Patel V, Venkatarayappa I, Prayson MJ, Goswami T. Biomechanical evaluation of hybrid locking plate constructs. Hard Tissue 2013 Jun 01;2(4):32.

Transcription

1 Patel V, Venkatarayappa I, Prayson MJ, Goswami T. Biomechanical evaluation of hybrid locking plate constructs. Hard Tissue 2013 Jun 01;2(4):32. Licensee OA Publishing London Creative Commons Attribution License (CC-BY) Competing interests & Conflict of interests: None declared. All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript. All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.

2 Biomechanical evaluation of hybrid locking plate constructs V Patel 1, I Venkatarayappa 2, MJ Prayson 2, T Goswami 1, 2 1. Department of Biomedical, Industrial and Human Factors Engineering 2. Orthopaedic Surgery, Sports Medicine and Rehabilitation Wright State University, 3640 Colonel Glenn Hwy, Dayton, OH Abstract Hybrid locking compression plate constructs is commonly used in fracture stabilization of long bones. Locking compression plates may be used with locking and non-locking screws. Locking screws have the threaded heads, which enable to lock the screw within the plate hole and eliminate motion between screw and plate. They provide stability to the fixation and are most suited for osteoporotic bones and in stabilization of fracture with bone defects and significant comminution. Non-locking screws are suitable for normal bones and for simple fractures. Recommendation for the optimal combination of locking and non-locking screws is desired in an operating room setting though limited information available in the literature. Cost of locking screw is up to ten times more than non-locking screw. In order to reduce operating room costs, there is a need to determine an optimum arrangement of locking and non-locking screws that provides the best dynamic stiffness and stability of the construct during and after the healing occurred. Fracture models with synthetic femora were created using locking compression plate and different combination of locking and non-locking screws. Four groups with five samples in each group were tested for axial and rotational stiffness. Simulated solid models were created using Solidworks modeling 1

3 and finite element analysis was performed and the results were compared with experimental results. The results from this experimental program indicate that locking screws should be placed near the fracture gap to gain axial and torsional strength. Also, locking screws impart higher resistance to loosening of the adjacent level non-locking screws as found in this synthetic femur study. Keywords locking compression plate, locking screw, non-locking screw, dynamic stiffness, finite element model, displacement 2

4 1.0 Introduction An open reduction and internal fixation (ORIF) procedure is performed over 4 million times in the United States annually. A hybrid plate construct involves the use of locking screw and conventional cortical or cancellous screw. Internal fixation devices work as load sharing or load bearing devices. Plate osteosynthesis in comminuted fracture acts as load bearing device where the plate assumes all the loads. Compression plating in simple fracture pattern works as load sharing devices where bone assumes most of the load. Fracture healing leading to bone union depends on the devices used and their biomechanical behavior. Hence it is important to obtain stable fixation to prevent implant failure. Implants are not routinely removed unless there is infection or mechanical failures. Internal fixation devices are of many types. Some of the commonly used devices include K-wires, Plates, Screws, intramedullary rods etc., K-wires are mainly used as reduction tool and sometimes in stabilizing fractures of small bone in hand and feet. Different types of plates used for internal fixation include locking compression plate, dynamic compression plate, and reconstruction plate. Dynamic and locking compression plates are more commonly used. Locking compression plate acts as internal external fixator. They have very limited contact with the bone and hence reduce vascular damage. This has a great advantage in osteoporotic bone and in fractures with significant comminution and bone loss. Locking compression plates provide relative stability at the fracture site and facilitate bone healing by secondary intention Conventional plate osteosynthesis offers stability by friction between plate-bone interface. On axial loading, bending force is applied on the screws, which generate a 3

5 shear force. When axial load exceeds the friction force, movement occurs between the plate and screws, known as toggle, and may result in bending or fracture of screws. The change in plate screw angle limits failure loads to the strength of thread purchase. Screws alone may be used as primary mode of fixation in small bone fractures in hands and feet. Intramedullary rods are more commonly used in midshaft fractures of femur and tibia. Previous studies have evaluated the mechanical properties of hybrid plate construct in fourth-generation composite bone model 11. The potential advantage of locked screws in improving the axial and torsional strength and different combination of locked and nonlocked screws have been studied 12. To our knowledge there is no study, evaluating the dynamic stiffness of Hybrid construct at different stages of healing process. 2.0 Materials and Methods In this study synthetic femur models (Model 3403, pacific Research Laboratories, Vashon, USA) which contain epoxy glass fiber as shallow cylinders filled with polyethylene were used. Fracture comminution was simulated by an osteotomy gap of 2cm in the metaphyseal region using twenty synthetic femur models. Ten hole precontoured lateral distal femoral locking plate implants were used and simulated fractures were stabilized with different combination of locking and non-locking screws. The femur fracture constructs were divided into four groups as shown in Table-1. Five femurs were prepared for each of the 4 groups. The test protocol was kept same for all 4 groups of testing where load was cycled in the negative Y direction from 50 to 700 N and a torsion rotation of + 5º /cycle applied in a sinusoidal waveform. At the end of 50,000 cycles tests were stopped. The details of the specimen preparation and specimen testing are available 4

6 elsewhere Simulated solid models were constructed to perform finite element analysis so that these results could be compared with the experimental results. 2.1 Solidworks Modeling Solid models of LCP were created similar to Synthes condylar plates through Solidworks 2007, a three-dimensional (3D) Computer Aided Design (CAD) software. The nonlocking screw heads were minimized in size to decrease contact with the plate hole as shown in Figure 1. The distal condylar flared portion of the plate was designed with loft and fillet (Loft and Fillet are features used in Solidworks for creating transitions and external/internal round surfaces, respectively) operations as shown in Figure 2. A final assembly was created joining the various individual model portions. This assembly was similar in appearance to the femoral constructs tested experimentally. Additional groups were created for completeness. Plating constructs with three locking screws and one nonlocking screw were developed as well as plates with four locking screws. The bony segments (condylar and shaft) were subtracted for further analysis in ANSYS, (ANSYS Inc, Canonsburg, PA). Appropriate constraints were applied at the plate ends and on the screws. The model was then saved in a para-solid format. Three-dimensional models of the femur were also created using Mimics (Materialise, Inc, Plymouth, MI), which is a 3D modeling software that imports computed tomography (CT) imaging data. The CT images of the specimens were acquired from Miami Valley Hospital in Dayton, OH. Three-dimensional models, similar to the anatomical specimens, were achieved using the Mimics tools as seen in Figure 3. 5

7 2.2 FEA analysis in ANSYS The Solidworks models stored in parasolid format were imported into ANSYS. Stainless steel (316L) properties were assigned to the plate and screws built with SHELL 93 elements. Loads were applied to the screws with the plate and screws constrained. A time harmonic mode was chosen which converts loads in terms of displacement and stress. Loads of 300N were applied to each screw in a negative Y direction (downward). From the obtained results, construct dynamic stiffness and deformation were measured. 3.0 RESULTS 3.1 Finite Element Analysis Table-2 summarizes the results of finite element analysis in terms of maximum stress and maximum displacement that occurred in different screws. Maximum stress occurred in plates with all non-locking screws. Stresses were the lowest in plates with all locking screws. A maximum stress of 449 N/m 2 occurred in plates with all non-locking screws, 427 N/m 2 in plates with three non-locking screws, 373 N/m 2 and 341 N/m 2 in plates with two and one non-locking screws, respectively. A lowest maximum stress of 190 N/m 2 occurred in plates with all 4 locking screws. Total displacement of locked construct was, however, higher than all other combinations. It may be noted that the displacement considered here were in nano-meters, and operating room practice allows 10% of the fracture gap as a rule of thumb. Figure 4 shows the ANSYS models with all 4 locked and non-locked constructs. 3.2 Experimental Program Experimental data generated on 20 synthetic femurs has been tabulated in Table-3. Groups 2 and 3 exhibited higher torque to loosen the screws. However, locking screws 6

8 when positioned either immediately after the osteotomy gap, or at the distal end of the plate, exhibited similar amount of torque required to loosen the screw. More than onehalf of torque was retained in the screw at the end of 50,000 cycles. Constructs made with conventional screw, on the other hand, loosened most of their insertion torque and were loose enough that torque meter did not read the remaining torque. Remaining torque was measured by subtracting the loosening torque from insertion torque. The conventional screw, located in Screw-2 position, maintained somewhat lower torque than Screw-3. The average torque of locking screws, farthest and closest from the osteotomy gap, remained very similar Similarly, non-locking screws farthest from the gap showed low loosening torque compared to the non-locking screws nearer to the fracture gap. Table-4 summarizes average remaining torque, percentage loosening torque, axial stiffness, average torsional stiffness, and average deformation of each of the screws. It may be noted that a construct, collectively with locking and non-locking screws, had an axial stiffness range from 200 to 4000 N/mm, whereas the torsional stiffness range was N mm/º rotation and average deformation range was mm. The dynamic stiffness was plotted with respect to the number of cycles for all the 4 groups, Figures 6-9. The dynamic stiffness in Group 1 varied with respect to number of cycles in a sine function, representative of harmonic oscillation and may indicate the density and/or porosity variation in the construct, and is not of interest in this study. However, the Group 1 harmonics may indicate severe displacement changes during the tests. The displacement data was reviewed for all the groups and no trends were found that can be reported here. In all the tests, the displacement at the end of 50,000 cycles was the highest. The total deformation varied, as noted earlier from 0.5 to a worst case 7

9 scenario of 2.5 mm. The displacement data for Group 1 is shown in Figure10 showing normal, gradual accruing of deformation as the cycles continue. Since the significant drop in stiffness occurs during the initial stages of cyclic loading, as shown in Figures 6-9, the first 20,000 cycles are therefore very important in fracture fixation and healing. The remaining torque begins to deplete at this point and may be the reason why loosening of screws or screw pull-out is observed as a failure mode. The negative slope of the dynamic stiffness for Group 4 constructs may be described by Dynamic stiffness = -1E-20N 5 + 2E-15N 4-1E-10N 3 + 3E-06N N Where, N is the number of cycles. The efficacy of a hybrid construct depends on the dynamic stiffness exhibited by the construct. From Figures 6-9 it is obvious that lower stiffness ranges found for Groups 1-3 may loosen the constructs in the long run, and must show the remaining stiffness higher than N/mm. Therefore, Group 4 provides the most effective fixation stability with 2 locking screws one on each proximal and distal end from the osteotomy. Fixation efficacy was found to be with the use of 2 locking screws for the study groups investigated in this paper with dynamic stiffness range of 200 N/mm at the end of 50,000 cycles. However, the dynamic stiffness was expected to be much higher during the insertion of the device, 500 N/mm, as shown in Figure 9 using 2 locking screws. None of the other constructs show higher starting stiffness as Group 4. The starting stiffness ranges from 100 to 300 N/mm. This study used 2 cm osteotomy gap, where high stress/strain was likely to develop causing significant amount of axial and angular displacements. Additional work is recommended for smaller osteotomy gaps to determine the efficacy of the hybrid locking plate constructs. 8

10 4. DISCUSSION The LCP is subjected to static and cyclic loads in vivo which generate extremely complicated stress systems in the device 2. Materials with a very high Young s Modulus make the plate stiff. Stiff plates do not transfer as much stress to the bone in the local area causing the bone to thin or deplete its composition. This phenomenon is known as stress shielding and causes osteopenia 3. It is usually visible on the X-rays showing a region darker than the superior or inferior regions. Therefore, materials that possess suitable stiffness must be considered for use as LCP. Materials with a low modulus of elasticity do not provide enough rigidity to the bone to heal the fracture, and materials with a high elastic modulus increase rigidity and lower stresses in the bone; thus causing stress shielding. For the fixation devices, stainless steel- 316L, remains the preferred material of choice and used in this study. Efficacy of the LCP performance depends on fracture type. Conventional compression plate performs well for normal quality of bone and fracture with normal or partial contact between fragments. When both ends of the bone fragments are not in contact with each other, a bridge plate technique may be used either with locked or standard screws based on the bone quality. A combination technique is employed for simple oblique or articular fracture with more standard screws and fewer locking screws. An osteotomy gap of 2 cm was used in this study, simulating significant bone loss and fragmented fracture. The length of LCP is dependent on fracture length and loads being transferred to the plate (e.g. bending, pull out) 4. The ratio of the plate length to the fracture length is called plate span width 5. Guatier and Sommer recommended plate span width to be from 2 to 3 for comminuted fracture and 8 to 10 for simple fracture 6. This suggests that for more 9

11 comminuted fracture; a long plate provides higher axial and torsional stability than the short plate 7. Working length is the length between two screws of two different fracture fragments. With a small fracture length and working length the bone ends do not come in contact with each other thus reducing callus formation 8. This increases stress and strain during torsion loading. Stresses induced in a plate with a 6 mm fracture gap are higher than plate with a 1 mm fracture gap. The stress in a screw decreases with smaller fracture gap 7. In osteoporotic bone, the working length is kept small because of small bone thickness compared to the normal bones. When torsion load is applied, the screws tend to pullout. The LCP plates have point contact at the undersurface thus preserving periosteum blood supply. Conventional plates exert N force when screws are tightened to the bone [9]. LCP plates reduce this load and preserve the blood supply. Reduced contact between the bone and the plate improves bone growth 9. Stoffel suggested that by increasing distance of plate from the bone 2 to 6 mm, the axial and torsional stability decreased by 10-15% 7. Freeman et al. have studied the effect of the number and location of locked screws on the mechanical properties of hybrid plate construction on osteoporotic bone model. Seven different constructs with 2 unlocked and 5 hybrid constructs were tested. They determined that 4 screws per side increased the stiffness to 33% when compared to 3 screws in each fragment. They did not find any difference when 1 or 2 unlocked screws were replaced by locked screws on each fragment. However, replacing 3 unlocked screws on each side with locked screws increased the torsional stiffness by 24%. They concluded that at least 3 bicortical locked screws on each side of the fracture are needed in a 10

12 osteoporotic bone model to increase the torsional strength. Locked screws when placed between the fracture and unlocked screws protect the unlocked screws from loosening 12. While literature cited effects of these parameters on resulting stiffness and stability of the fixation, limited amount of data was available showing the dynamic stiffness of the construct. The dynamic stability is pivotal in holding the construct especially in early phase during fracture healing. Therefore, this paper provides the most desired dynamic stiffness of hybrid constructs with different screw configurations at various stages of healing process that will be useful in the treatment of comminuted fractures. 11

13 Conclusion The hybrid locking compression plate constructs tested during this research show the following trends: The remaining torque in each of the locking screw remained similar, even though their placement changed Conventional screws loosened at a faster rate than the locking screws The dynamic stiffness of the construct was found as a function of number of locking screws and where they were placed, proximal and distal end placement was found to be an effective position Fixation efficacy was proposed to be a function of dynamic stiffness and may be predicted using simple mathematic models, for the conditions investigated in this study for a 2 cm osteotomy gap. Depletion of dynamic stiffness occurs with respect to time and cycles, therefore, needs to be accounted for in surgery. Acknowledgement Rinki Goswami, Senior in Biological Engineering, Cornell University, edited the manuscript. 12

14 References 1. Sheng H., Ching-C.H., J. L. Wang, Ching-Kong Chao, Jinn Lin., Mechanical tests and finite element models for bone holding power of tibial locking screws, Clinical Biomechanics 19, Sudhakar K.V., Metallurgical investigation of a failure in 316L stainless steel orthopaedic implant. Engineering Failure Analysis 12, ASTM F136, Standard specification for wrought titanium-6 aluminum- 4 vanadium ELI (Extra Low Interstitial) alloy for surgical implant applications (UNS R56401), ASTM International, West Conshohocken, PA 4. Miller D., Goswami T., A review of locking compression plate biomechanics and their advantages as internal fixators in fracture healing. Clinical Biomechanics 22, Rozbruch, S.R., Muller, U., Gautier, E., Ganz, R., The evolution of femoral shaft plating technique, Clin. Orthop., Gautier E, Sommer C., 2003.Guidelines for the clinical application of the LCP. Injury 34, SB63 SB76 7. Stoffel K, Dieter U, Stachowiak G, Gachter A, Kuster M., 2003.Biomechanical testing of the LCP how can stability in locked internal fixators be controlled? Injury 34, SB11 SB Kubiak E, Fulkerson E, Strauss E, Egol K, The evolution of locked plates. The Journal of Bone and Joint Surgery, 88-A, Supplement 4 9. Perren, S.M., Evolution and rational of locked internal fixator technologyintroductory remarks, Injury 32 (Suppl 2), S-B3-S-B9 13

15 10. Klaue, K., Fengels, I., Perren, S.M., Long-term effects of plate osteosynthesis: comparison of four different plates, Injury 31, S-B51- S-B Patel, V., Biomechanical evaluation of locked and non-locked constructs under axial and torsion loading, M.S. Thesis, Wright State University, Dayton, OH, USA. 11. Goswami, T, Patel, V. Dalstrom, D., and Prayson, MJ Mechanical evaluation of fourth-generation composite femur hybrid locking plate constructs, Materials in Medicine, Journal of Materials Science, 22, 9, Freeman AL, Tornetta P 3 rd, Schmidt A, Bechtold J, Ricci W, Flemming M.2010, How much do locked screws add to the fixation of hybrid plate constructs in osteoporotic bone? J Orthop Trauma. 24, (3),

16 Table-1. Specifications of femurs in the experimental program Group Femur No Screw-1 Screw-2 Screw-3 Screw-4 1 1,2,3,4,5 NL NL NL NL 2 1,2,3,4,5 NL NL NL L 3 1,2,3,4,5 L NL NL NL 4 1,2,3,4,5 L NL NL L Notes, L: locking, NL: non-locking. 15

17 Table-2: Stress and displacement development in analytical models, ANSYS results Femoral Construct Stress (N/m 2 ) Displacement(m) 4 Non-Locking *10^-8 1 Locking *10^-9 3 Non-Locking 2 Non-Locking 373 5*10^-9 2 Locking 1 Non-Locking *10^-11 3 locking 4 Locking *10^-8 16

18 Table-3. Remaining torque (RT) measured in Nm, in each of the screws. Femur type RT in Screw-1 RT in Screw-2 RT in Screw-3 RT in Screw-3 Gr.1, Femur Gr.1, Femur Gr. 1, Femur Gr. 1, Femur Gr. 1, Femur Gr. 2, Femur Gr. 2, Femur Gr. 2, Femur Gr. 2, Femur Gr. 2, Femur Gr. 3, Femur Gr. 3, Femur Gr. 3, Femur Gr. 3, Femur Gr. 3, Femur Gr. 4, Femur Gr. 4, Femur Gr. 4, Femur Gr. 4, Femur Gr. 4, Femur Color code- yellow- locked screws, green- non-locked screws 17

19 Table-4. Results of individual femur constructs showing average remaining torque (ART) in Nm, percentage loosening (PL), average axial stiffness (AAS) in N/mm, average torsional stiffness (ATS) in Nmm/ rotation, average deformation (AD) in mm and observation on each test Test No ART PL AAS ATS AD Observations Gr.1, F Gr.1, F Failed at 44,3365 cycles Gr. 1, F Failed at 26,923 cycles Gr. 1, F Gr. 1, F Failed at cycles Gr. 2, F Gr. 2, F Gr. 2, F Gr. 2 F Gr. 2, F Gr. 3, F Gr. 3, F Gr. 3, F Screw-1 failed Gr. 3, F Gr. 3, F Screw-1 pullout 18

20 Gr. 4, F Gr. 4, F Gr. 4, F Gr. 4, F

21

22

23

24 Figure 4: Finite element analysis results from ANSYS, one typical data sheet illustrated above and numerical data tabulated in Table 2. 23

25

26

27

28

29

30