Optimal additional support screw position for prevention of hinge fracture in biplanar closed wedge distal femoral osteotomy

Abstract

Background

The purpose of this study was to examine the biomechanical significance of supplemental fixation using a positional screw in prevention of the hinge fracture in lateral closed-wedge distal femoral osteotomy (LCW-DFO) by means of a three-dimensional finite element analysis.

Methods

The three-dimensional numerical knee models with LCW-DFO were developed. To assess the mechanical efficacy of the positional screw and determine its optimal position and orientation, in total, 13 screwing methods were analyzed. In the first four methods, the screw was supported by the cortical bone only on the medial surface (mono-cortical). In the other 9 models, the screw was supported by both medial and lateral cortical bones (bi-cortical). Under 1000 N of vertical force and 5 Nm of rotational torques, the highest shear stress value around the medial hinge area was adopted as an analytical parameter.

Results

In mono-cortical methods, with the cancellous bone support, all methods were able to reduce the highest stress value compared to the value without the screw, while the efficacy was rather inferior when the screw was in horizontal direction. Without the cancellous bone support, however, all methods were not able to reduce the stress value. In bi-cortical methods, with the cancellous bone support, almost all screw augmentation methods were able to reduce the stress value. When screwing from the medial to the lateral, it only gets worse when going extremely posterior. Without the cancellous bone support, all methods were able to reduce the stress value.

Conclusion

The mechanical efficacy of the bi-cortical method was proven regardless of the quality of the local cancellous bone.

1. Introduction

Osteotomy around the knee is a commonly employed surgical option for relatively young and active patients with uni-compartmental knee osteoarthritis, and favorable outcomes have been reported in literatures.1, 2, 3 Among the various types of osteotomy procedures, double level osteotomy (DLO) has been our principal surgical option for severe varus knee deformity correcting both femoral and tibial deformities while reconstructing physiologic joint geometry.4, 5, 6, 7, 8 DLO for varus knee consists of lateral closed wedge distal femoral osteotomy (LCW-DFO) and high tibial osteotomy (HTO). On the other hand, medial closed wedge distal femoral osteotomy (MCW-DFO) has been generally indicated for correction of valgus knee deformity in our practice because distal femoral valgus is a primary component of deformity in the majority of the cases.9, 10, 11

In recent years, surgical technique of DFO has been evolved with introduction of biplane-cut osteotomy and fixation following the minimally invasive plate osteosynthesis (MIPO) technique.¹⁰ The wide bony contact area attained by the biplanar osteotomy and rigid fixation using a locking compression plate (LCP) can induce enhancement of bony healing and improvement in fixation properties.^11,12 However, there still are remaining shortcomings and complications related to its surgical procedure. Among those, hinge fracture extending from the medial end of the wedge is a rather frequently encountered complication with the incidence ranging from 30.6% to 57%.13, 14, 15

Hinge fracture in wedge osteotomy was first documented as a complication in open wedge HTO and there have been several studies dealing with this issue,^16,17 while there have been only a few reports regarding this complication in closed wedge DFO.^13,14 Due to inferior stability of closed wedge DFO compared with open wedge HTO even with the fixation using an angular locking plate,¹² consequences of hinge fractures in closed wedge DFO can be more evident potentially leading to delayed bony healing or change in alignment.^16,17 In order to prevent this problematic complication, supplemental fixation with an additional plate on the contralateral side is generally recommended. When hinge fracture occurs in MCW-DFO, lateral additional plate fixation is needed, which may induce irritation of the iliotibial band. Consequently, hinge fracture in close wedge DFO adds to surgical invasion and may impair the surgical outcome, and thus its prevention is a critically important issue. In our current clinical practice, supplemental fixation of the osteotomy site using a positional screw has been adopted as a measure to reduce the incidence of hinge fracture without increasing surgical invasion; however, its preventive efficacy has not been proven.

The purpose of this study, therefore, was to examine the biomechanical significance of the positional screw in prevention of the hinge fracture in closed wedge DFO by means of a three-dimensional finite element (FE) analysis. It was hypothesized that one positional screw fixation would effectively reduce the stress at the hinge under loading conditions, and its efficacy would be influenced by the screw position.

2. Materials and methods

2.1. Simulation of LCW-DFO with or without the positional screw

A standardized three-dimensional (3D) commercially available digital model of a left femur (#3972, Pacific Research Laboratories, Inc., WA, USA) was used to simulate LCW-DFO. The bone geometry was derived from a healthy male adult without any osteoarthritis or deformities. An orthogonal coordinate system, which was based on the mechanical axis, was used to describe three distinct planes: coronal, sagittal, and axial.^18,19 LCW-DFO procedure on the femur model followed the surgical method developed by a working group from the Netherlands.¹⁰ Osteotomies in the biplanar technique are composed of three cuts: one ascending cut and two transverse cuts. The three cuts are made from the lateral side of the femur and intersect at the medial epicondyle. To simulate the lateral closed wedge (LCW), a wedge-shaped bone between the two transverse cuts was removed, then the distal bone was rotated in valgus direction on the intersection point, which acts like a hinge. As a result, in the analytical model, mechanical lateral distal femoral angle (mLDFA) was set at 85°.⁶ The initial osteotomy cut level in the LCW-DFO was set at 4 cm proximally from the lateral femoral epicondyle. After the femoral osteotomy using the biplanar technique was completed, fixation of the osteotomy site was accomplished with the TomoFix medial distal femur (MDF) anatomical plate (DePuy Synthes, Solothurn, Switzerland) and eight locking screws.

To assess the mechanical efficacy of the positional screw and determine its optimal position and orientation, in total, 13 screwing methods were analyzed. In the first four methods, the screw was supported by the cortical bone only on the medial surface (mono-cortical). Four angles were analyzed in the mono-cortical method as shown in Fig. 1. The screw was vertical in the method 0 and horizontal in the method 3, while the screws in the methods 1 and 2 were angled in between those two directions. These four models have the same entry point on the medial cortical bone surface. It simulates that the screw was inserted from the same position, but changed its direction for each model. The position of the entry point was set at approximately 2 cm proximally from the hinge area to prevent the screw hole causing any fractures around this area. In the other 9 models, the screw was supported by both medial and lateral cortical bones (bi-cortical). The entry and exit sites on the both bone surfaces are shown in Fig. 2. Two locations on the lateral bone surface and five locations on the medial bone surface were defined. Thus, ten (two times five) screw positions were defined. However, the A-5 positioning is considered not feasible due to the interference between the plate/locking screws and the positional screw. Thus, the A-5 position was excluded from the analysis and remaining 9 bi-cortical screwing methods were modeled and analyzed. Consequently, a total of 13 FE models with different screw positions were subjected to the analysis, and the results were compared to those of the control condition without positional screw. Hence, actually 14 models were analyzed in this study.

Fig. 1.

Four types of screw angles of mono-cortical method were analyzed. The insertion position on the medial bone surface were identical through the four models.

Fig. 2.

In bi-cortical method, nine types of screw positions were analyzed. Two locations on the lateral bone surface and five locations on the medial bone surface were marked as the screw holes. Although a product of two and five is ten, A-5 was impossible due to the plate and its screws. Thus, nine types were analyzed.

2.2. Analytical conditions

The 14 3D models, which included those with and without the positional screw, were converted to finite element (FE) models using the FE analysis program Nastran (Autodesk Nastran version 2021, Autodesk, San Rafael, CA, USA). Average of 4.2 mm tetrahedral second-order elements were used. To preserve continuity between the proximal and distal bony segments, two transverse cuts were connected from the intersection point of the three cuts to 1 cm inside, simulating two transverse cuts that were adjusted to meet at 1 cm inside from the medial bony surface. The rest of the two transverse cuts were separated. The locking screws and the positional screw were completely fixed to the bone and the locking screws were also completely fixed to the anatomical plate, while interfaces between the anatomical plate and the bone and osteotomy surfaces were defined as contact areas without friction. Because the coefficient of frictions of those contact areas were not considered in the analysis, the analytical model represented the worst-case situation.

All the material properties assigned to the elements were assumed to be isotropic and homogeneous. For the anatomical plate and screws made of titanium alloy, an elastic modulus of 193 GPa and a Poisson's ratio of 0.3 were adopted in the analysis. Elastic moduli of the cortical bone and the cancellous bone were set at 11.5 GPa and 0.6 GPa with their Poisson's ratios of 0.3 and 0.2 respectively.²⁰ Since the material properties of the cancellous bone are widely varied depending on aging and/or associated pathological conditions such as osteoporosis or diabetes,^21,22 models without cancellous bone component were also analyzed as a model to simulate a condition in patients who lack substantial mechanical support provided by the cancellous bone due to aging or severe osteoporosis.

As loading status applied to the model, two types of boundary conditions, 5 Nm of internal rotation (IR) with 1000 N of vertical force and 5 Nm of external rotation (ER) with 1000 N of vertical force, were used^12,23 (Fig. 3). The distal end of the femur was completely constrained. The loadings, which are axial loading with torque, occur during flexion-extension of the femur, during walking with full weight bearing.¹² IR and ER are defined depending on a rotational direction of the distal segment of the femur relative to the proximal part. As for a measure to estimate the effectiveness of the positional screw in reducing the risk of hinge fracture, the highest shear stress value around the medial hinge area under loading, which may induce the hinge fracture, was adopted as an analytical parameter.

Fig. 3.

Two types of boundary conditions: 5 Nm of internal rotation (IR) with 1000 N of vertical force (left figure) and 5 Nm of external rotation (ER) with 1000 N of vertical force (right figure) were used. The distal end of the femur was completely constrained.

3. Results

The principal shearing stress distribution around the hinge under the loading of 1000 N of vertical force and 5 Nm of internal rotation with cancellous bone without the screw is shown (Fig. 4). Two stress concentrations at the edge of the hinge were observed. Fig. 5 shows the highest shear stress values calculated for the mono-cortical screw methods. Stress values under IR (blue) and ER (orange) loading conditions with cancellous bone (left) and without cancellous bone (right) are shown respectively. The two horizontal lines in the graphs indicate the stress values without the screw. Without the screw, the highest stress value was 6 MPa with the cancellous bone (the blue line in the left graph), while the corresponding value without the cancellous bone was 8 MPa (the orange line in the right graph). With the cancellous bone, all methods were able to reduce the highest stress value with the lowest value observed in the method 2, while the efficacy was rather inferior in the method 3 (horizontal direction). Without the cancellous bone, however, all methods were not able to reduce the stress value.

Fig. 4.

The principal shearing stress distribution around the medial bone contact area (expressed by red circles) was measured. Since there were always two stress concentrations, only the higher value was extracted.

Fig. 5.

The highest shear stress values with mono-cortical screw method were extracted. The two vertical lines were the stress values without the screw. Stress values at IR (blue) and ER (orange) with cancellous bone (left) and without cancellous bone (right) are shown respectively.

Fig. 6 shows the highest shear stress values calculated for bi-cortical screw methods. The horizontal lines indicating the stress values without the screw are the same as in Fig. 5. With the cancellous bone, all screw augmentation methods but A-1 were able to reduce the stress value compared to the value without the screw. The lowest values were observed in the method A-3 with 4.9 MPa for IR and 3.3 MPa for ER. Without the cancellous bone, all methods were able to reduce the stress value and the lowest value was observed in the method B-2 with 6.7 MPa for IR and 7.0 MPa for ER.

Fig. 6.

The highest shear stress values with bi-cortical screw method were extracted. The two vertical lines were the stress values without the screw. Stress values at IR (blue) and ER (orange) with cancellous bone (left) and without cancellous bone (right) are shown respectively.

4. Discussions

Considering the purpose of the current study, the present study has shown that the positional screw is overall useful to reduce the stress at the hinge area potentially preventing the hinge fracture. The most important finding is that the use of the positional screw in both mono-cortical and bi-cortical methods could provide effective mechanical support in DFO. Comparing with the additional plate application on the contralateral side, augmentation with a single positional screw can afford substantial benefit for patients and surgeons because of its technical ease and less-invasiveness.

The mono-cortical method takes precedence over the bi-cortical method in terms of invasion; however, close attention should be paid to the host bone quality from a mechanical point of view. By contrast, mechanical support can be attained regardless of the cancellous bone quality in the bi-cortical method. Therefore, in cases with poor cancellous bone quality, use of the bi-cortical method seems preferable. As for the angles of the mono-cortical screw, the study results showed that the mechanical efficacy of horizontally directed screw was rather inferior among the four directions and may not be recommended. In patients with poor cancellous bone quality, however, no recognizable benefit was shown for any of the mono-cortical screwing methods.

In the analyses for the bi-cortical screw fixation in this study, mechanical efficacy in reducing the stress at the hinge area was shown for all the screwing positions tested. From a mechanical point of view, the bi-cortical method is overall more preferable than the mono-cortical method; however, from a practical standpoint, the A-1 positioning is not recommended because the screw trajectory extends backwards posing a risk for neurovascular injuries while the B-5 positioning may be too anterior. Potential problems of the bi-cortical method as compared with the mono-cortical method are greater surgical invasion to the host bone, and increased possibility of impingement against the anatomical plate or the locking screw. Direction of the screw should be carefully selected during the procedure so as to avoid the mechanical interference.

The material properties of the bone have been controversial and investigated in a lot of studies.^21,22,24 Its mechanical characteristics are anisotropic, inhomogeneous, viscoelastic, nonlinear and/or heterogeneous.²¹ In consideration of those complexities, only two obvious patterns of the bone conditions were applied in this study to be concise and technically workable assuring the results to be reliable. The first model combined the cancellous bone and the cortical bone. This model is assumed as a patient who is young and/or healthy without any osteoporosis. The elastic modulus of the cancellous bone in the model was set to be 600 MPa.²⁰ The second model was a patient whose cancellous bone doesn't have enough mechanical support due to aging or osteoporosis. To simulate such a patient, we adopted a model without the cancellous bone. To determine the validity of those two models, we conducted another analysis in which the mechanical property of the cancellous bone was varied. Fig. 7 shows the highest shear stress values around the medial hinge area with varied mechanical properties of the cancellous bone. One of the screwing conditions in the bi-cortical methods (A-3) was used here. With stepwise increase in the mechanical properties of the cancellous bone, the stress values under IR condition decrease and look saturated thereafter, while the stress values markedly increase corresponding to the reduction in the mechanical properties especially under ER loading. In the ER condition, the distal part of the femur externally rotates while fixed only at both edges: the anatomical plate on the lateral side and the bone connected hinge area on the medial side. Thus, the stress value around the hinge area markedly increases with decrease in the mechanical properties of the cancellous bone. Whereas, under IR application, the ascending cut surfaces are compressed with each other taking shear stress from the transverse cuts, even if the cancellous bone support was lost. Based on this extra analysis, the two distinct examples were selected for the analysis as two representative types, since the real cases would be somewhere between the two examples. Even if the elastic modulus of the cancellous bone is higher than 600 MPa, it is assumed that the results obtained from the present study analysis can be extrapolated to the clinical situation.

Fig. 7.

The highest shear stress values with one of the bi-cortical screw methods (A3) were extracted varying the elastic modulus of the cancellous bone.

There are several limitations included in the study. First, although the efficacy of the mono-cortical method is affected by the quality of the cancellous bone at the osteotomy area, its influence may not be accurately predicted. If the cancellous bone has enough stiffness, the mono-cortical method can be useful to reduce the stress around the hinge area. Second, the geometrical bone data in this study were derived from a single person in the digital anatomy media, and thus the present study results cannot be generalized and applied to all patients with osteoarthritis undergoing osteotomy. Third, residual stress was not considered. In this study, the hinge area was modeled without any residual stress before loading. During surgery, the hinge area was not a point, but an area, so some residual stress was likely present. And finally, it should be noted that this study was limited to LCW-DFO with the biplanar technique.

5. Conclusions

Use of additional positional screw fixation as a supplemental fixation in close wedge DFO could reduce the stress around the hinge area in both the mono-cortical and bi-cortical methods, potentially reducing the risk for hinge fracture; however, the efficacy of the mono-cortical method was not demonstrated in cases without support of the cancellous bone. The mechanical efficacy of the bi-cortical method was proven regardless of the quality of the local cancellous bone, while problem of interference with the plate/locking screw may be more frequently encountered in the bi-cortical screwing method. From a practical standpoint, the direction of the bi-cortical screw should be carefully selected.

This simulation study helps decision-making for surgeons who try to use the positional screw to avoid hinge fractures. In the future, we have the same simulations using various bone geometries to make the knowledge robust.

Funding

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

Authors contribution

MH: Software, Writing-Original Draft; HN: Formal analysis, Conceptualization, Writing-Original Draft; RK: Methodology, Writing-Review and Editing; SO: Methodology, Writing-Review and Editing; SY: Project administration, Writing-Review and Editing; TT: Supervision, Writing-Review and Editing; TI: Data Curation, Methodology, Writing-Review and Editing.

Declaration of competing interest

None.