Application of a combined cancellous lag screw enhances the stability of locking plate fixation of osteoporotic lateral tibial plateau fracture by providing interfragmentary compression force

Background Insufficient interfragmentary compression force (IFCF) frequently leads to unstable fixation of osteoporotic lateral tibial plateau fractures (OLTPFs). A combined cancellous lag screw (CCLS) enhances IFCF; however, its effect on OLTPF fixation stability remains unclear. Therefore, we investigated the effect of CCLS on OLTPF stability using locking plate fixation (LPF). Materials and methods Twelve synthetic osteoporotic tibial bones were used to simulate OLTPFs, which were fixed using LPF, LPF-AO cancellous lag screws (LPF-AOCLS), and LPF-CCLS. Subsequently, 10,000 cyclic loadings from 30 to 400 N were performed. The initial axial stiffness (IAS), maximal axial micromotion of the lateral fragment (MAM-LF) measured every 1000 cycles, and failure load after 10,000 cycles were tested. The same three fixations for OLTPF were simulated using finite element analysis (FEA). IFCFs of 0, 225, and 300 N were applied to the LPF, LPF-AOCLS, and LPF-CCLS, respectively, with a 1000-N axial compressive force. The MAM-LF, peak von Mises stress (VMS), peak equivalent elastic strain of the lateral fragment (EES-LF), and nodes of EES-LF > 2% (considered bone destruction) were calculated. Results Biomechanical tests revealed the LPF-AOCLS and LPF-CCLS groups to be superior to the LPF group in terms of the IAS, MAM-LF, and failure load (all p < 0.05). FEA revealed that the MAM-LF, peak VMS, peak EES-LF, and nodes with EES-LF > 2% in the LPF were higher than those in the LPF-AOCLS and LPF-CCLS. Conclusion IFCF was shown to enhance the stability of OLTPFs using LPF. Considering overscrewing, CCLS is preferably recommended, although there were no significant differences between CCLS and AOCLS.


Background
Osteoporotic tibial plateau fracture (TPF) is prevalent among elderly populations [1].Due to specific knee geometry and tibiofemoral joint force, over 60% of osteoporotic TPFs occur in the lateral column [2].The prevalence of osteoporotic lateral TPF (OLTPF) gradually increases with age [3], and this injury can have a catastrophic effect on the health of the elderly.
Stable fixation for OLTPF is critical for accelerating recovery in elderly patients and avoiding complications related to long-term bed rest [4,5].Among the various available fixation methods, locking plate (LP) fixation (LPF) is the most widely used for OLTPF [6,7].However, 11% of patients with LP-fixed lateral TPF experience lateral platform collapse [8,9].Gardner et al. found that locking screw-cutting in the cancellous epiphyseal area was an important contributing factor, possibly due to the increasing shear stresses at the locking screw-bone interface [10].Therefore, additional lag screws were applied to increase the interfragmentary compression force (IFCF), thereby reducing the shear stresses at the locking screw-bone interface and ultimately enhancing the stability of the LPF of the lateral TPF [11,12].However, due to severe bone mass reduction in the osteoporotic tibia, the commonly used cancellous lag screws are prone to overscrewing, resulting in a decrease in IFCF and a high risk of lateral platform collapse [13,14].
The combined cancellous lag screw (CCLS) previously developed by our research team may offer promise for solving this issue [13].The major improvement in this device is that the screw rod of the CCLS is divided into two parts and connected by fine threads.This allows the screwing angle range to be expanded through fine threads, enabling surgeons to accurately determine the time to stop screw insertion to obtain a greater IFCF and avoid overscrewing.Moreover, this device facilitates the avoidance of further cutting damage to the osteoporotic cancellous bone by screw threads, as it fastens fragments by shortening the screw length through fine threads.Based on these characteristics, CCLS obtained a 25% higher IFCF than AO cancellous lag screws (AOCLS) in osteoporotic bones [15].However, no studies have investigated the effect of CCLS on the stability of OLTPF using LPF.
Consequently, this study aimed to investigate the effect of CCLS on the stability of OLTPF using biomechanical testing and finite element analysis (FEA).We hypothesized that the CCLS would effectively enhance the stability of the LPF of OLTPF by providing IFCF.

Biomechanical testing Materials
Twelve large fourth-generation osteoporotic synthetic left tibial bones (No. #3402 Sawbones, 0.16 g/cc, Pacific Research Laboratories, Vashon, WA, USA) were used in this study [16].The synthetic tibial bones were transversely truncated 200 mm from the lateral plateau and fixed using a dental tray powder (polymethyl methacrylate, Shanghai New Century Dental Materials Co., Ltd, Shanghai, China).

Fracture models and test groups
A reproducible cut was mechanically created by the same surgeon who used a thin blade saw based on a single template to create a lateral tibial plateau fracture (Schatzker type I).Following anatomical reduction under direct vision, the synthetic tibial bones were randomly instrumented into three groups of four samples each: length was reduced using a custom-designed locked screwdriver to obtain the maximum IFCF.Finally, the LP was implanted with seven locking screws, and all locking screws were tightened with a torque of 4 Nm.

Test procedure
All samples were subjected to compression loading to replicate the shearing forces on the tibial plateau during complete knee extension using a specially designed loading applicator.A hard gasket was attached to the upper surface of the lateral fragment, and a four-camera marker-based motion capture system (120 Hz, Qualysys AB, Gothenburg, Sweden) was used to obtain interfragmentary displacements.The mechanical properties of the samples were measured using a BOSE3510-AT testing machine (load measuring range, ± 7500 N, displacement measuring range, ± 25 mm, frequency range, 0.00001-100 Hz, Bose Corporation, Force Systems Group, Eden Prairie, MN, USA; Fig. 2).
After the models were created, the bones were subjected to 10,000 cyclic loadings with forces ranging from 30 to 400 N. The waveform of the cyclic loading was sinusoidal, with a frequency of 3 Hz.Fixation failure was defined as synthetic bone fracture, implant fracture, or disengagement of the bone-implant relationship.If the samples did not exhibit failure within 10,000 cycles, a load-to-failure test was performed at a loading speed of 5 mm/min until a displacement of 3 mm could be achieved.
The initial axial stiffness (IAS), maximal axial micromotion of the lateral fragment (MAM-LF) during cyclic loading, failure loads, and number of failure cycles (for structures that failed within 10,000 cycles) were assessed.IAS was defined as the force-displacement ratio at the third loading [17,18].The load that caused a 2-mm displacement was identified as the failure load.

Experimental models
Experimental models, including the AOCLS, CCLS, and LP systems, were constructed based on the specifications provided by their manufacturers.The left proximal tibia model was generated using Mimics v19.0 software (Materialize Mimics, Leuven, Belgium) to create a threedimensional reconstruction of computed tomography scan data obtained from a 27-year-old healthy volunteer (male; height, 174 cm; weight, 70 kg), from whom informed consent was obtained prior to data collection.Standard radiographs were performed to exclude lowerextremity fractures, abnormalities, and pathologic bone lesions.The scans were performed on a Philips Ingenuity 64 CT scanner.The scan range was from the anterior superior iliac spine to the ankle.The scan parameters were as follows: 140 kV; 350 mAs; slice thickness: 1 mm; scanning interval: 0.5 mm.The Digital Imaging and Communications in Medicine data of 1000 layers were copied and recorded.Ethical approval was granted by the Institutional Ethics Committee of our institution (Ethical Clearance Certificate No. 2022-01).Subsequently, the model was imported into SolidWorks (version 2017; Dassault Systèmes, Waltham, MA, USA) to create the lateral TPF (Schatzker I) and to virtually implant all devices into the fractured proximal tibia to simulate the three fixations: LPF, LPF-AOCLS, and LPF-CCLS (Fig. 3a, b).An appropriate position was maintained between the LP and tibia during the modeling to guarantee that all models maintained the precise relative positions of the LP and tibia in order to ensure accurate calculations.Once the relative positions of the LP and tibia were established, they were replicated and employed as the foundational model for subsequent models [19].

Boundary and loading conditions
According to clinical practice and previous studies [19,21], the contact between the LP and screws was fully bound to imitate the locking mechanism.The screws were fully tied to the tibia, and a friction coefficient of 0.3 was used for the fragment-screws and fragment-tibia interactions.The degrees of freedom on the surface of the distal tibia were fully constrained.The distal ends of the tibial models were fully constrained as the boundary condition.IFCFs of 0, 225, and 300 N were then applied to the LPF, LPF-AOCLS, and LPF-CCLS, respectively [15].An axial compressive force of 1000 N was applied to simulate the walking load in an adult patient.Of these, 60% of the selected force was attributed to the medial tibial plateau and 40% to the lateral tibial plateau (Fig. 4).

Convergence analysis, model validation, and analysis
Convergence analysis of the meshes was performed to determine the appropriate mesh quality (convergence change rate < 2%) [24].The MAM-LF, peak von Mises stress (VMS), and peak equivalent elastic strain of the lateral fragment (EES-LF) were then assessed.The MAM-LF was evaluated to validate the finite element (FE) model, which was compared using biomechanical tests.MAM-LF > 2 mm and EES-LF > 2% indicated failure displacement and bone destruction, respectively [25].

Statistical analysis
Statistical analyses were performed using IBM SPSS (version 27.0;IBM Corp., Armonk, NY, USA).The Shapiro-Wilk normality test was performed to check for data normality.An analysis of variance was used to compare differences among the three groups, and Fisher's Least Significant Difference test was used as a post hoc test.
No fixation failure was observed in any samples during the loading cycles.The MAM-LF of the LPF group showed statistically significant differences compared to the LPF-AOCLS and LPF-CCLS groups when measured after every 1000 cycles (p < 0.05; Fig. 5b).

Convergence analysis and FE model validation
Mesh sizes of 2.0 mm for bones and 0.5 mm for implants were applied in this study according to the results of the mesh convergence analysis (Table 2).The MAM-LF in the FE model was 0.596 mm, similar to the results of the biomechanical test (0.5446 ± 0.1092 mm).The FEA model herein was, therefore, deemed valid.

MAM-LF
No failed displacements of the lateral fragments were observed in any model.The MAM-LFs were 0.5909, 0.4255, and 0.4192 mm for the LPF, LPF-AOCLS, and LPF-CCLS models, respectively (Fig. 6).

Peak VMS of implants
The peak VMS of the implants in the three fixation models occurred at the bending of the locking plate.In the LPF model, the peak VMS of the implant (246.17mPa) was higher than those in the LPF-AOCLS (232.47 mPa) and LPF-CCLS (231.85 mPa) models (Fig. 7).

EES-LF
In the three fixation models, the peak EES-LF was observed at the posterior screw tunnel close to the fracture plane.The peak EES-LF and nodes with EES-LF > 2% in the LPF model (6.93%, 4660) were higher than those in the LPF-AOCLS (3.97%, 656) and LPF-CCLS (3.74%, 649) models (Fig. 8).

Discussion
This study investigated the effect of a CCLS on OLTPF stability using an LPF through biomechanical testing and FEA.Biomechanical tests showed that the stability of the OLTPF was enhanced through the IFCF provided by the lag screw.Furthermore, the FEA showed that adding a lag screw reduced the peak VMS of the LP and EES-LF.
Previous studies have shown that the IFCF produced by the lag screws improves the stability of the distal femoral fractures in non-osteoporotic bones [26][27][28].However, the role of IFCF in OLTPFs remains unclear.In our study, the effect of IFCF on OLTPF was investigated using biomechanical testing and FEA.The results of the biomechanical tests showed that the LPF-AOCLS and LPF-CCLS groups exhibited significantly higher IAS than that of the LPF group.This was similar to the results of the study by Plecko et al. [28] and is significant for patients to bear weight early, restore knee function, and avoid complications related to longterm bed rest.Additionally, we performed a fatigue test consisting of 10,000 loading cycles from 30 to 400 N to simulate a 70-kg adult walking during the 6-week fracture-healing process [29].No failure in which the MAM-LF was greater than 2 mm was observed in any samples during the loading cycles.However, there were significant differences in MAM-LF in the LPF group compared with those in the LPF-AOCLS and LPF-CCLS groups, indicating that the stability of LPF-AOCLS and LPF-CCLS was superior to that of LPF during the fracture-healing process.In addition, the load-to-failure test demonstrated a significant increase in the failure load in the LPF-AOCLS and LPF-CCLS groups compared to the LPF group.This finding was aligned with existing research on non-osteoporotic bones [19] and supported the hypothesis that the IFCF provided by the lag screw plays a crucial role in enhancing the stability of the LPF in OLTPF.
FEA showed findings similar to those of the biomechanical tests.The results of the MAM-LF in the LPF model were greater than those in the LPF-AOCLS and LPF-CCLS models.This was also reflected in the VMS results.The LPF model exhibited a higher peak VMS for the LP compared to the LPF-AOCLS and LPF-CCLS models.Thus, the IFCF provided by the lag screw was shown to effectively reduce the MAM-LF and load on the LP to protect it from damage.This result agrees with the study by Zhang et al. [19], in which FEA of distal femoral fractures was performed.Moreover, the LPF model exhibited a higher peak EES-LF and a greater number of nodes with EES-LF > 2% compared to the LPF-AOCLS and LPF-CCLS models.This finding implies that the lateral fragment suffered more severe bone damage when fixed using the LPF, suggesting that the IFCF provided by the lag screw could decrease the cutting effect caused by locking screws However, the present analysis revealed no significant differences between the LPF-CCLS and LPF-AOCLS groups in the IAS, fatigue test, or load-to-failure test, although the LPF-CCLS group exhibited numerically superior performance compared to the LPF-AOCLS group.There are two possible explanations for this phenomenon.First, although patients with osteoporosis have a low bone density and the trabecular bone structure is too weak to maintain the holding force of the lag screw [30], both AOCLS and CCLS have sufficient holding force in synthetic tibial bones; hence, it was difficult for the AOCLS and CCLS to produce significant differences in the holding force.Second, the limited sample size of the study and slight disparity in IFCF provided by the CCLS and AOCLS may have resulted in a lack of significant differences in the stability of LPF of OLTPF.Based on the biomechanical testing and FEA findings, both CCLS and AOCLS may enhance the stability of the LPF of OLTPF by providing an IFCF.

Limitations
This study has some limitations.First, only synthetic tibia bones were used, and no experiments were conducted on actual bones.Although this model may not accurately reproduce clinical osteoporotic fracture fixation, it allows for highly reproducible testing using homogeneous material.Second, the force exerted by body weight on the tibia was subject to variation based on the degree of flexion and extension in real life.Accurately replicating these intricate dynamics in the laboratory setting has proven challenging.However, according to published methods, the single test setting used in this study was sufficient to examine fixation stability [31].

Conclusion
Overall, this analysis showed that both CCLS and AOCLS effectively enhance the stability of OLTPFs using LPF by providing IFCF.Although no significant differences were observed between the CCLS and AOCLS groups, considering the risk of overscrewing in osteoporotic bones, CCLS is recommended for improving the stability of LPF in patients with OLTPF.

Fig. 1
Fig. 1 Locking plate system for biomechanical testing and modeling in our study.a Locking plate system; b Locking screw; c Head of the locking screw

Fig. 2
Fig. 2 Preparation of samples and accessories.a Views of constructs mounted for testing; b Two different lag screws for osteoporotic lateral tibial plateau fracture (OLTPF) reduction; c) A hard gasket was matched between the upper surface of the lateral fragment and the loading applicator

Fig. 5
Fig.5 Results of biomechanical testing.a Initial axial stiffness (IAS): The locking plate fixation (LPF) group showed significant differences compared with the LPF-AO cancellous lag screw (AOCLS) and LPF-combined cancellous lag screw (CCLS) groups (p < 0.05); b Maximal axial micromotion of the lateral fragment (MAM-LF) in cyclic compression loading in every 1000 cycles: MAM-LF in the LPF group showed significant differences compared with that in the LPF-AOCLS and LPF-CCLS groups measured at every 1000 cycles (p < 0.05); c Failure load in the load-to-failure test: There was a significant difference in failure load when the LPF group was compared with the LPF-AOCLS and LPF-CCLS groups (p < 0.05).*p value < 0.05

Fig. 6
Fig. 6 Comparison of the three fixation models of MAM-LF in the finite element analysis.a LPF model; b LPF-AOCLS model; c LPF-CCLS model

Fig. 7
Fig. 7 Comparison of implants of peak von Mises Stress (VMS) of the three fixation models; the red label indicates where peak VMS occurred.a LPF model; b LPF-AOCLS model; c LPF-CCLS model

Table 1
Material properties of the FE models used in this study Fig. 4 Depictions of the loads applied in the finite element models

Table 2
Mesh convergence analysis of the finite element models