Effective Treatment of Femur Diaphyseal Fracture with Compression Plate – A Finite Element and In Vivo Study Comparing the Healing Outcomes of Nailing and Plating

Abstract

Background

The rigidity in osteosynthesis causes primary healing, and it takes longer to heal. The flexibility provided to the fixation allows micromotion between fragments which accelerates secondary healing.

Methods

In this study, the healing outcomes of nailing and plating in different fixation stabilities were investigated and compared by using a finite element tool. The clinical observational study was also performed to verify the results of the finite element analysis. The nonlinear contact analysis was performed on 5 different fixation configurations capturing nail and plate in immediate post-surgery.

Results

The finite element analysis results showed that flexibility instead of rigidity in interlock nail implantation increases the axial and shear micromotion near the fracture site by 47.4% (P < 0.05) and 12.4% (P < 0.05), respectively. For LCDCP implantation, the flexible fixation increases the axial and shear micromotion near fracture site by 75.7% (P < 0.05) and 25.3% (P < 0.05), respectively.

Conclusion

Our findings suggest that flexible fixations of interlock nail and LCDCP provide a preferred mechanical environment for healing, and hence, the LCDCP in flexible mode can be an effective alternative to interlock nail for the femur diaphyseal fracture.

Supplementary Information

The online version contains supplementary material available at 10.1007/s43465-022-00795-1.

Introduction

The long bone fracture can be treated by external as well as internal fracture fixation tools. The femur diaphyseal fracture is the most common bone injury. Internal fixation devices, such as compression plates with screws and intramedullary nails, are now widely used. The surgeons employ these implants based on the patient's post-accident diagnosis. The environment in which the fracture heals is greatly influenced by the proper supply of blood and soft tissue interactions near the fracture site [1, 2]. It was evident that the minimum surgical exposure is beneficial for early healing since the supply of blood to the bone is preserved and it can be re-established quickly [3]. Both the compression plate and the intramedullary nail can be fixed rigidly or flexibly, i.e., rigid and flexible fixation. The bone healing process is predominantly determined by the stability provided to the fracture fixation [4]. By choosing some of the screws not to be inserted, the fixation can be made flexible and it can be made more flexible with the minimum possible number of screws inserted. Secondary fracture healing is dependent on a certain amount of interfragmentary movement, which can be obtained through flexible fixation. These interfragmentary movements (axial compression and shear) can be quite large [5]. The fixation rigidity introduces a new concept of stability, i.e., absolute and relative stability [1]. The fracture treated with rigid fixation heals with absolute stability, i.e., without callus formation and the fracture treated with flexible fixation heals with relative stability, i.e., with callus formation [1]. Although, it can be understood that perfect rigidity is not possible to achieve in the clinic. Therefore, even with the rigid fixation, there will still be some callus formation occurred. The reason behind the difference in the formation of the amount of callus is the micromotion between the fracture gap. The movement of fracture surfaces in the direction of physiological loading, i.e., axial micromotion accelerates the fracture healing, but the shear movement or shear micromotion of fracture surface results in delay or nonunion of fracture [6]. During the early stages of healing, endochondral ossification results in callus development by the production of cartilage [7]. It was evident that callus formation is beneficial for fracture healing [8]. The more the callus forms, the faster the bone heals. From the past studies, it is noticeable that the less rigid fixation leads to faster healing and also boosts the quality of callus formed, but it also results in less stability to the fracture fixation [1, 9]. The surgeon has to confirm that the fixation should not be too flexible because too much interfragmentary motion leads to more callus formation but it will be less stable. It will give rise to the condition of nonunion or delayed healing because the less stable callus formed takes a much longer time to heal completely [10].

None of the studies have reported the optimal fixation rigidity that should be provided during surgery to provide the best healing environment. There have been previous studies that reported the effect of fixation stability on bone healing. None of the studies compared the performance of nailing and plating while varying the fixation stability. To treat the femur diaphyseal fracture, the interlock nail is preferred over the compression plate in the majority of cases. However, no study has proven the efficacy of interlock nails over compression plates for diaphyseal fracture. Furthermore, while the interlock nail may emerge as a better implant, a comparison with the compression plate in flexible mode has yet to be performed. The limited contact dynamic compression plate (LCDCP) was designed and developed to reduce the plate interference with cortical perfusion, thus reducing cortical porosity [11, 12]. It has advantages over conventional plating, including a lower risk of infection due to improved blood supply, higher rates of union, and a lower risk of refracture [13]. For diaphyseal fractures, osteosynthesis by plating (LCP/LCDCP) is a viable option. The locking compression plate (LCP) is more expensive than the LCDCP, but it provides more stable strut and angle fixation, making it more useful in osteoporotic bones [14]. Compression plates can be fixed to bone in a flexible mode by not inserting some of the screws, and the fixation can be made more flexible by using fewer screws. As flexible fixation allows the micromotion between the fracture fragment, which allows secondary healing, the compression plates may be an effective alternative to interlock nails for the femur diaphyseal fracture. Hence, there is a need to study the comparison of the healing outcomes of nailing and plating while varying the fixation stability.

This study aims to investigate and compare the healing outcomes of the intramedullary nail and compression plate when they were fixed with different fixation stabilities. The interlocking nail and limited contact dynamic compression plate (LCDCP) were the two implants selected for the analysis. For this study, the LCDCP was chosen as the compression plate over LCP because LCP is more useful for osteoporotic patients and the current study is only applicable to fresh healthy adult bone as the effect of osteoporosis is not considered. Also, the equally good LCDCP is less expensive and best suited to the scope of this study. Both the implants were fixed rigidly in one case with all screws inserted and flexibly in another case by not inserting some screws. There were 5 cases to investigate as represented in the flowchart (Fig. 1). The nonlinear contact analysis considering the one-legged instance physiological loading configuration is performed to simulate the biomechanical behavior of bone-implant fixation. The evaluations of the mechanical performance of these implants were done by considering the various governing mechanical factors involved in bone fracture healing. The medical-grade titanium (Ti) implants were already proven to have a lower rate of failure and fewer complications than similar stainless steel (SS) implants [15]. Also, stainless steel is twice as stiff as titanium. Internal fixation of fractures with stainless steel plates is likely to produce more rigid fixation than with titanium plates. As a result, stainless steel plate fixation is likely to result in fracture healing with minimal callus/primary healing [16]. Therefore, in this study, both the implants (Interlock nail and LCDCP) were considered to be made of medical-grade titanium material (Ti-6Al-4 V). This study does not intend to investigate the effect of implant material, which has already been done in the past with metals and fibrous composite prostheses too [6, 17]. The titanium implant material was chosen for the analysis because it is commonly used in most osteosynthesis procedures. To verify the results of finite element analysis, the clinical observational study was performed with the patients implanted with the same implants of titanium material.

Fig. 1.

Flow chart representing the fixations configurations used in the study

Finite Element Analysis

Computer-Aided Design (CAD) Models

The three-dimensional solid model of the left femur used in this analysis was employed from the repository called ‘GrabCAD Community,’ a cloud-based collaboration environment made available in the public domain [18]. The CAD model of the femur was assumed to be the femur of an adult, disease-free male. The length of the femur was 487 mm, and the height of the male with this femur was believed to be around 1740 mm [19]. Initially, it was a hollow femur model that represented the cortical bone. SolidWorks 2021 (Dassault Systems, Waltham, Massachusetts, USA) computer-aided design software was used for the further development of the model. The internal region was created by using the intersection feature in SolidWorks. As a result, the CAD models of the femur with dual-zone (cortical and cancellous) were obtained. For the placement of the interlock nail (diameter, 12 mm), the intramedullary canal was created by inserting a dummy nail (diameter, 13 mm) in the cancellous bone and cutting away the common region using the cavity option in SolidWorks.

The geometry of the interlocking femoral nail and LCDCP was obtained by reverse engineering at VNIT Nagpur, India. The implant samples were collected from a clinical laboratory. SolidWorks 2021 software was used to create the CAD model of the interlocking nail (length 420 mm, diameter 12 mm) consisting of two distal holes and two proximal holes. The CAD model of LCDCP (length 236.5 mm, thickness 6 mm, and width 17 mm) with 13 holes was also prepared in SolidWorks. The interlocking screws (diameter, 4.9 mm) used with the interlocking nail and the locking head screws (diameter, 5 mm) used with the LCDCP were also modeled. The length of the screws was taken according to the respective hole position capturing the non-uniform diameters of bone. The CAD model of interlocking femoral nails and LCDCP is shown in Fig. 2a and Fig. 2b.

Fig. 2.

CAD geometry models of implants – a 420-mm long interlock nail, b 13-hole limited contact dynamic compression plate (LCDCP)

The assemblies of CAD models of implants (interlock nail and LCDCP plate) with the CAD model of bone were created in SolidWorks 2021. During the assembly, the common intersecting volumes of bone screws were cut away from the models by using the cavity feature. The use of threaded screws was avoided in the assemblies to ease the simulation in ANSYS Workbench, and the contacts between the screws and the bone were set to be bonded.

Fracture Configuration

For a complete understanding of the effect of implantation on bone fracture healing, both the implants were assembled in rigid and flexible modes and simulated in ANSYS Workbench (ANSYS Inc., Canonsburg, PA, USA). The aim was to analyze the healing outcomes under the effect of varying fixation stability. Therefore, a 30-degree inclined, proximal lateral to distal medial (PLDM) fracture of minute size 0.1 mm was created in the middle of the shaft. This minute fracture was created only to separate the bone into two parts as the effect of different fracture sizes was not considered in this study. Both implants (interlock femoral nail and the LCDCP) were fixed rigidly in one case with all screws inserted and flexibly in another case with some screws not inserted. Two flexible fixation modes were considered for LCDCP and one flexible fixation mode was considered for interlock nail. Thus, there were 5 cases to investigate for the immediate post-surgery phase. The immediate post-surgery models of the femur fastened with two different implants (interlock nail and LCDCP) in different fixation configurations (rigid and flexible) are shown in Fig. 3.

Fig. 3.

Immediate post-surgery CAD models of the fractured femur treated with a interlock nail rigidly, b interlock nail flexibly, c LCDCP rigidly, d LCDCP flexibly 1, and e LCDCP flexibly

Material Properties and Meshing

The bone was assumed to be linearly elastic and isotropic. The material properties for cortical bone (E = 16.7 GPa, v = 0.3) and cancellous bone (E = 0.155 Gpa, v = 0.3) were provided based on the previous studies [20–25]. The material properties for medical grade titanium (Ti-6Al-4 V) implant (E = 110 Gpa, v = 0.3) were also taken from the past study [17].

The assembled CAD model was imported into ANSYS Workbench to prepare finite element models. It was concluded in the past study that the tetrahedral 10-noded elements are the best choice for the meshing of human femurs as they are best suited to the nonlinear geometry of the bone [26]. Therefore, tetrahedral 10-noded elements with three translational degrees of freedom at each node were employed to mesh the cortical bone, cancellous bone, interlock nail, and screws. The LCDCP plate with locking head screws was also meshed with tetrahedral 10-noded elements. The contact sizing was provided to bone-screw, bone-nail/plate, and screw-nail/plate contact regions with adjustable element sizes. This ensured that the load was transferred efficiently between the contacts.

The mesh convergence study was done to get the optimal element size (Table 1). It was observed that when the mesh size was changed from 5 to 4 mm for all the parts of the assembly, the change in deformation is negligible (0.00566%) as shown in Table 1 (row 3). To improve the mesh element quality, the element size for cortical bone and implants was further reduced up to 2 mm while keeping constant the element size for cancellous bone (4 mm). That seems to improve the mesh element quality of the finite element model and further reduce the percentage change in deformation to 0.00283% as shown in Table 1 (row 5). Thus, the element size of 4 mm was provided to the cancellous bone, and the 2 mm element size was given to cortical bone and implants.

Table 1.

Mesh convergence study done on bone-LCDCP fixation

Element size		No. of nodes	No. of elements	Deformation (mm)	% Change
Cancellous bone	Cortical bone/implants
6 mm	6 mm	624,123	399,443	7.0615
5 mm	5 mm	624,926	400,368	7.0621	0.00850
4 mm	4 mm	626,096	401,132	7.0625	0.00566
4 mm	3 mm	626,924	401,719	7.0628	0.00425
4 mm	2 mm	627,809	402,293	7.0629	0.00283

Contacts

The interlock nail (diameter 12 mm) was inserted in the intramedullary canal (diameter 13 mm) created in the cancellous bone (discussed in Sect Computer-aided design (CAD) models). The bone and the nail surfaces were considered to have nonlinear frictionless contact, with an initial gap permitted by creating the intramedullary canal of a slightly larger diameter [6]. That allowed the surfaces to come into contact during the simulation [27]. The contact between cortical and cancellous bones was set to be bonded. The use of threaded screws was avoided in the assemblies to simplify the simulation in ANSYS Workbench, and the contacts between the screws and the bone (cortical/cancellous) were assumed to be bonded [24].

The bonded contact was provided between the screws head and the LCDCP plate because the LCDCP and their screws are also having threads on their mating surfaces. The threads of the locking head screws were also not modeled to ease the simulation and screw-bone (cortical/cancellous) contacts were set to be bonded [24]. The frictional contact (frictional coefficient, 0.3) was provided to the LCDCP and cortical bone contact [28].

For all the bone-implant contact pairs, the implant surface (harder surface) was considered as the target surface, and the bone surface (softer surface than the implant) was considered the contact surface. It was to ensure that the bone surface cannot penetrate through the implant surface.

One-Legged Stance Loading and Boundary Conditions

In the current study, the one-legged instance loading configuration mentioned in past studies [29] was used to investigate the biomechanical behavior of bone-implant fixation in various aforementioned modes. The load setup consists of the joint reaction force, abductor force, vastus lateral force, and iliopsoas force. The magnitude and orientation of the muscle and joint reaction forces are given in Table 2. During the FE analysis, the distal end of the femur bone in the assemblies was set to be fully fixed to restrain any translation and rotation at this end (zero displacements) and forces were applied as shown in Fig. 4.

Table 2.

Muscle and joint reaction forces applied to the femur

Muscle and joint reaction forces	Magnitude of resultant force (N)	Orientation (degree)
		φ	ϒ
Joint reaction force (JRF)	730	291	273
Abductors (ABD)	300	110	90
Vastus lateralis (VL)	292	270	270
Iliopsoas (LP)	188	99	137

Fig. 4.

Application of muscle and joint reaction forces in ANSYS Workbench

Output Parameters to be Analyzed

Several influential mechanical factors involved in bone fracture healing were captured based on the published literature. Some factors are found to be beneficial for healing while some are known to be detrimental. The influential mechanical factors include compressive normal force [30], mean stress in the bone around the fracture [6], shear deformation or shear micromotion near the fracture site [5], the maximum stress in implants [31], axial deformation or axial micromotion near fracture site [32], and the contact pressure near fracture site [29]. The importance of all these parameters in fracture healing is discussed later in Sect Finite element analysis. The ‘near fracture site’ is a region or location that lies on the bone surface nearest to the fracture zone and which is indicative of the healing zone. All the ‘near fracture site’ parameters were measured by using the probe tool in Ansys Workbench.

Statistical Analysis

Considering both the implants with different flexibility modes led to a total of 3 data sets that cover all the possible combinations. The significance of the impact of modifying the fixation stability on the influencing output parameters in interlock nail and LCDCP fixations was analyzed by utilizing a paired-sample T test with a P value less than 0.05 considered significant.

Results

The various decisive parameters that assess implantation efficacy were investigated by FEA. Table 3 summarizes the results obtained by the finite element analysis. The maximum penetration in the nonlinear contacts among all the configurations was found to be 0.0104 mm.

Table 3.

Finite element analysis results for the immediate post-surgery phase

Output parameters	Fixation configuration
	N(R)	N(F)	P(R)	P(F1)	P(F2)
Normal stress (MPa)	138.2	209.3	87.5	76.2	219.4
Mean stress (MPa)	166.2	227	100.4	103.5	260
Shear micromotion (mm)	1.79	2.11	1.54	1.55	1.93
Axial micromotion (mm)	0.76	1.12	0.37	0.38	0.65
Contact pressure (MPa)	150.1	198.1	117.6	131.4	246.6
Mean stress in implant (MPa)	350.7	685	296.9	349.7	619.6

N(R) & N(F) represent interlock nail fixed rigidly and flexibly, respectively. P(R) represents LCDCP fixed rigidly. P(F1) and P(F2) represent LCDCP fixed flexibly with 8 screws and 6 screws, respectively

Compressive Normal Stress and Mean Stress

The compressive normal stress (P = 0.02) near the fracture site was raised by 71.1 MPa (51.4%) when the interlock nail was fastened flexibly rather than rigidly. When rigid and flexible fixation was compared in LCDCP implantation, the flexible fixation was found to reduce the compressive normal stress by 11.3 MPa (12.9%) when 8 screws were used but increased it massively by 131.9 MPa (150.7%) when 6 screws were used (Table 3, row 1). As discussed in Sect Output parameters to be analyzed, all the ‘near fracture site’ parameters were obtained by using the probe tool in the Ansys Workbench.

When the interlock nail was fastened flexibly rather than rigidly, the mean stress in the bone (P = 0.02) near the fracture site increased by 60.8 MPa (36.6%). When rigid and flexible fixation was compared in LCDCP implantation, the flexible fixation was shown to increase mean stress by 3.1 MPa (3.09%) and 159.6 MPa (159%), respectively, when 8 screws and 6 screws were used in fixation (Table 3, row 2).

Shear Micromotion and Axial Micromotion

The shear micromotion near the fracture site was always found to be higher in the anterior–posterior direction than in the medial–lateral direction as shown in Table 4. Therefore, the shear micromotion in the anterior–posterior direction was considered to examine its effect on fracture healing.

Table 4.

Shear micromotion (mm) in the anterior–posterior (A-P) and medial–lateral (M-L) direction of the femur obtained by FEA for the post-surgery phase

Direction of shear micromotion	Fixation configuration
	N(R)	N(F)	P(R)	P(F1)	P(F2)
Anterior–Posterior	1.61	1.81	1.54	1.55	1.93
(A-P)
Medial–Lateral	0.64	0.97	0.36	0.38	1.43
(M-L)

N(R) & N(F) represent interlock nail fixed rigidly and flexibly, respectively. P(R) represents LCDCP fixed rigidly. P(F1) and P(F2) represent LCDCP fixed flexibly with 8 screws and 6 screws, respectively

The shear micromotion (P = 0.005) and axial micromotion (P = 0.05) near the fracture site were increased by 0.2 mm (12.4%) and 0.36 mm (47.4%), respectively, when the interlocking nail was fastened flexibly rather than rigidly. There was very little difference in both output parameters when the LCDCP was fixed flexibly with 8 screws rather than rigidly. In comparison to rigid fixation, shear micromotion and axial micromotion increased by 0.39 mm (25.3%) and 0.28 mm (75.7%), respectively, when 6 screws were utilized with LCDCP (Table 3, row 3, and row 4).

Contact Pressure and Mean Stress in Implant

The contact pressure (P = 0.003) in the vicinity of the fracture increased by 48 MPa (32%) when the interlock nail was fastened flexibly rather than rigidly. When rigid and flexible fixation was compared in LCDCP implantation, the flexible fixation was shown to increase the contact pressure by 13.8 MPa (11.7%) and 129 MPa (109.7%), respectively, when 8 screws and 6 screws were used in fixation (Table 3, row 5).

The mean stress in the implant (interlock nail and screws) (P = 0.002) was found to be enhanced by 334.3 MPa (95.3%) when the interlock nail was fixed flexibly instead of rigidly. When rigid and flexible fixation in LCDCP implantation was examined, the flexible fixation was found to increase the mean stress in the implant (LCDCP and screws) by 52.8 MPa (17.8%) when 8 screws were used in fixation. When compared to rigid fixation, the flexibly fixed LCDCP with 6 screws caused a high rise of 322.7 MPa (108.7%) in mean stress in the implant (LCDCP and screws) (Table 3, row 6). However, the location of maximum stress in implants was not identical always as discussed later.

Clinical Observational Study

The clinical study was performed on patients implanted with both implants in various fixation modes to provide conclusive evidence for the superiority of implants in various fixation modes. A search turned up 283 patients with femoral shaft fractures, with 69 of them meeting our criteria for concerned implants (LCDCP and Interlock Nail), each with two fixation modes. Both implants were made up of medical-grade titanium alloy (Ti-6Al-4 V). All 69 patients were between the ages of 19 and 37. Patients were followed up on after 3 months and 9 months of surgery. All the clinical data were verified by an experienced orthopedic surgeon. Out of 69, the clinical observations of 63 patients (91.3%) were found to justify the results of the current finite element study as discussed later. The position of the screws in interlock nail fixations (rigid and flexible) was identical to that in the FE analysis. Although the position of screws inserted in the LCDCP differed from that used in the FE analysis for this study, the flexible fixation can be thought of as a fixation with fewer screws inserted.

Discussion

Bone fracture healing is a highly complicated mechano-biological phenomenon, and it requires an optimum environment at every stage of bone fracture healing [33]. This study employed a comprehensive finite element model to investigate and compare the performance of the intramedullary nail (interlocking femoral nail) and the compression plate (limited contact dynamic compression plate) which are used to treat femur shaft fracture. During FE analysis, the various important mechanical factors which play a vital role in the healing of fractures were considered. The clinical trials were also performed to check the reliability of the outcomes of this study. Also, to test the reliability of the FE analysis, the bone-implant fixations (Interlock nail and LCDCP) were analyzed experimentally and numerically using finite element analysis. The bone-implant fixation was rigidly fixed from the distal end of the bone in both analyses (experimental and FEA), and a vertical compressive load was applied at the top of the bone. The detailed process can be seen in the supplementary material (Supplementary file 1).

Finite Element Analysis

For the post-surgery phase, the comparison of all the influencing output parameters obtained from finite element analysis are shown in Fig. 5.

Fig. 5.

Comparison of output parameters. N(R) & N(F) represent interlock nail fixed rigidly & flexibly, respectively. P(R) represents LCDCP fixed rigidly. P(F1) and P(F2) represent LCDCP fixed flexibly with 8 screws and 6 screws, respectively. ‘NF’ represents ‘Near fracture location of implant.’ H1, H3, and H4 represent the implant’s first, third, and fourth proximal holes corresponding to the respective first, third, and fourth proximal screws (S1, S3, S4). Numbers have not been applied to holes that have not been used for screws put

Compressive Normal Stress and Mean Stress

Many in vivo investigations have found that increasing compressive normal load near the fracture site promotes callus development and hence improves fracture healing, whereas decreasing the load causes fracture healing to be sluggish [6]. The flexibly fixed interlock nail (209.3 MPa) and the flexibly fixed LCDCP with 6 screws (219.4 MPa) produced high values of compressive normal stress and thus greater compressive force in the bone near the fracture site. This indicates that healing was improved with these configurations for both implant materials. In comparison to interlock nail flexible fixation, the aforementioned LCDCP fixation has a slightly higher stress value (Fig. 5a).

Stress shielding is a process detrimental to bone health that results in the loss of bone density (osteopenia) and the loosening of screws due to the removal of stress from the bone by the stiffer implant [34]. The mean stress in the bone around the fracture indicates that stress shielding by the implant has a limited effect. Figure 5b shows that among all the configurations, the flexibly fixed LCDCP (Ti) with 6 screws was found to have the maximum value of mean stress in the bone around the fracture (260 MPa). For interlock nails, it was observed maximum (227 MPa) for flexible fixation (Ti). It shows that compromising stability produces a higher transfer of stress from implant to bone. The lowest mean value of stress was found in LCDCP fixed rigidly indicating the low transfer of stress from implant to the bone.

Shear Micromotion and Axial Micromotion

While allowing axial load at fracture level by compromising with the fracture stability, the shear deformation/shear micromotion also increases, which results in poor callus formation and hence delayed union even when a sufficient axial load is present [5]. Figure 5c shows that LCDCP flexible fixation with 6 screws had the highest value of shear micromotion among all five configurations. As a result, the same configuration (LCDCP’s 6 screws flexible fixation) that improved healing with a higher normal stress magnitude can also delay healing. Moreover, the maximum shear movement observed for interlock nail fixed flexibly (1.81 mm) and for LCDCP flexible fixation with 6 screws (1.93 mm) was found below the limit of shear movement (2–4 mm) being inimical to the healing process [35].

Another mechanical factor that aids fracture healing is axial micromotion near the fracture site [32], which was shown to be maximum (1.12 mm) for flexibly fixed interlock nail (Ti). Also, this was observed to be much greater than the maximum axial micromotion observed in LCDCP (Ti) fixation (0.65 mm) (Fig. 5d).

When the rigid and flexible fixation was compared in LCDCP implantation, the flexibly fixed LCDCP with 8 screws was observed to have a negligible effect on shear and axial micromotion (Fig. 5c, d). However, using 6 screws with LCDCP resulted in a significant rise in shear micromotion (25.3%) and axial micromotion (75.7%). It reveals that when 6 screws were utilized in LCDCP, the fixation stability was effectively changed. (Fig. 5c, d).

Contact Pressure and Mean Stress in Implant

The more the contact pressure at the fracture surface, the faster will be the healing process [29]. The contact surface must be adequately stressed for proper healing to occur. Figure 5e shows that for both implants, the flexible implantation resulted in more contact pressure near the fracture site. The maximum contact pressure was observed for flexibly fixed LCDCP with 6 screws (246.6 MPa) making it a beneficial configuration for fracture healing. It shows that the contact surface was adequately stressed. For interlock nails, the maximum contact pressure observed was 198.1 MPa for flexible fixation.

Among all the configurations, the flexibly fixed interlocking nail produced the highest mean stress in the implant (Fig. 5f). The location of maximum stress in the implant was at the proximal screw hole (Fig. 6b). This is due to a stress concentration at the proximal screw hole of the implant. Otherwise, the stress near the fracture was found to be around 400 MPa. (Fig. 6b), somewhat greater than the maximum stress induced in the implant for the rigid fixation of the nail (350.7 MPa) (Fig. 6a). Since one of the proximal screws was not inserted to make the fixation flexible, only one screw had to counter the bending moment raised due to greater axial micromotion (Fig. 5d). That was responsible for the rise in stress concentration at the implant’s screw hole. However, the location of maximum stress in the implant was found to be different among the configurations as shown in Fig. 6.

Fig. 6.

Mean stress induced in the implant for – a Interlock nail fixed rigidly, b interlock nail fixed flexibly, c LCDCP fixed rigidly, d LCDCP fixed flexibly with 8 screws, and e LCDCP fixed flexibly with 6 screws

Inference from the Finite Element Analysis

For the post-surgery phase, all of the output parameters were examined and a comparison was made. The flexibly fastened LCDCP with 6 screws had the greatest magnitude of beneficial normal compressive stress and mean stress in the bone around the fracture site (Fig. 5a, Fig. 5b). In addition, it had the highest contact pressure of any of the configurations (Fig. 5e). However, with flexibly fixed interlock nails, these parameters were also proven to be of the optimal magnitude. The aforementioned LCDCP fixation also showed the highest amount of detrimental shear micromotion around the fracture site (1.93 mm), which was only 6.6% higher than the shear micromotion seen with the flexibly fixed interlock nail (1.81 mm) (Fig. 5c). However, the magnitude of shear micromotion for both the configurations was found below the limit of shear micromotion (2–4 mm) being detrimental to the healing process [35]. The axial micromotion near the fracture site was found to be substantially higher (72.3%) for the flexibly fixed interlock nail than the axial micromotion seen for the flexibly fastened LCDCP with 6 screws (Fig. 5d). Apart from axial micromotion, in terms of other output parameters, the flexibly fixed LCDCP (6 screws) was found to be a better fixation than the flexibly fixed interlock nail. With a substantially higher magnitude of axial micromotion, the flexibly fixed interlock nail had the optimal magnitude of all output parameters. Thus, for the post-surgery phase, it can be concluded that flexible fixation of interlock nail and LCDCP with 6 screws provides a good mechanical environment for fracture healing.

Clinical Observations and Inference from the Clinical Study

It was evident that callus formation is beneficial for fracture healing. The more the callus forms, the faster the bone heals. From the past studies, it is noticeable that the less rigid fixation leads to faster healing and also boosts the quality of the callus formed [1, 9]. The reason behind the difference in the formation of the amount of callus is the micromotion between the fracture gap. Figure 7 shows immediate post-surgery and the follow-up clinical data of some patients meeting our criteria. LCDCP was implanted in some patients (P1, P2, P3, and P4), while interlock nail was implanted in others (P5, P6, P7, P8, and P9). The 3 months post-surgery clinical data of patients implanted with rigidly fixed LCDCP (P1 and P2) demonstrate that there was very little callus developed at the fracture site. In the 3 months post-surgery observation of another patient (P3) implanted with LCDCP in flexible mode or with fewer screws inserted, the callus formation was found to be greater and adequate. A 9 month of clinical observation of another patient (P4) implanted with the flexibly fixed LCDCP revealed that the fracture had healed completely and that the bone had reshaped to its original dimensions.

Fig. 7.

Clinical observations of the patients implanted with LCDCP and interlock nail in different fixation stabilities. P1, P2, P3 and P4 represent patient 1, patient 2, patient 3 and patient 4, respectively, implanted with LCDCP. P5, P6, P7, P8 and P9 represent patient 5, patient 6, patient 7, patient 8 and patient 9, respectively, implanted with interlock nail. ‘A’ represents immediate post-surgery observation. 'B and C' represent patient observations after 3 and 9 months of surgery, respectively

The 3 months post-surgery clinical data of patients implanted with rigidly fixed interlock nail, i.e., both proximal screws inserted (P5 and P6) depict that there was very little callus developed at the fracture site. The callus formation was observed to be substantial in the 3 months post-surgery observation of the patient implanted with a flexibly fixed interlock nail (P7). The 3 months post-surgery observation of another patient (P8) implanted with the same fixation showed a very good amount of dense callus accumulated around the fracture. A 9-month post-surgery clinical observation of another patient implanted with the same fixation (P9) revealed that the fracture had completely healed and that the bone had reshaped and appeared to be fresh healthy bone with no fracture.

Clinically, the flexible fixations for both the implants were proven to be beneficial in this study, as they enhanced the callus generation at the fracture site when compared with their rigid fixation. The 9 months post-surgery clinical observation of patient 4 and patient 9 implanted with LCDCP and interlock nail, respectively, revealed that both implants in flexible fixation mode were equally effective. This supports the findings of the current finite element analysis, which show that the LCDCP with fewer screws is also a very efficient and effective way of treating femur diaphyseal fractures.

Limitations and Conclusion

There are some shortcomings of this study that needs to be pointed out and should be considered before the direct application of the results. Firstly, the bone was assumed to be linear isotropic, but the real bone is viscoelastic, nonlinear, and anisotropic [36]. However, some past studies proved that the linear characteristics of bone are considered a fair approximation of real bone [25, 37]. Therefore, many studies in the past used the isotropic properties of the bone to simulate their FE model [28, 38–40]. The current model is only applicable to fresh healthy adult bone, and the effect of osteoporosis is not considered. Secondly, bone healing is a very complex process that may be affected by many factors, and most of the factors are biological. Apart from the mechanical parameters, biological factors like the blood supply and contact of soft tissues at the fracture region are the critical biological conditions for the healing process [41, 42]. These biological conditions cannot be simulated by using FEA. However, the effectiveness of this finite element study was validated by in vivo clinical observation. The current analysis includes the majority of the relevant mechanical parameters that can have a significant impact on the fracture healing phenomenon.

Despite its shortcomings, this finite element study will be highly valuable in simulating the fracture healing process and developing the best feasible treatment for the fracture using various implants. This research investigated the performance of nailing and plating with different fixation stabilities by finite element analysis and in vivo clinical observation. Both investigations revealed that the flexibly fastened LCDCP with fewer screws could be an alternative to flexibly fixed interlock nails for the femur diaphyseal fracture, as both fixations have been shown to generate an equally good mechanical environment during the healing phase.

Clinical Importance

The findings of this study could help bioengineers or surgeons better understand how nailing and plating affect fracture healing. It should be emphasized that a less stiff material implant that was flexibly attached to the fractured bone was sufficiently robust to keep the fracture straight and to allow for beneficial interfragmentary mobility, which is known to promote callus production at the fracture junction. This study is part of ongoing research aimed at developing the best possible fracture treatment method using IM nails and compression plates [43].

This study with the current FE model can be applied to the subject of age within 50 years because a past study on age-related changes in bone mineral density revealed that up to the age of 50, bone mineral density was stable, but after that, it began to decrease [44]. The men in their 50 s were observed to have a 4.0% prevalence of osteoporosis, 7.2% in their 60 s, 15.1% in their 70 s, and 26.7% in their 80 s. Women's figures were 15.2%, 36.5%, 62.7%, and 85.8% [44].

Future Scope

Cortical and trabecular bone are the materials used in the bone analysis. The elastic properties of both types of bone vary with age and species, and thus, there is variation across all subjects. This inter-subject variability, combined with the anisotropic nature of bone, makes numerical simulation extremely complicated. Despite this, the majority of studies have attempted to describe bone with varying degrees of success by modeling bone properties as transversely isotropic or fully isotropic and assuming a homogeneous structure [43]. Osteoporosis study of bone is a subject of bone disease that can be studied separately. Therefore, a similar study can be performed in future considering the osteoporotic nature of bone. Also, in light of the more advanced locking compression plate (LCP), a similar study comparing the biomechanical performance of the LCP, LCDCP, and intramedullary nail with varying fixation stability can be performed on healthy and osteoporotic bone.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declarations

Conflict of interest

Sandeep Rathor, Rashmi Uddanwadiker, Nandram Saryam, and Ashutosh Apte declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human or animal subjects performed by the any of the authors.

Informed consent

For this type of study informed consent is not required.