In Vitro Biomechanical Comparison of 3.5 String of Pearl Plate Fixation to 3.5 Locking Compression Plate Fixation in a Canine Fracture Gap Model Renee‐Claire Malenfant, DVM, and Gary A. Sod, DVM, PhD Department of Veterinary Clinical Science, School of Veterinary Medicine, Louisiana State University, Baton Rouge, Louisiana
Corresponding Author Renee‐Claire Malenfant, DVM, Department of Clinical Science, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA 70803. E‐mail: rmalenfant@lsu. edu Submitted December 2011 Accepted June 2012 DOI:10.1111/j.1532-950X.2014.12095.x
Objective: To compare the 3.5 string of pearls (SOP) plate with a 3.5 mm locking compression plate (LCP) using a fracture model in static loading and cyclic fatigue testing. Study Design: In vitro biomechanical testing of paired tibias with a mid‐diaphyseal ostectomy, stabilized by 1 of 2 locking systems. Sample Population: Cadaveric canine tibiae (n ¼ 24 pairs). Methods: Tibias were randomly divided into 4 equal groups: (1) 4‐point bending single cycle to failure, (2) 4‐point bending cyclic fatigue, (3) torsion single cycle to failure, and (4) torsion cyclic fatigue. Randomly assigned SOP and LCP bridged a 20 mm mid‐diaphyseal ostectomy. Mean test variables values for each method were compared using a paired t‐test within each group with significance set at P 20 kg (mean SD, 21.3 3.43 kg). Dogs were euthanatized for reasons unrelated to orthopedic abnormalities or to this study and the cadavers were donated by local shelters. Tibiae were collected, stripped of all soft tissue, visually inspected for gross abnormalities, wrapped in saline (0.9% NaCl) solution soaked gauzes, bagged in pairs, and stored at 23.3°C until testing. Before preparation for testing, bones were thawed at room temperature (20–22°C) and randomly divided into 4 groups (6 pairs each) for: (1) 4‐ point bending single cycle to failure testing; (2) 4‐point bending cyclic fatigue testing; (3) torsional testing single cycle to failure; and (4) torsion cyclic fatigue testing. Right and left tibias of each group were assigned using a random number generator to LCP or SOP plates and it was assumed that the left and right tibias had no significant differences for the purpose of this study.
Construct Preparation An 8‐hole 3.5 SOP plate was fixed on the medial aspect of each randomly selected tibia with six 3.5 self‐ tapping cortical screws (Synthes Vet). Because there was adequate plate/bone contact, the plate was not contoured before fixation. Screws were place in the following order with numbering of the holes starting proximal: 3rd, 6th, 2nd, 7th, 1st, and finally the 8th screw. The 4th and 5th holes, which would overlie the ostectomy site, were left empty. Screws were tightened but not completely before the ostectomy to ensure the saw would not come in contact with the plate. The plate‐bone assembly was placed in a positioning jig and a 20 mm ostectomy was created in the mid tibia underlying the 4th and 5th screw holes using an oscillating saw (Synthes Ltd.). Care was taken to create an exact 20 mm ostectomy and to make sure that the saw did not come in contact with the plate. Finally, the screws were fully inserted in the same order they were first placed with no modification once the screws were completely screwed in place. An 8‐hole 3.5 LCP plate was fixed on the medial aspect of the contralateral tibia with two 3.5 self‐tapping cortical screws (Synthes Vet) and four 3.5 locking self‐tapping screws (Synthes Vet). These plates also did not require contouring. The cortical screws we placed first in the 2nd and 7th hole
466
followed by the locking screws in the 3rd, 6th, 1st, and 8th hole. Again the 4th and 5th hole were left empty. A 20 mm ostectomy was created in the same location and using the same technique as the with the SOP plates. The screws were then tightened in the same order as for placement (Fig 1). Mechanical Testing A modified MTESTWindows material testing system (AD‐ MET Inc., Norwood, MA) controlling a servo‐hydraulic material testing machine (designed and built by 1 author [GAS], LSU‐SVM, Baton Rouge, LA) equipped with 3 different uniaxial load cells; 1 kN for the 4‐point bending, single cycle to failure; 50 N m for the torsion, single cycle to failure; and 500 N for the 4‐point bending cycling testing. The MTESTWindows system provided closed loop servo control, managed calibration and provided analog/digital conversion of the data acquired and storage in a computer data file. Calibration was verified by an external contractor every 6 months, when the MTS was moved or when a new load cell was installed. MTESTWindows was used to perform load cell calibration, using the shunt calibration technique, and position calibration, before each cyclic fatigue test and before each group of single cycle to failure tests. 4‐Point Bending, Single Cycle to Failure. Plate‐bone assemblies were placed on a bending fixture within the material testing machine. The inner support points were in contact with the lateral cortex of the tibias and the outer support points were in contact with the medial cortex creating the tension and compression forces on the proper cortices. Both inner and outer support points were centered relative to the ostectomy site. The inner load spans were 7.2 and 7.8 cm for the SOP and LCP constructs, respectively, to adapt for the difference in length of the plates (SOP: 10.2 cm and LCP: 10.7 cm) to have the inner loading points between the 1st and
Figure 1
A 3.5 mm LCP construct.
Veterinary Surgery 43 (2014) 465–470 © Copyright 2014 by The American College of Veterinary Surgeons
Malenfant and Sod
2nd and the 7th and 8th holes. The outer support span was 11 cm for all constructs. The distance between the outer and inner supports was 1.9 cm for the SOP and 1.6 cm for the LCP. Bending tests were performed over a single cycle under a displacement control at a constant rate of 0.1 mm/s to failure using the MTS. Load and displacement data was acquired at 0.02‐second intervals throughout the test by analog/digital conversion and stored in a computer data file. 4‐Point Bending, Cyclic Fatigue Testing. Plate‐bone assemblies were placed on the bending fixture within the material testing machine in using the same protocol as the 4‐point bending, single cycle to failure group. A cycle load of 0–200 N was applied at 10 Hz. The maximum load of 200 N was extrapolated from the 4‐point bending single cycle to failure by using 50% of the minimum mean yield load for either plates, thus within the elastic region of the load deformation curve.9 The number of cycles to failure was recorded. In this study failure was described as plastic deformation of implants or catastrophic failure of implants, defined by a sudden decrease in load. Torsional Testing, Single Cycle to Failure. The proximal and distal extremities of tibia‐plate construct were embedded in fiberglass potting material (Light Weight 3 Fiberglass, Evercoat Co., Cincinnati, OH) in pedestals with 1 cm of potting material between the bone and the bottom of the pedestal, as described previously.10 The pedestals incorporated a 2.0 cm portion of each extremity of the bone and were at least 1.0 cm from the proximal and distal end of the SOP or the LCP. The pedestals were positioned within torsional test fixtures attached to the load frame, with the axis of loading aligned along the longitudinal axis of the midshaft of the tibia and not the longitudinal axis of the plate. A universal joint was incorporated within proximal test fixture to prevent generation of a bending moment. The plate‐bone assemblies were loaded under displacement control as a constant rate of 0.0175 rad/s until a rotation of 0.175 rad was attained. Torque and angle data were acquired at 0.01‐second intervals throughout the test by analog/digital conversion and stored in a computer data file. Torsion, Cyclic Fatigue Testing. Plate‐bone assemblies were placed on the torsional test fixtures within the material testing machine using the same protocol as the torsional testing, single cycle to failure group. A cycle load of 0–4.0 N m was applied at 5 Hz. The maximum load of 4.0 N m was extrapolated from the torsional single cycle to failure by using 50% of the minimum mean yield load for either plates, thus within the elastic region of the load deformation curve.9 The number of cycles to failure was recorded. Again, failure was described as plastic deformation of implants or catastrophic failure of implants, defined by a sudden decrease in load.
Comparison of SOP to LCP Fracture Gap Model
of each curve was used to determine stiffness. Yield points were determined as those points where the load–deformation curve deviated from the linear (elastic) region. The yield point was determined using the offset method with a 2% displacement in 4‐point bending and 5% rotation in torsion offset from the linear regression line.9 The yield load for both test and associated displacement were obtained by extrapolation from the yield point to the load axis and displacement axis, respectively. Failure load in all tests and associated displacement were determined from the first indication of failure, that is, the point on the load–deformation curve immediately preceding the first decrease in load. Bending moments and composite rigidities under 4‐point bending, single cycle to failure, were calculated by using the formulas:
BM ¼ F
L 2
where BM is the bending moment, F is the force (yield load or failure load), and L the distance between the inner and outer supports consequently 1.6 and 1.9 cm for the LCP and SOP plates respectively and,
AD ¼ tan1
D L
where AD is the angular displacement, D the displacement (cm), and L is as the bending moment formula. Failure Mode All plate‐bone construct were dismantled after testing to evaluate the site and extend of failure. Any suspicious abnormality was examined with a materials microscope (Leica DM2500 MH) to confirm or reject the failure. Statistical Analysis Data were reported as mean SD for each construct for the yield load, yield displacement, yield bending moment, composite rigidity, failure load, failure displacement and failure bending moment for 4‐point bending in single cycle to failure and cycles to failure. The mean SD was also calculated for each construct for yield load, yield displacement, composite rigidity, failure load, failure displacement for torsional testing in single cycle to failure and cycles to failure. The effect of the plate type was evaluated by paired samples for each construct using t‐tests for paired sample means within each testing group. Significance was set at P
Corresponding Author Renee‐Claire Malenfant, DVM, Department of Clinical Science, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA 70803. E‐mail: rmalenfant@lsu. edu Submitted December 2011 Accepted June 2012 DOI:10.1111/j.1532-950X.2014.12095.x
Objective: To compare the 3.5 string of pearls (SOP) plate with a 3.5 mm locking compression plate (LCP) using a fracture model in static loading and cyclic fatigue testing. Study Design: In vitro biomechanical testing of paired tibias with a mid‐diaphyseal ostectomy, stabilized by 1 of 2 locking systems. Sample Population: Cadaveric canine tibiae (n ¼ 24 pairs). Methods: Tibias were randomly divided into 4 equal groups: (1) 4‐point bending single cycle to failure, (2) 4‐point bending cyclic fatigue, (3) torsion single cycle to failure, and (4) torsion cyclic fatigue. Randomly assigned SOP and LCP bridged a 20 mm mid‐diaphyseal ostectomy. Mean test variables values for each method were compared using a paired t‐test within each group with significance set at P 20 kg (mean SD, 21.3 3.43 kg). Dogs were euthanatized for reasons unrelated to orthopedic abnormalities or to this study and the cadavers were donated by local shelters. Tibiae were collected, stripped of all soft tissue, visually inspected for gross abnormalities, wrapped in saline (0.9% NaCl) solution soaked gauzes, bagged in pairs, and stored at 23.3°C until testing. Before preparation for testing, bones were thawed at room temperature (20–22°C) and randomly divided into 4 groups (6 pairs each) for: (1) 4‐ point bending single cycle to failure testing; (2) 4‐point bending cyclic fatigue testing; (3) torsional testing single cycle to failure; and (4) torsion cyclic fatigue testing. Right and left tibias of each group were assigned using a random number generator to LCP or SOP plates and it was assumed that the left and right tibias had no significant differences for the purpose of this study.
Construct Preparation An 8‐hole 3.5 SOP plate was fixed on the medial aspect of each randomly selected tibia with six 3.5 self‐ tapping cortical screws (Synthes Vet). Because there was adequate plate/bone contact, the plate was not contoured before fixation. Screws were place in the following order with numbering of the holes starting proximal: 3rd, 6th, 2nd, 7th, 1st, and finally the 8th screw. The 4th and 5th holes, which would overlie the ostectomy site, were left empty. Screws were tightened but not completely before the ostectomy to ensure the saw would not come in contact with the plate. The plate‐bone assembly was placed in a positioning jig and a 20 mm ostectomy was created in the mid tibia underlying the 4th and 5th screw holes using an oscillating saw (Synthes Ltd.). Care was taken to create an exact 20 mm ostectomy and to make sure that the saw did not come in contact with the plate. Finally, the screws were fully inserted in the same order they were first placed with no modification once the screws were completely screwed in place. An 8‐hole 3.5 LCP plate was fixed on the medial aspect of the contralateral tibia with two 3.5 self‐tapping cortical screws (Synthes Vet) and four 3.5 locking self‐tapping screws (Synthes Vet). These plates also did not require contouring. The cortical screws we placed first in the 2nd and 7th hole
466
followed by the locking screws in the 3rd, 6th, 1st, and 8th hole. Again the 4th and 5th hole were left empty. A 20 mm ostectomy was created in the same location and using the same technique as the with the SOP plates. The screws were then tightened in the same order as for placement (Fig 1). Mechanical Testing A modified MTESTWindows material testing system (AD‐ MET Inc., Norwood, MA) controlling a servo‐hydraulic material testing machine (designed and built by 1 author [GAS], LSU‐SVM, Baton Rouge, LA) equipped with 3 different uniaxial load cells; 1 kN for the 4‐point bending, single cycle to failure; 50 N m for the torsion, single cycle to failure; and 500 N for the 4‐point bending cycling testing. The MTESTWindows system provided closed loop servo control, managed calibration and provided analog/digital conversion of the data acquired and storage in a computer data file. Calibration was verified by an external contractor every 6 months, when the MTS was moved or when a new load cell was installed. MTESTWindows was used to perform load cell calibration, using the shunt calibration technique, and position calibration, before each cyclic fatigue test and before each group of single cycle to failure tests. 4‐Point Bending, Single Cycle to Failure. Plate‐bone assemblies were placed on a bending fixture within the material testing machine. The inner support points were in contact with the lateral cortex of the tibias and the outer support points were in contact with the medial cortex creating the tension and compression forces on the proper cortices. Both inner and outer support points were centered relative to the ostectomy site. The inner load spans were 7.2 and 7.8 cm for the SOP and LCP constructs, respectively, to adapt for the difference in length of the plates (SOP: 10.2 cm and LCP: 10.7 cm) to have the inner loading points between the 1st and
Figure 1
A 3.5 mm LCP construct.
Veterinary Surgery 43 (2014) 465–470 © Copyright 2014 by The American College of Veterinary Surgeons
Malenfant and Sod
2nd and the 7th and 8th holes. The outer support span was 11 cm for all constructs. The distance between the outer and inner supports was 1.9 cm for the SOP and 1.6 cm for the LCP. Bending tests were performed over a single cycle under a displacement control at a constant rate of 0.1 mm/s to failure using the MTS. Load and displacement data was acquired at 0.02‐second intervals throughout the test by analog/digital conversion and stored in a computer data file. 4‐Point Bending, Cyclic Fatigue Testing. Plate‐bone assemblies were placed on the bending fixture within the material testing machine in using the same protocol as the 4‐point bending, single cycle to failure group. A cycle load of 0–200 N was applied at 10 Hz. The maximum load of 200 N was extrapolated from the 4‐point bending single cycle to failure by using 50% of the minimum mean yield load for either plates, thus within the elastic region of the load deformation curve.9 The number of cycles to failure was recorded. In this study failure was described as plastic deformation of implants or catastrophic failure of implants, defined by a sudden decrease in load. Torsional Testing, Single Cycle to Failure. The proximal and distal extremities of tibia‐plate construct were embedded in fiberglass potting material (Light Weight 3 Fiberglass, Evercoat Co., Cincinnati, OH) in pedestals with 1 cm of potting material between the bone and the bottom of the pedestal, as described previously.10 The pedestals incorporated a 2.0 cm portion of each extremity of the bone and were at least 1.0 cm from the proximal and distal end of the SOP or the LCP. The pedestals were positioned within torsional test fixtures attached to the load frame, with the axis of loading aligned along the longitudinal axis of the midshaft of the tibia and not the longitudinal axis of the plate. A universal joint was incorporated within proximal test fixture to prevent generation of a bending moment. The plate‐bone assemblies were loaded under displacement control as a constant rate of 0.0175 rad/s until a rotation of 0.175 rad was attained. Torque and angle data were acquired at 0.01‐second intervals throughout the test by analog/digital conversion and stored in a computer data file. Torsion, Cyclic Fatigue Testing. Plate‐bone assemblies were placed on the torsional test fixtures within the material testing machine using the same protocol as the torsional testing, single cycle to failure group. A cycle load of 0–4.0 N m was applied at 5 Hz. The maximum load of 4.0 N m was extrapolated from the torsional single cycle to failure by using 50% of the minimum mean yield load for either plates, thus within the elastic region of the load deformation curve.9 The number of cycles to failure was recorded. Again, failure was described as plastic deformation of implants or catastrophic failure of implants, defined by a sudden decrease in load.
Comparison of SOP to LCP Fracture Gap Model
of each curve was used to determine stiffness. Yield points were determined as those points where the load–deformation curve deviated from the linear (elastic) region. The yield point was determined using the offset method with a 2% displacement in 4‐point bending and 5% rotation in torsion offset from the linear regression line.9 The yield load for both test and associated displacement were obtained by extrapolation from the yield point to the load axis and displacement axis, respectively. Failure load in all tests and associated displacement were determined from the first indication of failure, that is, the point on the load–deformation curve immediately preceding the first decrease in load. Bending moments and composite rigidities under 4‐point bending, single cycle to failure, were calculated by using the formulas:
BM ¼ F
L 2
where BM is the bending moment, F is the force (yield load or failure load), and L the distance between the inner and outer supports consequently 1.6 and 1.9 cm for the LCP and SOP plates respectively and,
AD ¼ tan1
D L
where AD is the angular displacement, D the displacement (cm), and L is as the bending moment formula. Failure Mode All plate‐bone construct were dismantled after testing to evaluate the site and extend of failure. Any suspicious abnormality was examined with a materials microscope (Leica DM2500 MH) to confirm or reject the failure. Statistical Analysis Data were reported as mean SD for each construct for the yield load, yield displacement, yield bending moment, composite rigidity, failure load, failure displacement and failure bending moment for 4‐point bending in single cycle to failure and cycles to failure. The mean SD was also calculated for each construct for yield load, yield displacement, composite rigidity, failure load, failure displacement for torsional testing in single cycle to failure and cycles to failure. The effect of the plate type was evaluated by paired samples for each construct using t‐tests for paired sample means within each testing group. Significance was set at P
