33.3 MPa. In case of the fixation with a thread,
the magnitude of these stresses was 19.2% less and equaled to 26.9 MPa. The stress
also developed in the transverse plane of the model symmetry but in the center of
the section from the outside ([Fig. 8B ]). The lowest loads were obtained in the control model (20.9 MPa) and were 37.2%
less compared with the model with a mesh and 22.3% less than the model with a thread.
These stresses appeared in the center of the section from the inside of the model
([Fig. 8C ]).
Fig. 8 Stress distribution in the “capsule–head of the hip joint endoprosthesis” model.
(A–C ) In the capsule. (D ) In the head.
The magnitudes of the maximum stress in the head were distributed between the models
in the same way. The major stress was fixed in a model with the mesh and peaked 87.4
MPa. In the model with the thread fixation, the stress was 37.6% less. And the smallest
values were obtained from the model without fixation, which is 68.6 and 49.7% less
compared with the models with mesh and thread fixation methods, respectively. The
indicated stress developed at the point of the ball's connection with the parallelepiped
([Fig. 8 ]), that is, in the stress concentrator, which explains their value.
As mentioned previously, the kinematic calculation proved that the stress distribution
indicates the rigidity characteristics of the studied models. The higher the stress
under the same load conditions, the more rigid the model. The obtained values show
that in regard to fixation rigidity, the model with the mesh is more rigid.
At the next stage of the work, a static calculation was performed, that is, the magnitude
of the applied load was recorded. It is noteworthy that in case of a fixed amount
of force, models with different stiffness will also have different displacements—the
harder the model, the less displacement. Therefore, the main indicator of this study
is the size of the opening of the section, and the stress values of the model can
be used to further assess the strength.
Analysis of the results in [Table 3 ] showed that the smallest displacements in the direction of opening were obtained
from the model of fixation of the incision with a mesh and amounted to 1.01 mm. When
the incision was fixed with a thread, this value was 1.68 mm, which turned out to
be higher by 0.67 mm or 66.3%. The largest displacements were obtained from the control
model (without fixation), which were equal to 3.42 mm, and were larger than the models
with fixation by threads and mesh by 1.74 mm or 103.6% and 2.41 mm or 238.6%, respectively.
Table 3
The magnitude of the displacements and stresses in the system “capsule–head of the
hip joint endoprosthesis” with a static calculation
Fixation method
Displacement in the capsule δ , mm
Stress σMiz , MPa
Head displacement w , mm
δ
max
δ
is
δ
os
Capsule
Head
Model without closing capsulotomy (control model)
3.42
2.87
3.42
10.5
13.6
3.90
Model of the system with the closure of the defect with sutures
1.68
1.43
1.60
8.12
13.8
2.06
Model of the system with the closure of the defect with polypropylene mesh
1.01
0.86
0.99
7.59
13.5
1.38
It is noteworthy that in the kinematic calculation as well, the values of displacements
at the edges of the incision inside and outside the capsule differ ([Table 3 ]).
Taking into account the ratio of 2 * δ (δ is the deviation of the points of the capsule from the axis of the incision), we
obtain the full disclosure of the incision, which is shown in [Table 4 ].
Table 4
Values of incision disclosure depending on the model fixing during the static calculation
Fixation method
Displacement, mm
Δmax
Δis
Δos
Model without closing capsulotomy (control model)
6.84
5.74
6.84
Model of the system with the closure of the defect with sutures
3.36
2.86
3.20
Model of the system with the closure of the defect with polypropylene mesh
2.02
1.72
1.98
Regarding the magnitudes of the disclosure, it can be noted that in the model of fixation
with the mesh, it was also the smallest both on the inside and on the outside. The
difference in these values is 0.26 mm. In the fixation model, the opening was higher
by 66.3% from the inside and by 61.6% from the outside, and the difference between
them was 0.34 mm. In the control model, the disclosures were greatest and exceeded
these indicators by 233.7% from the inside and 245.5% from the outside for the model
with a grid and by 100.7% from the inside and 113.8% from the outside for the model
with a thread. The difference in the disclosures for the control model was 1.10 mm.
It should be noted that in [Table 3 ], in addition to the values described previously and used to assess the effectiveness
of the method for closing the capsule (displacement/stresses in the capsule and head),
the magnitudes of displacement of the head also depend on the method of fixation.
This is because when a fixed force is applied to the head, the movements in the model
are not controlled but are the result. Therefore, the opening values are the indicator
of not only stiffness but also the amount of movement of the head in the direction
of the cut (w = displacement in the direction of the applied force).
[Table 3 ] shows the smallest movements in the model of fixation with a mesh, which is 1.38 mm.
In the model of fixation with thread, this value was equal to 2.06 mm, which is 49.3%
more. The largest displacements of the head obtained for the model without fixation,
which equaled 3.90 mm, which larger than the model with the mesh by 182.6 and 89.3%
than the model with the thread, respectively.
Note that displacements in models under static load are distributed in the same way
as in the kinematic calculation ([Fig. 7 ]). However, the magnitudes of displacements in the static calculation differ significantly
for different models. Therefore, [Fig. 9 ] shows patterns of the distribution of displacements in the circumferential direction
for each model. To compare the magnitudes of the deformations of the models, the images
are shown on the same scale. As can be seen from [Fig. 9 ], the smallest disclosure is fixed in the mesh fixation model ([Fig. 9C ]), while fixation with a thread is medium ([Fig. 9B ]), and the largest values are fived in the model without fixation ([Fig. 9A ]).
Fig. 9 Deformations of the “capsule–head of the hip joint endoprosthesis” system in static
calculation.
To estimate the stress state, Mises equivalent stresses both in the capsule and in
the head were selected.
[Table 3 ] shows the greatest stresses in the capsule arise in the model without fixation,
which, in magnitude, equaled 10.5 MPa. During fixation with a thread, these stresses
are 22.7% less and amounted to 8.12 MPa. The lowest stresses occurred in the mesh
fixation model (7.59 MPa) and are less by 27.7% than the control model and 6.53% less
than the model with the thread. These stresses arose, in all models, at the point
of contact between the head and the capsule.
The maximum stresses in the head are not practically differing in magnitude for different
models of fixation. The largest of this stress is achieved with the thread model and
amounted to 13.8 MPa. The model without latching stress is less by 1.45%. And the
smallest ones are obtained for a model with a mesh, which is smaller than that of
models with thread fixing and control by 2.17 and 0.74%, respectively. The indicated
stresses appeared, as in the kinematic calculation, at the point of the sphere's connection
of the head with the parallelepiped ([Fig. 8 ]), that is, in the stress concentrator.
As mentioned previously, in a static calculation, the distribution of displacements
indicates the rigidity characteristics of the models under consideration. The lower
the displacement under the same loading conditions, the more rigid the model. The
resulting movements show that in terms of rigidity of fixation, the model with the
mesh is more rigid. In addition, the magnitude of the resulting stresses in the capsule
indicates that from the point of view of strength, the mesh model is also more durable.
