Citation
Establishing metrics for comparison of ankle fusion devices

Material Information

Title:
Establishing metrics for comparison of ankle fusion devices patient-specific-finite-element analysis of three intramedullary nails for tibiotalocalcaneal fusion
Added title page title:
Patient-specific-finite-element analysis of three intramedullary nails for tibiotalocalcaneal fusion
Creator:
Terrill, Patrick K. ( author )
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (29 pages) : ;

Thesis/Dissertation Information

Degree:
Master's ( Master of science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Mechanical Engineering, CU Denver
Degree Disciplines:
Mechanical engineering

Subjects

Subjects / Keywords:
Ankle ( lcsh )
Ankle ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Introduction/Purpose: Tibiotalocalcaneal (TTC) arthrodesis is a salvage procedure for patients with severe osteoarthritis and other degenerative ankle conditions. Oftentimes, an intramedullary (IM) nail is implanted across the joints and then fixed with tibial and calcaneal screws. Maintaining compression and load sharing are both largely desired to promote fusion via primary bone healing; however, compression can be lost due to small amounts of bone resorption and IM nails are now being made from carbon-fiber epoxy to minimize stress shielding. To date, no one has been able to directly characterize or compare the specific amount of these parameters across nails in a single model. The purpose of this study is to compare influence of nail design and materials for compressive and load-sharing properties using a patient-specific finite-element model. Methods: A titanium nail, a pseudoelastic nickel-titanium nail, and carbon fiber-epoxy nail were investigated for (1) load sharing between the nail body and the tibia under gait loading and (2) compression less as a function of resorption in the talus. A patient-specific model of the ankle, both in geometry and material properties, was generated from a quantitative computed tomography (QCT) scan of a healthy leg. The nail models were generated from a 3D scan imported into SCANIP. Compression in the nickel-titanium nail was simulated by prestretching the pseudoelastic compressive element. Conversely, compression in the titanium and carbon-fiber nails were generated by giving the nail jacket an orthotropic contraction coefficient in the model. After compression, each nail was subjected to an applied gait load that peaked at 1121N. Resorption was simulated using a thin compressible layer of bone in the talus and decreasing the modulus and Poisson's ratio. Results: Surprisingly, the carbon-fiber nail showed similar stress shielding to the titanium nail, with 72% and 77% of the stress being transferred through the devices instead of the ankle, respectively. Even though carbon fiber-epoxy has a significantly lower modulus than titanium (75 GPa vs 110 GPa), the overall stiffness of the nails was still much greater than that of bone (~30,000 N/mm vs. ~44,000 N/mm vs. ~3,000 N/mm, respectively). The pseudoelastic nail only shielded 32% of the stress values by comparison. For the titanium and carbon-fiber nails, over 85% of the initial compression provided by the nail drops with 0.10 mm of resorption. The pseudoelastic nail maintained 90% of its initial compression after 0.10 mm of resorption. Conclusions: IM nail design and materials played a significant role in maintaining compression and load sharing. The pseudoelastic nail had the lowest degree of stress-shielding (32%) and maintained compression for over 0.10 mm of simulated resorption. Constant compression and the avoidance of "resorption gapping" is paramount to drive primary bone healing in joint fusions due to lack of periosteal/endosteal anatomy crossing the fusion site, thus crippling the ability for secondary bone healing (callus healing). This model allows for direct comparison between devices and can be used pre-operatively to predict patient-specific performance and help aid in device selection for TTC fusion. ( ,,,,,, )
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Patrick K. Terrill.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10153 ( NOTIS )
1015342802 ( OCLC )
on1015342802
Classification:
LD1193.E55 2017m T47 ( lcc )

Downloads

This item has the following downloads:


Full Text
ESTABLISHING METRICS FOR COMPARISON OF ANKLE FUSION DEVICES: PATIENT-
SPECIFIC FINITE-ELEMENT ANALSYIS OF THREE INTRAMEDULLARY NAILS FOR
TIBIOTALOCALCANEAL FUSION by
PATRICK K. TERRILL
B.S., Virginia Polytechnic Institute and State University, 2011
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Mechanical Engineering Program
2017


2017
PATRICK K. TERRILL ALL RIGHTS RESERVED
11


This thesis for the Master of Science in Mechanical Engineering degree by
Patrick K. Terrill
has been approved for the Mechanical Engineering Program by
R. Dana Carpenter, Chair Christopher M. Yakacki Kai Yu
Date: April 28 2017
m


Terrill, Patrick K. (M.S., Mechanical Engineering Program)
Establishing Metrics for Comparison of Ankle Fusion Devices: Patient-Specific Finite-Element Analysis of Three Intramedullary Nails for Tibiotalocalcaneal Fusion Thesis directed by Professor R. Dana Carpenter
ABSTRACT
Tibiotalocalcaneal (TTC) arthrodesis is a salvage procedure for patients with severe osteoarthritis and other severe ankle conditions. The procedure is done by inserting an intramedullary (IM) nail through bottom of the calcaneus and up through the talus and the tibia. The nail is fixed with several screws and compression is applied by the nail and screws to the three bones in order to promote fusion. The purpose of this study is to compare different IM nails in a simulated ankle fusion using Finite Element Analysis (FEA). A pseudo-elastic, conventional titanium, and conventional carbon-fiber nail were all tested for load sharing between the nail body and the tibia under gait loading and compression with a resorption zone. A model of the bones was generated from a quantitative computed tomography (QCT) scan of a healthy leg. Pseudo-elastic compression was simulated by pre-stretching the pseudo-elastic compressive element in the by itself and then inserting it into the nail jacket. The conventional static nails, were compressed by giving the nail jacket an orthotropic thermal expansion coefficient and reducing the temperature in the model. After compression, each nail was subjected to an applied gait load. In the pseudo-elastic model, the tibia was subjected to 68% of the peak load during the gait cycle. The titanium and carbon-fiber nail models had tibia load values of 23% and 28%, respectively. The resorption for each model was simulated using a layer of bone in the talus with varying Youngs Moduli and a Poissons ratio of 0.01. For the static nails, the amount of compression provided by the nail drops more than 80% when exposed to a prescribed resorption zone. The pseudo-elastic nail maintains compression when exposed to resorption up to 4% of the locked strain value.
The form and content of this abstract are approved. I recommend its publication.
Approved: R. Dana Carpenter
IV


ACKNOWLEDGEMENTS
I would like to thank my advisor, Prof. Dana Carpenter for his continued encouragement and assistance throughout my research. His background knowledge of the subject matter and the associated software was critical for helping me complete my thesis in a timely manner. He also has excellent taste in movies.
I would like to thank my other advisor, Prof. Chris Yakacki for his assistance throughout my research. He has a wealth of knowledge on the subject matter and the real world applications, problems, and needs in the field of ankle arthrodesis nails. His previous research helped direct mine to the topic I have been working on. His extensive knowledge of movie quotes has also been there to lighten the mood at all times.
I would like to thank the members of the Smart Materials and Biomechanics Laboratory (SMAB LAB) at CU Denver for their support during my studies and research. Sam Mills, Nick Traugutt, Mohand Saed, Ross Volpe, Hannah Tifft. I would especially like to thank Ravi Patel and Ryan Anderson for their previous work on my research topic and their help with learning the software required and subject matter.
I would finally like to thank my family. My parents, Cathy and Allen Terrill, and my sister, Samantha, supported and encouraged me when I decided to quit my job and pursue my graduate degree.
v


TABLE OF CONTENTS
L Introduction...............................................................................1
Ankle Fusion Impact.......................................................................1
Background................................................................................2
Current Technology........................................................................2
Research Goals............................................................................3
II. Methods..................................................................................4
Pseudo-elastic Nail Model.................................................................4
Static Nail Model.........................................................................7
Resorption Models.........................................................................9
Analyzing the Results....................................................................10
HE Results..................................................................................12
Pseudo-elastic Nail with Gait............................................................12
Titanium Nail with Gait..................................................................12
Carbon Fiber Nail with Gait..............................................................13
Gait-Foad Sharing Characteristics of Each Nail...........................................14
Compression Foss with Different Resorption Values........................................15
IV. Discussion..............................................................................17
Ankle Arthrodesis Nails Under Gait Foad..............................................17
Ankle Arthrodesis Nails with a Resorption Zone...........................................17
vi


LIST OF FIGURES
Figure 1: An Ankle Arthrodesis Nail...................................................1
Figure 2: Intramedullary nail types by generation (Yakacki)...........................3
Figure 3: ScanIP model comparison for the Pseudo-elastic nail masks (Right) and the
unmodified bone model masks (Left)..........................................4
Figure 4: Load on Heel During Gait in Percent Body Weights (Sacco)....................7
Figure 5 Gait Load Force Sites and Magnitude...........................................7
Figure 6 QCT Scan and Solidworks Model Combined in ScanIP..............................8
Figure 7 ScanIP Model with the Addition of a Resorption Zone..........................10
Figure 8 Reaction Force of Bone in Z-direction at Slice 250.......................... 10
Figure 9: Force Distribution during Gait Load for a Pseudo-elastic Nail Model.........12
Figure 10: Force Distribution during Gait Load for a Conventional Static Titanium Ankle
Arthrodesis Nail...........................................................13
Figure 11: Force Distribution during Gait Load for a Conventional Static Carbon Fiber Body
Ankle Arthrodesis Nail.....................................................14
Figure 12: Load Sharing of Different Ankle Arthrodesis Nails at Peak Gait Load........15
Figure 13: Compression Values of Different Ankle Arthrodesis Nails when Exposed to a
Specified Resorption Zone Modulus..........................................16
vii


CHAPTER 1
INTRODUCTION
Ankle Fusion Impact
Tibio-talo-calcaneal (TTC) arthrodesis is a surgical salvage procedure performed to treat ankle pain and trauma related to pathologies such as osteoarthritis, rheumatoid arthritis, neuropathic arthropathy, talar osteonecrosis, osteomyelitis, failed total ankle replacement and other diseases of the ankle.720 TTC arthrodesis is often performed as an alternative to amputation. The purpose of the procedure is to provide a stable, pain-free union of the bones of the hindfoot. Reviews of surgical procedures have found the union rate of TTC arthrodesis procedures to be as high as 94.6% (195 fusions of 206 procedures); however, the occurrence of a non-union can be seen in up to 40% of procedures involving complex cases of ankle disease.5811"13151621242528 in a cost nnalysis of ankle procedures, tibio-talo fusion alone is seen to cost as much as $32,683 for a single non-arthroscopic procedure.23
Figure 1: An Ankle Arthrodesis Nail
1


Background
In order to promote union, an arthrodesis device must provide compression as well as stability as described by AO/ASIF principles. Compression induces stability and promotes union by maximizing bone-to-bone contact across the fusion site. Compression also helps promote primary bone growth by preventing excessive micro-motion of the joint, which is necessary for proper fusion. In addition, the device must provide rigidity to the fusion site to prevent excessive bending and torsional motions of the TTC complex. 14,27,29 Several methods of arthrodesis have been proposed since its inception in 1879; however, the most common methods are external fixation and intramedullary (IM) nailing. External fixators are ring frames placed around the foot and ankle with tensioned wires passing through the bones and allow for compression to be adjusted over time.7 IM nails are stainless steel or titanium nails inserted through the calcaneus into the tibia to provide rigidity to the ankle during fusion. IM nails have become more popular as they permit earlier weight bearing after surgery and have fewer complications.17 Current Technology
Recent advancements in IM nailing have focused on providing initial compression to the fusion site and providing the option for dynamization. A recent review by Yakacki et al.32 detailed the development of titanium-based IM nails. First generation IM nails require the surgeon to manually compress the joint by hammering a strike plate on the installation hardware, while second generation nails offer a method to apply compression externally by using compression rods installed through the proximal portion of the tibia. The latest 3rd generation of IM nails have incorporated internal compression mechanisms to help maintain compression during installation. These devices generate compression via an internal screw that axially presses upon the distal locking screws. Most IM nails offer a dynamized option in which a locked tibial screw is removed, leaving a tibial screw inserted through a slotted hole. Dynamization promotes compression from weight bearing, eliminates stress shielding, and is often utilized in cases of delayed healing.110,26
2


First generation Second generation Third generation Pseudoejgstic Dyngngil
Yakacki et at. 20 /1
Figure 2: Intramedullary nail types by generation (Yakacki)
Research Goals
Previous studies and development efforts have mainly focused on the stiffness of the TTC-nail construct21118'2"22 or compression generated from the installation of the IM nail;3931 however, to the best of the authors' knowledge, no studies have investigated how IM nails affect the distribution of weight-bearing load across the fusion site. Some reports have stated IM nails create a load-sharing construct;4 however, there has been no systematic investigation between load transfer and IM nailing techniques. Therefore, the purpose of this study was to model and test the load-sharing characteristics between the TTC joint and IM nail for different methods of IM nailing. Static methods were evaluated using a titanium nail, while a pseudo-elastic method was evaluated using a newly FDA-cleared pseudo-elastic nail, which utilizes Nickel-Titanium (NiTi) to maintain axial compression. To model a carbon fiber based nail, the titanium model was used with a modified modulus to match carbon-fiber reinforced PEEK material properties.
3


CHAPTER II
METHODS
Pseudo-elastic Nail Model
A model for an implanted pseudo-elastic ankle arthrodesis nail (AAN) was created based on the steps described by Ryan Anderson et al.46 Models were first imported into ScanIP+FE+CAD (Simpleware LTD). The model was created using a QCT scan of healthy leg bones and a CAD based model of a pseudo-elastic nail provided by the manufacturer (MedShape).The healthy tibia and talus had portions of the bone removed, cut, and mated in the model as they would be for an ankle fusion procedure. The pseudo-elastic and its associated components were oriented as it would be after a successful surgery. The tibia, talus, calcaneus, and pseudo-elastic nail were all given individual masks in the ScanIP software. Any floating elements were cleaned up using the island removal tool. Once floating elements were removed, small gaps in the model were closed using the cavity fill tool. The model was oriented such that the z-axis is pointing down the length of the leg. Upon completion of model clean up the model is ready for meshing and then exportation.
Figure 3: ScanIP model comparison for the Pseudo-elastic nail masks (Right) and the unmodified
bone model masks (Le ft)
4


In ScanIPs model configuration module is used to prepare the model for meshing. The masks were smoothed using greyscale values and exported with length units in millimeters. The model configuration was given a mesh coarseness of -25. A Convergence study by Ryan Anderson46 showed that this was an appropriate mesh coarseness to get a sufficient number of elements without over saturating the models. After coarseness, the material properties of each component needed to be defined. Mass density of the pseudo-elastic nail components does not affect the results of FEA that will be conducted on the model. All of the components of the pseudo-elastic nail, except for the compressive element, were assigned the material properties of titanium, Youngs modulus of l.lOxlO5 MPa and a Poissons ratio of 0.3. Since Abaqus operates without units it is important to make sure Youngs modulus is exported from ScanIP in the proper units. For each bone mask a greyscale based material property definition was used. Mass density of the bones were calculated by ScanIP based on greyscale brightness of the QCT scan and Youngs modulus was calculated based on the determined density. Equations for the mass density were the same as described by Keyak et al. (Keyak paper). Once again it was important to make sure that the equations matched the units being used in ScanIP, millimeters and megapascals, before exporting to Abaqus. In order to prevent zero or negative modulus for portions of bone with very faint brightness a blanket material property of IMPa was used for all very low brightness values. A Poissons ratio of 0.3 was assigned for all bones. After material properties were defined, contact sets were defined between the pseudo-elastic nail jacket and each bone. These contact sets allow for the bone and nail body to move relative to one another, rather than being bonded together. The model is finally meshed using the full model tool and exported as an .inp file that can be used in Abaqus software.
The model is then imported into Abaqus software. The first step in Abaqus is to remove the locking manually remove the locking material elements and replace them with a new pair of elements that will connect the compressive NiTi element to the nail body or jacket. To do this a group of nodes on the compressive element and a group of three nodes on the jacket are selected. Then, the .inp needs to be edited by deleting every element from the locking material mask except for the first two. Those
5


two elements will have their nodes replaced with the nodes chosen above. The next step, once the locking material has been added is to update the material properties of the compressive element to those of Nickel Titanium. The nickel titanium material is assigned the name ABQSUPERELASTICN3D and the associated material definition as described by Ryan Anderson46. The necessary mechanical constants for the pseudo-elastic material are described in the UMAT literature built into Abaqus. Next, the sliding element of the pseudo-elastic nail is pinned in place, and then the NiTi rod is stretched eight (8) millimeters in the z-direction. During this step the interaction between the bones, pseudo-elastic nail, and the locking material are turned off. In the next step the displacement of the NiTi rod is reduced to six (6) millimeters. This stretch followed by relaxation step programs the NiTi rod to maintain a level of nearly constant compression. After the programming steps, the bones and locking material are reactivated and then the proximal end of the tibia is pinned in place. After this, the NiTi rod is released by disabling the prescribed displacement from the first step. The locking material, after reactivation, holds the NiTi rod in a stretched state. After the NiTi rod is programmed and released then the gait load is applied. A peak gait load of 112 IN is divided amongst 4 points, 280.25 per node, at the distal end of the Calcaneus surrounding the opening where the AAN is implanted. 1121 N was chosen to represent the maximum ground reaction force, 1.1 Body Weights (Sacco), during gait for a 104kg person. The applied load was also applied exclusively in the z-direction because it is the direction of highest stress during gait. Once the model is completely set up then it is passed to the job module and submitted for analysis.
6


Figure 4: Load on Heel During Gait in Percent Body Weights (Sacco)
Figure 5 Gait Load Force Sites and Magnitude
Static Nail Model
For the static nail models a physical titanium static nail was measured using calipers and created in Solidworks. After the nail body was created, as set of screws were created and mated to the
7


nail body. Diameters of the screws were determined based on the diameter of the anchoring and guide holes in the titanium nail body and lengths were assigned based on the nail lengths from the pseudoelastic nail model. The nail model was then exported from Solidworks and imported into ScanIP+FE+CAD with the same leg bones as the pseudo-elastic model. The pseudo-elastic masks were turned off and the titanium surfaces were oriented in the same way and converted into masks. For model configuration the same settings were used with the lack of locking material and compressive element. The nail jacket is assigned its own material name with the material properties of Ti 6AL-4V and the screws are assigned the same properties with a separate name. This differentiation is important for editing in Abaqus. Contact sets and bone material properties are assigned the same as they were for the NiTi model. After model configuration is complete, the model is export as an .inp file for Abaqus.
Figure 6 OCT Scan and Solidworks Model Combined in ScanIP Once in Abaqus the material for the nail body needs to be edited. The material needs to have an orthotropic expansion coefficient added, for this model an expansion coefficient of a3=0.0005 is used. In addition to an expansion coefficient, the material has a specific heat of 1 and conductivity of 100 added to the material definition. The material definition additions allow for the nail body to contract when cooled and expand when heated up. For a carbon-fiber nail, the material definition is
8


altered even further by changing the elastic material properties to reflect a modulus of 75GPa, this value was chosen based on material properties listed in the documentation for the carbon-fiber AAN. The elements of the nail body then need to be selected and have their element type changed to the coupled temperature displacement, C3D4T, element type. After this, the same elements have their material orientation defined the same as the global coordinate system. For Abaqus steps, one step needs to be created to pin the most proximal end of the tibia in place. Once the tibia is pinned in place the nail body needs to have the temperature reduced by one, this is done by setting a boundary condition of temperature -1 for a selected region of nail body elements. Since Abaqus operates on a unitless system of measurement, the -1 temperature change triggers the expansion response of the nail body without affecting any other parts of the model. No other elements will be affected by the temperature change. The final step is to apply a gait load with the same amplitude, magnitude, and location as used in the NiTi model. The model is then passed to the job module and submitted. Resorption Models
For the resorption models of both the NiTi and Static Nail the steps within Abaqus will be exactly the same. The only changes will be present in the steps concerning ScanIP. For each model the most proximal slices of the talus, slices 300-309 totaling approximately 3mm in thickness, will have material properties changed from a greyscale based modulus to more compliant material properties. To do this the slices are copied to a new mask and the new mask given the new material properties, IMPa Youngs modulus and 0.01 Poissons ratio for example. Those slices are then removed from the talus. The model is then exported to Abaqus and run as described above.
9


Figure 7 SccmlP Model with the Addition of a Resorption Zone Analyzing the Results
Once the job module is finished in Abaqus the results can be viewed by right clicking on the completed job and selecting the results option. Once in the results viewport the tibia, nail body, and compressive element (when applicable) need to be selected by using the display group manager. After isolating one group of elements, the tibia for example, a view cut needs to be made on the z-axis to observe the force in the tibia. The cut should be made at the same slice for each group of elements, for this study slice 250 was used for each material. This slice is located on the distal end of the most proximal AAN screws.
Figure 8 Reaction Force of Bone in Z-direction at Slice 250
10


After choosing the view cut the reaction force values need to be displayed. This is done by enabling the free body slice check box and going into the options to disable heat flow display along with moments and enabling only the z direction component of force. After this is done the reaction for data can be exported using the XYData manager. To do this a new set of XYData needs to be created from the free body data. In the XYData creation menu it is important to double-check that the z-direction component is checked. The data can then be exported from Abaqus and then imported into Microsoft Excel, iGor, or MATLAB for visual representation. The data is exported with reaction force in the z-direction at each time step. When comparing each nail configuration the force in the z direction is most closely related to the force experienced from gait loading. For each configuration it is important to note the load in each material at the start of gait load or swing phase, where the gait load is zero. The next important data point is the load in each material at max gait load or toe off. For comparison of each configuration with a localized resorption zone, gait load is not as important as the amount of stress at zero load. The load in each nail jacket, tibia, and compressive element, where applicable, should be noted in the models with no resorption zone and then noted in the models with a resorption zone. Even more data is acquired by incrementally raising and lowering the modulus of the resorption zone.
11


CHAPTER III
RESULTS
Pseudo-elastic Nail with Gait
The pseudo-elastic Pseudo-elastic maintained compression in the NiTi element at a 51544 N throughout the gait cycle. At zero gait load, also known as swing phase, the tibia had a higher force value load then the nail jacket, -364 N compared to -156 N. Throughout the gait cycle the nail body, or jacket, only varied from -156 N to -534 N at toe off. The tibia by comparison had a minimum load of -364 N and rose as high as -1116 N at toe off.
Figure 9: Force Distribution during Gait Load for a Pseudo-elastic Nail Model Titanium Nail with Gait
The static titanium nail was given a prescribed contraction of 0.5%. This contraction provided a compression of about 770 N which was equal and opposite between the tibia and nail body during swing phase, zero gait load. Once gait load was applied, the tibia had force values that ranged from -
12


770 N to -965 N. The nail jacket had force values that ranged from +770 N to -156 N. This range means that the tibia only absorbed 195 N of the maximum gait load and the nail body absorbed 926 N of the gait load.
Static Titanium Nail Force Distribution
Figure 10: Force Distribution during Gait Load for a Conventional Static Titanium Ankle
Arthrodesis Nail
Carbon Fiber Nail with Gait
The carbon-fiber nail had a similar graph shape compared to the conventional titanium static nail. With an expansion coefficient of 0.5% the nail body generated 700 N of compression. In order to achieve the same compression value as the titanium nail an expansion coefficient of 0.55% needed to be applied. Similar to the conventional titanium nail, the carbon fiber nail body absorbed the bulk of the gait load. The nail body had force values ranging from 705 N during swing phase to -172 N during toe off. The tibia in this scenario had force values ranging from -705 N to -952 N during toe off, a difference of only -247 N. The tibia only experienced 22% of the gait load.
13


Static Carbon-fiber Nail Force Distribution
Figure 11: Force Distribution during Gait Load for a Conventional Static Carbon Fiber Body Ankle
Arthrodesis Nail
Gait-Load Sharing Characteristics of Each Nail
The load sharing of each nail configuration, shown in the figure below, shows that the pseudo-elastic, which has an internal compressive element allow for more of the gait load to be transferred to the bone as it would without a medical device rather than be absorbed by the device. The material used to manufacture a static nail without internal compression did very little to affect the load sharing characteristics. A static nail with a lower modulus, the carbon fiber based nail, allowed be more load to be transferred to the bone but not more than would be experience by the nail body.
14


Load Sharing Properties of Ankle Arthrodesis Nails
DynaNail (Toe Off) Versanail (Toe Off) Carbofix (Toe Off)
Tibia Nail Body
Figure 12: Load Sharing of Different Ankle Arthrodesis Nails at Peak Gait Load
Compression Loss with Different Resorption Values
Once a resorption zone was added to each configuration each nail lost some compression.
The pseudo-elastic nail model was able to maintain more than 90% of compression even with a resorption zone with a very low modulus, IMPa. Both static nail configurations lost the majority of the compression generated by the nail, more than 80%, when a low modulus resorption zone is added. Both the carbon fiber and the titanium static nails maintained compression until the modulus of the resorption zone dropped below lOOOMPa. Both the static nails shared similar curve shapes. Both of the static nails were prescribed the same strain of 0.05% to generate the forces shown at the different resorption zone moduli. The titanium static nail dropped to 109 N when given a resorption zone with a modulus of 1 MPa. The carbon fiber nail dropped to a similar 112 N when exposed to a resorption zone with the same modulus, 1 MPa. Under healthy bone conditions, the static titanium nail had a compression value 770 N which was similar to the values experienced when the resorption zone was assigned a modulus of 10,000 N.
15


Effects of a Resorption Zone on Ankle Arthrodesis Nail Compression
Figure 13: Compression Values of Different Ankle Arthrodesis Nails when Exposed to a Specified
Resorption Zone Modtdus
16


CHAPTER IV
DISCUSSION
Ankle Arthrodesis Nails Under Gait Uoad
Results showed that the pseudo-elastic nail was able to maintain nearly constant compression of the pseudo-elastic NiTi element throughout the gait cycle. The pseudo-elastic nail also transferred 66% of the gait load to the tibia. The remaining load was split between the nail body, 33%, and the compressive element, less than 1%. In contrast to the pseudo-elastic nail model, both static nail configurations transferred most of their load, 83% for the titanium nail and 78% for the carbon-fiber nail, to the nail jacket. The remaining force from gait loading, in both static nail models, was transferred to the tibia and other bones. According to Wolff s law47, bone growth is promoted by cyclical compression and tension of the bone. The load sharing properties in Figure 12 indicate that the pseudo-elastic nail is better at providing the needed cyclical loading of the tibia after installation. The static nails on the other hand do not allow most of the load to transfer to the bone.
Ankle Arthrodesis Nails with a Resorption Zone
IM nail design and materials played a significant role in maintaining compression and load sharing. The pseudoelastic nail had the lowest degree of stress-shielding (32%) and maintained compression for over 0.10 mm of simulated resorption. Constant compression and the avoidance of resorption gapping is paramount to drive primary bone healing in joint fusions due to lack of periosteal/endosteal anatomy crossing the fusion site, thus crippling the ability for secondary bone healing (callus healing). This model allows for direct comparison between devices and can be used pre-operatively to predict patient-specific performance and help aid in device selection for TTC fusion.
Uimitations
The aims of this study were to analyze different ankle arthrodesis nails under a variety of conditions. The first aim was to study each nail under gait loading conditions. The second was to analyze the nails when exposed to a localized resorption zone. For each of these aims a healthy leg
17


QCT scan was used. The osteotomy was done by cutting the bones from the QCT scan within the SCANIP software. This allowed for an osteotomy where the bones were all pressed directly against one another without any shifting or compaction required. This also allowed for an idealized resorption zone where the area of resorption was able to be chosen within the software and the modulus of the zone adjusted as needed. The contact between the bones and the nail jacket was also set to frictionless and there were no gaps allowed between the bones and the device.
Future Work
For future advancement of this study, there are several avenues of advancement and improvement. There should be a continuation of this study using QCT scans of several different legs. It would also benefit from scans of unhealthy legs and legs of varying ages. There is also room for comparison of the results from this experiment to values in a compression test of a cadaver leg with one of the medical devices installed. This would allow us to determine the validity of our results.
After the numbers from this experiment are confirmed, it would be useful to create a model of the static nails with their respective dynamization slots activated. In practice, this would require another surgery and is usually reserved for post union cases or as a last resort in non-union, to reduce the stress shielding effects of the device. Confirming or disproving these claims will help to determine the load sharing effects of the dynamization slots.
Conclusions
1. The pseudo-elastic nail allowed for 67% of the load experienced from the cycle to be transferred to the tibia. The static carbon-fiber and titanium nails allowed only 22% and 17% of the load to transfer, respectively.
2. The pseudo-elastic nail maintained compression, within 10% of original values, when subjected to the resorption zone. Both static nails lost more than 80% of original compression when subjected to the resorption zone.
18


REFERENCES
1. Basumallick, M. N., and Bandopadhyay, A.: Effect of dynamization in open interlocking nailing of femoral fractures. A prospective randomized comparative study of 50 cases with a 2-year follow-up. Ac to Orthop Belg, 68(1): 42-8, 2002.
2. Bennett, G. L.; Cameron, B.; Njus, G.; Saunders, M.; and Kay, D. B.: Tibiotalocalcaneal arthrodesis: a biomechanical assessment of stability. Foot Ankle Int, 26(7): 530-6, 2005.
3. Berson, L.; McGarvey, W. C.; and Clanton, T. O.: Evaluation of compression in intramedullary hindfoot arthrodesis. Foot Ankle Int, 23(11): 992-5, 2002.
4. Bevemage, B. D.; Deleu, P. A.; Maldague, P.; and Leemrijse, T.: Technique and early experience with posterior arthroscopic tibiotalocalcaneal arthrodesis. Orthop Traumatol Surg Res, 96(4): 469-75,2010.
5. Boer, R.; Mader, K.; Pennig, D.; and Verheyen, C. C.: Tibiotalocalcaneal arthrodesis using a reamed retrograde locking nail. Clin Orthop Relat Res, 463: 151-6, 2007.
6. Brumback, R. J.: The rationales of interlocking nailing of the femur, tibia, and humerus. Clin Orthop Relat Res, (324): 292-320, 1996.
7. Chamley, J.: Compression arthrodesis of the ankle and shoulder. J Bone Joint Surg, 33B: 180-91, 1951.
8. Conway, J. D.: Charcot salvage of the foot and ankle using external fixation. Foot Ankle Clin, 13(1): 157-73,2008.
9. Eichhom, S.; Lindner, T.; M,ckley, T.; Trapp, O.; and Steinhauser, E.: The loss of compression in intramedullary ankle Arthrodesis using two different types of compression rods A biomechanical study. Journal of Biomechanics, 39(Supplement 1): S136-S136, 2006.
10. Fazal, M. A.; Garrido, E.; and Williams, R. L.: Tibio-talo-calaneal arthrodesis by retrograde intramedullary nail and bone grafting. Foot and Ankle Surgery, 12(4): 185-190, 2006.
11. Fragomen, A. T.; Meyers, K. N.; Davis, N.; Shu, EL; Wright, T.; and Rozbruch, S. R.: A biomechanical comparison of micromotion after ankle fusion using 2 fixation techniques: intramedullary arthrodesis nail or Ilizarov external fixator. Foot Ankle Int, 29(3): 334-41, 2008.
12. Frey, C.; Halikus, N. M.; Vu-Rose, T.; and Ebramzadeh, E.: A review of ankle arthrodesis: predisposing factors to nonunion. Foot Ankle Int, 15(11): 581-4, 1994.
13. Gil, F. J., and Planell, J. A.: Shape memory alloys for medical applications. Proc InstMech EngH, 212(6): 473-88, 1998.
14. Gilbert, J. A.; Dahners, L. E.; and Atkinson, M. A.: The effect of external fixation stiffness on early healing of transverse osteotomies. J Orthop Res, 7(3): 389-97, 1989.
15. Haaker, R.; Kohja, E. Y.; Wojciechowski, M.; and Gruber, G.: Tibio-talo-calcaneal arthrodesis by a retrograde intramedullary nail. Ortop Traumatol Rehahil, 12(3): 245-9, 2010.
19


16. iDataResearchlnc.: US Markets for Orthopedic Trauma Devices. 2010.
17. Kim, C.; Catanzariti, A. R.; and Mendicino, R. W.: Tibiotalocalcaneal arthrodesis for salvage of severe ankle degeneration. Clin Podiatr Med Surg, 26(2): 283-302, 2009.
18. Lee, A. T.; Sundberg, E. B.; Lindsey, D. P.; Harris, A. H.; and Chou, L. B.: Biomechanical comparison of blade plate and intramedullary nail fixation for tibiocalcaneal arthrodesis. Foot Ankle Int, 31(2): 164-71,2010.
19. Means, K. R.; Parks, B. G.; Nguyen, A.; and Schon, L. C.: Intramedullary nail fixation with posterior-to-anterior compared to transverse distal screw placement for tibiotalocalcaneal arthrodesis: a biomechanical investigation. Foot Ankle Int, 27(12): 1137-42, 2006.
20. Miickley, T.; Eichom, S.; Hoffmeier, K.; von Oldenburg, G.; Speitling, A.; Hoffmann, G. O.; and Buhren, V.: Biomechanical evaluation of primary stiffness of tibiotalocalcaneal fusion with intramedullary nails. Foot Ankle Int, 28(2): 224-31, 2007.
21. Miickley, T. M.; Eichom, S.; von Oldenburg, G.; Speitling, A.; DiCicco, J. D., 3rd; Hofmann, G. O.; and Buhren, V.: Biomechanical evaluation of primary stiffness oftibiotalar arthrodesis with an intramedullary compression nail and four other fixation devices. Foot Ankle Int, 27(10): 814-20, 2006.
22. Mueckley, T. M.; Eichom, S.; von Oldenburg, G.; Speitling, A.; DiCicco, J. D., 3rd; Hofmann, G. O.; and Buhren, V.: Biomechanical evaluation of primary stiffness of tibiotalar arthrodesis with an intramedullary compression nail and four other fixation devices. Foot Ankle Int, 27(10): 814-20, 2006.
23. Nihal, A.; Gellman, R. E.; Embil, J. M.; and Trepman, E.: Ankle arthrodesis. Foot Ankle Surg, 14(1): 1-10, 2008.
24. Perlman, M. H., and Thordarson, D. B.: Ankle fusion in a high risk population: an assessment of nonunion risk factors. Foot Ankle Int, 20(8): 491-6, 1999.
25. Plaass, C.; Knupp, M.; Barg, A.; and Hintermann, B.: Anterior double plating for rigid fixation of isolated tibiotalar arthrodesis. Foot Ankle Int, 30(7): 631-9, 2009.
26. Sharma, D. K.; Yunas, I.; and Ramos, J.: Management of complications of tibiotalocalcaneal arthrodesis in patients with co-morbidity. The Foot, 16(1): 33-37, 2006.
27. Teijesen, T., and Apalset, K.: The influence of different degrees of stiffness of fixation plates on experimental bone healing. JOrthop Res, 6(2): 293-9, 1988.
28. Vesely, R.; Prochazka, V.; Visna, P.; Valentova, J.; and Savolt, J.: [Tibiotalocalcaneal arthrodesis using a retrograde nail locked in the sagittal plane]. Acta Chir Orthop Traumatol Cech, 75(2): 129-33, 2008.
29. Wu, J. J.; Shyr, H. S.; Chao, E. Y.; and Kelly, P. J.: Comparison of osteotomy healing under external fixation devices with different stiffness characteristics. J Bone Joint Surg Am, 66(8): 1258-64, 1984.
30. Yakacki, C. M.; Gall, K.; Dirschl, D. R.; and Pacaccio, D. J.: Pseudo-elastic intramedullary nailing for tibio-talo-calcaneal arthrodesis. Expert Rev Med Devices, 8(2): 159-66, 2011.
20


31. Yakacki, C. M.; Khalil, H. F.; Dixon, S. A.; Gall, K.; and Pacaccio, D. J.: Compression forces of internal and external ankle fixation devices with simulated bone resorption. Foot Ankle Int, 31(1): 76-85,2010.
32. Yakacki, Christopher M, etal. "Pseudoelastic intramedullary nailing for tibio-talo-calcaneal arthrodesis." Expert Review of Medical Devices 8.2 (2011): 159+. Academic OneFile. Web. 2 May 2016
33. Rho, Jae Young, Richard B. Ashman, and Charles H. Turner. "Young's modulus of trabecular and cortical bone material: ultrasonic and microtensile measurements." Journal of biomechanics 26.2 (1993): 111-119.
34. Sacco, Isabel CN, Paula MH Akashi, and Ewald M Hennig. A Comparison of Lower Limb EMG and Ground Reaction Forces between Barefoot and Shod Gait in Participants with Diabetic Neuropathic and Healthy Controls. BMC Musculoskeletal Disorders 11 (2010):
24. PMC. Web. 2 May 2016.
35. Hsu, Andrew R., J. Kent Ellington, and Samuel B. Adams. "Tibiotalocalcaneal Arthrodesis Using a Nitinol Intramedullary Hindfoot Nail." Foot & ankle specialist (2015): 1938640015598838.
36. Anderson, Ryan, et al. Finite Element Analysis of a Pseudoelastic Compression-Generating Intramedullary Ankle Arthrodesis Nail. Journal of the Mechanical Behavior of Biomedical Materials (2016)
37. Taylor, James et al. Tibiotalocalcaneal Arthrodesis Nails: A Comparison of Nails With and Without Internal Compression Foot Ankle Int March 2016 37: 294-299, first published on October 15, 2015 doi: 10.1177/1071100715611891
38. Yakacki, Christopher M., et al. "Compression forces of internal and external ankle fixation devices with simulated bone resorption." Foot & ankle international 31.1 (2010): 76-85.
39. ASM Material Data Sheet Titanium Ti-6A1-4V (Grade 5), Annealed http ://asm.matweb .com/search/SpecificMaterial .asp?bassnum=MTP641
40. L. Fortington and J. Geertzen, "Short and Long Term Mortality Rates after a Lower Limb Amputation", Sciencedirect.com, 2016. [Online]. Available:
http://www.sciencedirect.com/science/article/pii/S1078588413002256. [Accessed: 25- Nov-2016],
41. M. Erduran, "Biomechanical effects of the distance from the fracture zone to the interlocking fixation screw of intramedullary nail", Jbiomech.com, 2016. [Online]. Available: http://www.jbiomech.com/article/S0021-9290(l 1)00119-9/abstract. [Accessed: 25- Nov-2016],
42. A. Nihal, "Ankle Arthrodesis", Sciencedirect.com, 2016. [Online]. Available: http://www.sciencedirect.com/science/article/pii/S1268773107000744. [Accessed: 25- Nov-2016],
43. C. Yakacki, K. Gall, D. Dirschl and D. Pacaccio, "Pseudoelastic intramedullary nailing for tibio-talo-calcaneal arthrodesis", Tandfonline.com, 2016. [Online]. Available: http://www.tandfonline.com/doi/abs/10.1586/erd.10.93. [Accessed: 25-Nov-2016],
21


44. T2 Ankle Arthrodesis Nail | Stryker Internet: https://patients.stryker.com/assets/ankle_fusion-
97bd33028a61e413fcfall9bb288b664902eef5dff0ab03cl75081f786922f6f.jpg [Accessed 25-Nov-2016]
45. J. H. Keyak, T. S. Kaneko, J. Tehranzadeh, H. B. Skinner, Predicting proximal femoral strength using structural engineering models Clinical orthopaedics and related research, 219 ,August 2005. [Accessed 25-Nov-2016] Online:
http://joumals.lww.com/corr/Abstract/2005/10000/Predicting_the_Strength_of_Femoral_Sha fts_with_and. 3 0. aspx
46. Ryan T. Anderson, Douglas J. Pacaccio, Christopher M. Yakacki, R. Dana Carpenter, Finite element analysis of a pseudoelastic compression-generating intramedullary ankle arthrodesis nail Journal of the Mechanical Behavior ofMiomedical Materials, September 2016 Volume 62 Pages 83-92 [Online]
http ://www.sciencedirect.com/science/article/pii/S 1751616116301102
47. Wolff, Julius. The law of hone remodelling. Springer Science & Business Media, 2012.
22


Full Text

PAGE 1

ESTABLISHING METRICS FOR COMPARISON OF ANKLE FUSION DEVICES: PATIENT SPECIFIC FINITE ELEMENT ANAL S YIS OF THREE INTRAMEDULLARY NAILS FOR TIBIOTALOCALCANEAL FUSION by PATRICK K. TERRILL B.S., Virginia Polytechnic Institute and State University, 2011 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Mechanical E ngineering Program 2017

PAGE 2

ii 2017 PATRICK K. TERRILL ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Science in Mechanical Engineering degree by Patrick K. Terrill has been approved for the Mechanical Engineering Program by R. Dana Carpenter, Chair Christopher M. Yakacki Kai Yu Date: April 28 2017

PAGE 4

iv Terrill, Patrick K. ( M S Mechanical Engineering Program) Establishing Metrics for Comparison of Ankle Fusion Devices: Patient Specific Finite Element Analysis of Three Intramedullary Nails for Tibiotalocalcaneal Fusion Thesis directed by Professor R. Dana Carpenter ABSTRACT Tibiotalocalcaneal (TTC) arthrodesis is a salvage procedure for patients with severe osteoarthritis and other severe ankle conditions. The procedure is done by inserting an intramedullary (IM) nail through bottom of the calcaneus and up through the talus and the tibia. The nail is fixed with several screws and compression is applied by the nail and screws to the three bones in order to promote fusion. The purpose of this study is to compare dif ferent IM nails in a simulated ankle fusion using Finite Element Analysis (FEA). A pseudo elastic, conventional ti tanium, and conventional carbon fiber nail were all tested for load sharing between the nail body and the tibia under gait loading and compres sion with a resorption zone A model of the bone s was generated from a quantitative computed tomography (QCT) scan of a healthy leg. Pseudo elastic compression was simulated by pre stretching th e pseudo elastic compressive element in the by itself and then inserting it into the nail jacket. The conventional static nails were compressed by giving the nail jacket an orthotropic thermal expansion coefficient and reducing the temperature in the model. After comp ression each nail was subjected to an applied gait load. In the p seudo elastic model, the tibia was subjected to 68% of the peak load during the gait cycle. The t itanium and c arbon fiber nail models had tibia load values of 23% and 28% respectively. The resorption for each model was ratio of 0.01. For the static nails the amount of compression provided by the nail drops more than 80% when exposed to a prescribed resorption zone The pseudo elastic nail maintains compression when expo sed to resorption up to 4% of the locked strain val ue The form and content of this abstract are approved. I recommend its publi cation. Approved: R. Dana Carpenter

PAGE 5

v ACKNOWLEDGEMENTS I would like to thank my advisor Prof. Dana Carpenter for his continued encouragement and assistance throughout my research. His background knowledge of the subject matter and the associated software was critical for helping me complete my thesis in a timely manner. He also has excellent taste in movies. I would like to thank my other advisor, Prof. Chris Yakacki for his assistance throughout my research. He has a wealth of knowledge on the subject matter and the real world applications, problems, and needs in the field of a nkle a rthrodesis n ails. His previo us research helped direct mine to the topic I have been working on. His extensive knowledge of movie quotes has also b een there to lighten the mood a t all time s I would like to thank the members of the Smart Materials and Biomechanics Laboratory (SMAB LAB ) at CU Denver for their support during my studies and research. Sam Mills, Nick Traugutt, Mo hand Saed, Ross Volpe, Hannah Tif ft. I would especially like to thank Ravi Patel and Ryan Anderson for their previous work on my research topic and their help with learning the software required and subject matter. I would finally like to thank my family. My parents, Cathy and Allen Terri ll, and my sister, Samantha, supported and encouraged me when I decided to quit my job and pursue my graduate degree.

PAGE 6

vi TABLE OF CONTENTS I. Introduction ................................ ................................ ................................ ................................ ......... 1 Ankle Fusion Impact ................................ ................................ ................................ .......................... 1 Background ................................ ................................ ................................ ................................ ........ 2 Current Technology ................................ ................................ ................................ ............................ 2 Research Goals ................................ ................................ ................................ ................................ ... 3 II. Methods ................................ ................................ ................................ ................................ ............. 4 Pseudo elastic Nail Model ................................ ................................ ................................ .................. 4 Static Nail Model ................................ ................................ ................................ ................................ 7 Resorption Models ................................ ................................ ................................ ............................. 9 Analyzing the Results ................................ ................................ ................................ ....................... 10 III. Results ................................ ................................ ................................ ................................ ............ 12 Pseudo elastic Nail with Gait ................................ ................................ ................................ ........... 12 Titanium Nail with Gait ................................ ................................ ................................ .................... 12 Carbon Fiber Nail with Gait ................................ ................................ ................................ ............. 13 Gait Load Sharing Characteristics of Each Nail ................................ ................................ .............. 14 Compres sion Loss with Different Resorption Values ................................ ................................ ...... 15 IV. Discussion ................................ ................................ ................................ ................................ ...... 17 Ankle Arthrodesis Nails Under Gait Load ................................ ................................ ....................... 17 Ankle Arthrodesis Nails with a Resorption Zone ................................ ................................ ............. 17

PAGE 7

vii LIST OF FIGURES Figure 1: An Ankle Arthrodesis Nail ................................ ................................ ........................ 1 Figure 2: Intramedullary nail types by generation (Yakacki) ................................ ................... 3 Figure 3: ScanIP model comparison for the Pseudo elastic nail masks (Right) and the unmodified bone model masks (Left) ................................ ................................ ..... 4 Figure 4: Load on Heel During Gait in Percent Body Weights (Sacco) ................................ ... 7 Figure 5 Gait Load Force Sites and Magnitude ................................ ................................ ........ 7 Figure 6 QCT Scan and Sol idworks Model Combined in ScanIP ................................ ............ 8 Figure 7 ScanIP Model with the Addition of a Resorption Zone ................................ ........... 10 Figure 8 Reaction Force of Bone in Z direction at Slice 250 ................................ ................. 10 Figure 9 : Force Distribution during Gait Load for a Pseudo elastic Nail Model ................... 12 Figure 10: Force Distribution during Gait Load for a Conv entional Static Titanium Ankle Arthrodesis Nail ................................ ................................ ................................ .... 13 Figure 11: Force Distribution during Gait Load for a Conventional Static Carbon Fiber Body Ankle Arthrodesis Nail ................................ ................................ ......................... 14 Figure 12: Load Sharing of Different Ankle Arthrodesis Nails at Peak Gait Load ................ 15 Figure 13: Compression Values of Different Ankle Arthrodesis Nails when Exposed to a Specified Resorption Zone Modulus ................................ ................................ .... 16

PAGE 8

1 CHAPTER 1 INTRODUCTION Ankle F usion I mpact Tibio talo calcaneal (TTC) arthrodesis is a surgical salvage procedure performed to treat ankle pain and trauma related to pathologies such as osteoarthritis, rheumatoid arthritis, neuropathic arthropathy, talar osteonecrosis, osteomyelitis, failed total ankle replacement and other diseases of the ankle. 7 20 TTC arthrodesis is often performed as an alternative to amputation. The purpose of the procedure is to provide a stable, pain free union of the bones of the hindfoot. Reviews of surgical procedures have found the unio n rate of TTC arthrodesis procedures to be as high as 94.6% (195 fusions of 206 procedures); however, the occurrence of a non union can be seen in up to 40% of procedures involving complex cases of ankle disease. 5 8 11 13 15 16 21 24 25 28 In a cost analysis of ankle procedures, tibio talo fusion alone is seen to cost as much as $32,683 for a single non arthroscopic procedure. 23 Figure 1 : An Ankle Arthrodesis Nail

PAGE 9

2 Background In order to promote union, an arthrodesis device must provide compression as well as stability as described by AO/ASIF principles. Compression induces stability and promotes union by maximiz ing bone to bone contact across the fusion site. Compression also helps promote primary bone growth by preventing excessive micro motion of the joint, which is necessary for proper fusion. In addition, the device must provide rigidity to the fusion site to prevent excessive bending and torsional motions of the TTC complex. 14 27 29 Several methods of arthrodesis have been proposed since its inception in 1879; however, the most common methods are external fixation and intramedullary (IM) nailing. External fixators are r ing frames placed around the foot and ankle with tensioned wires passing through the bones and allow for compression to be adjusted over time. 7 IM nails are stainless steel or titanium nails inserted through the calcaneus into the tibia to pr ovide rigidity to the ankle during fusion. IM nails have become more popular as they permit earlier weight bearing after surgery and have fewer complications. 17 Current Technology Recent advancements in IM nailing have focused on providing initial compression to the fusion site and providing the option for dynamization. A recent review by Yakacki et al. 32 detailed the development of titanium based IM nails. First generation IM nails require the surgeon to manually compress the joint by hammering a strike plate on the installation hardware, while second gener ation nails offer a method to apply compression externally by using compression rods installed through the proximal portion of the tibia. The latest 3 rd generation of IM nails have incorporated internal compression mechanisms to help maintain compression during installation. These devices generate compression via an internal screw that axially presses upon the distal locking screws. Most IM nails through a slotted hole. Dynamization promotes compression from weight bearing, eliminates stress shielding, and is often utilized in cases of delayed healing. 1 10 26

PAGE 10

3 Figure 2 : Intramedullary nail types by generation (Yakacki) Research Goals Previous studies and development efforts have mainly focused on the stiffness of the TTC nail construct 2 11 18 20 22 or compression generated from the installation of the IM nail; 3 9 31 ho wever, to of weight bearing load across the fusion site. Some reports have stated IM nails create a load sharing construct; 4 however, there has been no systematic investigation b etween load transfer and IM nailing techniques. Therefore, the purpose of this study was to model and test the load sharing characteristics between the TTC joint and IM nail for different methods of IM nailing. Static methods were evaluated using a t itanium nail while a pseudo elastic method was evaluated usi ng a newly FDA cleared pseudo elastic nail which utilizes Nickel Titanium (NiTi) to maintain axial compression. To model a carbon fiber based nail, the t itanium model was used with a modified modulus to match carbon fiber reinforced PEEK material properties

PAGE 11

4 CHAPTER II METHODS Pseudo elastic Nail Model A model for an implanted pseudo elastic ankle arthrodesis nail (AAN) was created based on the steps des cribed by Ryan Anderson et al. 46 Models were first imported into ScanIP +FE+CAD (Simpleware LTD) The model was created using a QCT scan of healthy leg bones and a CAD based model of a p seudo elastic nail provided by the manufacturer (MedShape).The healthy tibia and talus had portions of the bone removed, cut, and mated in the model as they wou ld be for an ankle fusion procedure. The p seudo elas tic and its associated components were oriented as it would be after a successful surgery. The t ibia, t alus, c alcaneus, and p seudo elastic nail were all given individual masks in the ScanIP software. Any floating elements were cleaned up using the island removal tool. Once floating elements were removed, small gaps in the model were closed using the cavity fill tool. The model was oriented such that the z axis is pointing down the length of the leg. Upon completion of mode l clean up the model is ready for meshing and then exportation. Figure 3 : Sca nIP model comparison for the Pseudo elastic nail mas ks (Right) and the unmodified bone model masks (Left)

PAGE 12

5 In ScanIP is used to prepare th e model for meshing. T he masks were smoothed using greyscale values and exported with length units in millimeters. The m odel configuration w as given a mesh coarseness of 25. A Convergence study by Ryan Anderson 46 showed that this was an appropriate mesh coarseness to get a sufficient number of elements without over saturating the models. After coarseness, the material properties of each component needed to be defined. Mass density of the p seudo elastic nail components does not affect the results of FEA that will be conducted on the model All of the components of the p seudo elastic nail except for the compressive element, were assig ned the material properties of t itan ium, Young modulus of 1.10x10 5 MPa and a Poisson Abaqus operates without units it is important to make sure Young ScanIP in the proper units. For each bone mask a greyscale based material property definit ion was used. Mass density of the bones were calculated by ScanIP based on greyscale brightness of the QCT scan and Young the determined density. Equations for the mass density were the same as described by Keyak et al. ( K e yak paper). Once again it was important to make sure that the equations matched the units being u sed in ScanIP millimeters and m egapascals, before exporting to Abaqus In order to prevent zero or negative modulus for portions of bone with very faint brig htness a blanket material property of 1MPa was used for all very low brightness values. A Poisson After material properties were defined, contact sets were defined between the p seudo elastic nail jacket and each bone. These contact sets allow for the bone and nail body to move relative to one another, rather than being bonded together. The model is finally meshed using the f ull model tool and exported as an .inp file that can be used in Abaqus software. The model is then imported in to Abaqus software. The first step in Abaqus is to remove the locking manually remove the locking material elements and replace them with a new pair of elements that will connect the compressive NiTi element to the nail body or jacket. To do this a group of nodes on the compressive element and a grou p of three nodes on the jacket are selected. Then, the .inp needs to be edited by deleting every element from the locking material mask except for the first two. Those

PAGE 13

6 two elements will have their nodes replaced with the nodes chosen above. The next step, once the locking material has been added is to update the material properties of the compressive element to those of Nickel Titanium. The nickel titanium material is assigned the name bed by Ryan Anderson 46 The necessary mechanical constants for the pseudo elastic material are described in the UMAT literature built into Abaqus Next, the sliding element of the p seudo elastic nail is pinned in place, and then the NiTi rod is stretched eight (8) mill imeters in the z direction. During this step the in teraction between the bones, ps eudo elastic nail and the locking material are turned off. In the next step the displacement of the NiTi rod is reduced to six (6) millimeters. This stretch followed by relaxa to maintain a level of nearly constant compression. After the programming steps, the bones and locking material are reactivated and then the proximal end of the tibia is pinned in place. from the f irst step. The locking material, after reactivation, holds the NiTi rod in a stretched state. After the NiTi rod is programmed and released then the gait load is applied. A peak gait load of 1121N is d ivided amongst 4 points, 280.25 per node, at the distal end of the Calcaneus surrounding the opening where the AAN is implanted. 1121 N was chosen to represent the maximum ground reaction force, 1.1 Body Weights (Sacco), during gait for a 104kg person. The applied load was also applied exclusively in the z direction because it is the direction of highest stress during gait. Once the model is completely set up then it is passed to the job module and submitted for analysis.

PAGE 14

7 Figure 4 : Load on H eel D uring G ait in P ercent B ody W eights (Sacco) Figure 5 Gait L oad F orce S ites and M agnitude Static Nail Model For the static nail models a physical t itanium static nail was measured using calipers and created in Solidworks. After the nail body was created, as set of screws were created and mated to the

PAGE 15

8 nail body. Diameters of the screws were determined based on the diameter of the anchoring and guide holes in the t itanium nail body and lengths were assigned based on the nail lengths from the p seudo elastic nail model. The nail model was then exported from Solidwork s and imported into ScanIP +FE+CAD with the same leg bones as the p seudo elastic model. The p seudo elastic masks were turned off and the t itanium surfaces were oriented in the same way and converted into masks. For model configuration the same settings were used with the lack of locking material and compressive element. The nail jacket is assigned its own material name with the material properties of Ti 6AL 4V and the screws are assigned the same properties with a separate name. This differentiation is important for e diting in Abaqus Contact sets and bone material properties are assigned the same as they were for the NiTi model. After model configuration is complete, the model is export as an .inp file for Abaqus Figure 6 Q C T S can and Solidworks M o del C ombined in ScanIP Once in Abaqus the material for the nail body needs t o be edited. The material needs to have an orthotropic expansion coefficient added, for this model an expansion coefficient of 3 =0.0005 is used. In addition to an expansion coefficient, the material has a specific heat of 1 and conductivity of 100 add ed to the material definition. The material definition additions allow for the nail body to contract when cooled and expand when heated up. For a carbon fiber nail the material definition is

PAGE 16

9 altered even further by changing the elastic material proper ties to reflect a modulus of 75GPa, this value was chosen based o n material properties listed in the documentation for the c arbon fiber AAN. The elements of the nail body then need to be selected and have their element type changed to the coupled temperature displacement, C3D4T, element type. After this, the same elements have their material orientation defined the same as the global coordinate system. For Abaqus steps, one step needs to be created to pin the most proximal end of the tibia in place. Once the tibia is pinned in place the nail body needs to have the temperature reduced by one, this is done by setting a boundary condition of temperature 1 for a selected region of nail body elements. Since Abaqus operates on a unitless system of measurement, the 1 temperature change triggers the expansion response of the nail body without a ffecting any other parts of the model. No other elements will be affected by the t emperature change. The final step is to apply a gait load with the same amplitude, magnitude, and location as used in the NiTi model. The model is then passed to the job module and submitted. Resorption Models For the resorption models of both the NiTi an d Static Nail the steps within Abaqus will be exactly the same. The only changes will be present in the steps concerning ScanIP For each m odel the most proximal slices of the talus, slices 300 309 totaling approximately 3mm in thickness, will have materia l properties changed from a greyscale based modulus to more compliant material properties. To do this the slices are copied to a new mask and the new mask given the new material properties, 1MPa Y Poisson ratio for example. Those s lices are then removed from the talus. The model is then exported to Abaqus and run as described above.

PAGE 17

10 Figure 7 Sca nIP M odel with the A ddition of a R esorpti on Z one Analyzing the Results Once the job module is finished in Abaqus the results can be viewed by right clicking on the completed job and selecting the resul ts option. Once in the results viewport the tibia, nail body, and compressive element (when applicable) need to be selected by using the display group manager. After isolating one group of elements, the tibia for example, a view cut needs to be made on the z axis to observe the force in the tibia. The cut should be made at the same slice for each group of elements, for this study slice 250 was used for each material. This slice is located on the distal end of the most proximal AAN screws Figure 8 Reaction F orce of B one in Z direction at S lice 250

PAGE 18

11 After choosing the view cut the reaction force values need to be displayed. This is done by enabling the free body slice check box and going into the options to disable heat flow display along with moments and enabling only the z direction component of force. After this is done the reaction for da ta can be exported using the XYD ata manager. To do this a new set of XYData needs to be created from the free body data. In the XYData creation menu it is impo rtant to double check that the z d irection compon ent is checked. The data can then be exported from Abaqus and then imported into Microsoft Excel, i G or, or MATLAB for visual representation The data is expor ted with reaction force in the z direction at each time step. When comparing each nail configuration the force in the z direction is most closely related to the force experienced from gait loading. For each configuration it is important to note the load in each material at the start of gait load or swing phase, where the gait load is zero. The next important data point is the load in each material at max gait load or toe off. For comparison of each configuration with a localized resorption zone, gait load is not as important as the amount of stress at zero load. The load in each nai l jacket, tibia, and compressive element, where applicable, should be noted in the models with no resorption zone and then noted in the models with a resorption zone. Even more data is acquired by incrementally raising and lowering the modulus of the resor ption zone.

PAGE 19

12 CHAPTER III RESULTS Pseudo elastic Nail with G ait The pseudo elastic Pseudo elastic maintained compression in the NiTi element at a 515 44 N throughout the gait cycle. At zero gait load, also known as swing phase, the tibia had a higher force value load then the nail jacket, 364 N compared to 156 N. Throughout the gait cycle the nail body, or jacket, only varied from 156 N to 534 N at toe off. The tibia by comparison had a minimum load of 364 N and rose as high as 1116 N at toe off. Figure 9 : Force Distribution during Gait Load for a Pseudo elastic Nail Model Titanium Nail with G ait The static titanium nail was given a prescribed contraction of 0.5%. This contraction provided a compression of about 770 N which was equal and opposite between the tibia and nail body during swing phase, zero gait load. Once gait load was applied, the tibia had force values that ranged from

PAGE 20

13 770 N to 965 N. The nail jacket had force values t hat ranged from +770 N to 156 N. This range means th at the tibia only absorbed 195 N of the maximum gait load and the nail body absorbed 926 N of the gait load. Figure 10 : Force Distribution during Gait Load for a Conventional Sta tic Titanium Ankle Arthrodesis Nail Carbon Fiber Nail with G a it The carbon fiber nail had a similar graph shape compared to the conventional titanium static nail. With an expansion coefficient of 0.5% the nail body generated 700 N of compression. In order to achieve the same compression value as the titanium na il an expansion coefficient of 0.55% needed to be applied. S imilar to the conventional titanium nail, the carbon fiber nail body absorbed the bulk of the gait load. The nail body had force values ranging from 705 N during swing phase to 172 N during toe off The tibia in this scenario had force values ranging from 705 N to 952 N during toe off, a difference of only 247 N. The tibia only experienced 22% of the gait load.

PAGE 21

14 Figure 11 : Force Distribution during Gait Load for a Convention al Static Carbon Fiber Body Ankle Arthrodesis Nail Gait L oad Sharing Ch aracteristics of Each N ail The load sharing of each nail configuration, shown in the figure below, shows that the p seudo elastic which has an internal compressive element allow for more of the gait load to be transferred to the bone as it would without a medical device rather than be absorbed by the device. The material used to manufacture a static nail without internal compression did very little to affect the load sharing characteristics. A static nail with a lower modulus, the carbon fiber based nail, allowed be more load to be transferred to the bone but not more than would be experience by the nail body.

PAGE 22

15 Figure 12 : Load Sharing of Different Ankle Arthrodesis Nails at Pe ak Gait Load Compression L oss with D ifferent Resorption V alues Once a resorption zone was added to each configuration each nail lost some compression. The pseudo elastic nail model was able to maintain more than 90 % of compression even with a resorption zone wit h a very low modulus, 1MPa. Both static nail configurations lost the majority of the compression generated by the nail, more than 8 0 %, when a low modulus resorption zone is added. Both the carbon fiber and the titanium static nails maintained compression until the modulus of the resorption zone dropped below 1000MPa Both the static nails shared similar curve shapes Both of the static nails w ere prescribed the same strain of 0.05% to generate the forces shown at the different resorption zone moduli The titanium static nail dropped to 1 09 N when given a resorption zone with a modulus of 1 MPa. The carbon fiber nail dropped to a simil ar 112 N when exposed to a resorption zone with the same modulus, 1 MPa Under healthy bone conditions, the static titanium nail had a compression value 770 N which was similar to the values experienced when the resorption zone was assigned a modulus of 10,000 N 67% 17% 22% 34% 83% 78% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% DynaNail (Toe Off) Versanail (Toe Off) Carbofix (Toe Off) Percent of Load Load Sharing Properties of Ankle Arthrodesis Nails Tibia Nail Body

PAGE 23

16 Figure 13 : Compression V alues of D ifferent A nkle Arthrodesis Nails when Exposed to a Specified Resorption Zone Modulus

PAGE 24

17 CHAPTER IV DI SCUSSION Ankle Arthrodesis Nails Under Gait Load Results showed that the pseudo elastic nail was able to maintain nearly constant compression of the pseudo elastic NiTi element throughout the gait cycle. The p seudo elastic nail also transferred 66% of the gait load to the tibia. The remaining load was split between the nail body, 33%, and the compr essive element, less than 1%. In contrast to the p seudo elastic nail model both static nail configurations transferred m ost of their load, 83% for the t itanium nail and 78% for the carbon fiber nail to the nail jacket The remaining force from gait loading, in both static nail models, was transferred to the tibia and other bones. According to Wol f f s law 47 bone growth is promoted by c yclical compression and tension of the bone The load sharing properties in Figure 12 indicate that the pseudo elastic nail is better at providing the needed cyclical loading of the tibia after installation. The static na ils on the other hand do not allow most of the load to transfer to the bone. Ankle Arthrodesis Nails with a Resorption Zone IM nail design and materials played a significant role in maintaining compression and load sharing. The pseudoelastic nail had the lowest degree of stress shielding (32%) and maintained compression for over 0.10 mm of simulated resorption. C onstant compression and the avoidance of resorption gapping is paramount to drive primar y bone healing in joint fusions due to lack of perioste al/endosteal anatomy crossing the fusion site, thus crippling the ability for secondary bone healing (callus healing). This model allows for direct comparison between devices and can be used pre operatively to predict patient specific performance and help aid in device selection for TTC fusion. Limitations The aims of this study were to analyze different ankle arthrodesis nails under a variety of conditions. The first aim was to study each nail under gait loading conditions. The second was to analyze the nails when exposed to a localized resorption zone. For each of these aims a healthy leg

PAGE 25

18 QCT scan was used. The osteotomy was done by cutting the bones from the QCT scan within the SCANIP software. This allowed for an osteotomy where the bones were all pres sed directly against one another without any shifting or comp action required. This also allowed for an idealized resorption zone where the area of resorption was able to be chosen within the software and the modulus of the zone adjusted as needed. The cont act between the bones and the nail jacket was also set to frictionless and there were no gaps allowed between the bones and the device. Future Work For future advancement of this study, there are several avenues of advancement and improvement. There should be a continuation of this study using QCT scans of several different legs. It would also benefit from scans of unhealthy legs and legs of varying ages. There is also room for comparison of the results from this experiment to values in a compression test of a cadaver leg with one of the medical devices installed. This would allow us to determine the validity of our results. After the numbers from this experiment are confirmed, it would be useful to create a model of the static nails with their respect ive dynamization slots activated. In practice, this would require another surgery and is usually reserved for post union cases or as a last resort in non union to reduce the stress shielding effects of the device. Confirming or disproving these claims wil l help to determine the load sharing effects of the dynamization slots. Conclusions 1. The pseudo elastic nail allowed for 67% of the load experienced from the cycle to be transferred to the tibia T he static c arbon fiber and t itanium nails allowed only 22% and 17% of the load to transfer, respectively. 2. The pseudo elastic nail maintained compression, within 10% of original values, when subjected to the resorption zone Both static nails lost more than 8 0% of original compression when subjected to the resorption zone.

PAGE 26

19 REFERE NCES 1. Basumallick, M. N., and Bandopadhyay, A.: Effect of dynamization in open interlocking nailing of femoral fractures. A prospective randomized comparative study of 50 cases with a 2 year follow up. Acta Orthop Belg, 68(1): 42 8, 2002. 2. Bennett, G. L.; C ameron, B.; Njus, G.; Saunders, M.; and Kay, D. B.: Tibiotalocalcaneal arthrodesis: a biomechanical assessment of stability. Foot Ankle Int, 26(7): 530 6, 2005. 3. Berson, L.; McGarvey, W. C.; and Clanton, T. O.: Evaluation of compression in intramedullary hindfoot arthrodesis. Foot Ankle Int, 23(11): 992 5, 2002. 4. Bevernage, B. D.; Deleu, P. A.; Maldague, P.; and Leemrijse, T.: Technique and early experience with pos terior arthroscopic tibiotalocalcaneal arthrodesis. Orthop Traumatol Surg Res, 96(4): 469 75, 2010. 5. Boer, R.; Mader, K.; Pennig, D.; and Verheyen, C. C.: Tibiotalocalcaneal arthrodesis using a reamed retrograde locking nail. Clin Orthop Relat Res, 463: 151 6, 2007. 6. Brumback, R. J.: The rationales of interlocking nailing of the femur, tibia, and humerus. Clin Orthop Relat Res, (324): 292 320, 1996. 7. Charnley, J.: Compression arthrodesis of the ankle and shoulder. J Bone Joint Surg, 33B: 180 91, 1951. 8. Conway, J. D.: Charcot salvage of the foot and ankle using external fixation. Foot Ankle Clin, 13(1): 157 73, 2008. 9. Eichhorn, S.; Lindner, T.; Mckley, T.; Trapp, O.; and Steinhauser, E.: The loss of compression in intramedullary ankle Arthrodesis using two differ ent types of compression rods -A biomechanical study. Journal of Biomechanics, 39(Supplement 1): S136 S136, 2006. 10. Fazal, M. A.; Garrido, E.; and Williams, R. L.: Tibio talo calaneal arthrodesis by retrograde intramedullary nail and bone grafting. Foot and Ankle Surgery, 12(4): 185 190, 2006. 11. Fragomen, A. T.; Meyers, K. N.; Davis, N.; Shu, H.; Wright, T.; and Rozbruch, S. R.: A biomechanical comparison of micromotion after ankle fusion using 2 fixation techniques: intramedullary arthrodesis nail or Ilizarov external fixator. Foot Ankle Int, 29(3): 334 41, 2008. 12. Frey, C.; Halikus, N. M.; Vu Rose, T.; and Ebramzadeh, E.: A review of ankle arthrodesis: predisposing factors to nonunion. Foot Ankle Int, 15(11): 581 4, 1994. 13. Gil, F. J., and Planell, J. A.: Shape memory alloys for medical applications. Proc Inst Mech Eng H, 212(6): 473 88, 1998. 14. Gilbert, J. A.; Dahners, L. E.; and Atkinson, M. A.: The effect of external fixation stiffness on early healing of transverse osteotomies. J Orthop Res, 7(3): 389 97 1989. 15. Haaker, R.; Kohja, E. Y.; Wojciechowski, M.; and Gruber, G.: Tibio talo calcaneal arthrodesis by a retrograde intramedullary nail. Ortop Traumatol Rehabil, 12(3): 245 9, 2010.

PAGE 27

20 16. iData_Research_Inc.: US Markets for Orthopedic Trauma Devices. 2010. 17. Kim C.; Catanzariti, A. R.; and Mendicino, R. W.: Tibiotalocalcaneal arthrodesis for salvage of severe ankle degeneration. Clin Podiatr Med Surg, 26(2): 283 302, 2009. 18. Lee, A. T.; Sundberg, E. B.; Lindsey, D. P.; Harris, A. H.; and Chou, L. B.: Biomechanical comparison of blade plate and intramedullary nail fixation for tibiocalcaneal arthrodesis. Foot Ankle Int, 31(2): 164 71, 2010. 19. Means, K. R.; Parks, B. G.; Nguyen, A.; and Schon, L. C.: Intramedullary nail fixation with posterior to anterior compared to t ransverse distal screw placement for tibiotalocalcaneal arthrodesis: a biomechanical investigation. Foot Ankle Int, 27(12): 1137 42, 2006. 20. Mckley, T.; Eichorn, S.; Hoffmeier, K.; von Oldenburg, G.; Speitling, A.; Hoffmann, G. O.; and Buhren, V.: Biomechan ical evaluation of primary stiffness of tibiotalocalcaneal fusion with intramedullary nails. Foot Ankle Int, 28(2): 224 31, 2007. 21. Mckley, T. M.; Eichorn, S.; von Oldenburg, G.; Speitling, A.; DiCicco, J. D., 3rd; Hofmann, G. O.; and Buhren, V.: Biomechani cal evaluation of primary stiffness of tibiotalar arthrodesis with an intramedullary compression nail and four other fixation devices. Foot Ankle Int, 27(10): 814 20, 2006. 22. Mueckley, T. M.; Eichorn, S.; von Oldenburg, G.; Speitling, A.; DiCicco, J. D., 3rd ; Hofmann, G. O.; and Buhren, V.: Biomechanical evaluation of primary stiffness of tibiotalar arthrodesis with an intramedullary compression nail and four other fixation devices. Foot Ankle Int, 27(10): 814 20, 2006. 23. Nihal, A.; Gellman, R. E.; Embil, J. M. ; and Trepman, E.: Ankle arthrodesis. Foot Ankle Surg, 14(1): 1 10, 2008. 24. Perlman, M. H., and Thordarson, D. B.: Ankle fusion in a high risk population: an assessment of nonunion risk factors. Foot Ankle Int, 20(8): 491 6, 1999. 25. Plaass, C.; Knupp, M.; Barg, A.; and Hintermann, B.: Anterior double plating for rigid fixation of isolated tibiotalar arthrodesis. Foot Ankle Int, 30(7): 631 9, 2009. 26. Sharma, D. K.; Yunas, I.; and Ramos, J.: Management of complications of tibiotalocalcane al arthrodesis in patients with co morbidity. The Foot, 16(1): 33 37, 2006. 27. Terjesen, T., and Apalset, K.: The influence of different degrees of stiffness of fixation plates on experimental bone healing. J Orthop Res, 6(2): 293 9, 1988. 28. Vesely, R.; Prochaz ka, V.; Visna, P.; Valentova, J.; and Savolt, J.: [Tibiotalocalcaneal arthrodesis using a retrograde nail locked in the sagittal plane]. Acta Chir Orthop Traumatol Cech, 75(2): 129 33, 2008. 29. Wu, J. J.; Shyr, H. S.; Chao, E. Y.; and Kelly, P. J.: Comparison of osteotomy healing under external fixation devices with different stiffness characteristics. J Bone Joint Surg Am, 66(8): 1258 64, 1984. 30. Yakacki, C. M.; Gall, K.; Dirschl, D. R.; and Pacaccio, D. J.: Pseudo elastic intramedullary nailing for tibio talo calcaneal arthrodesis. Expert Rev Med Devices, 8(2): 159 66, 2011.

PAGE 28

21 31. Yakacki, C. M.; Khalil, H. F.; Dixon, S. A.; Gall, K.; and Pacaccio, D. J.: Compression forces of internal and external ankle fixation devices with simulated bone resorption. Foot Ankle Int 31(1): 76 85, 2010. 32. Yakacki, Christopher M, et al. "Pseudoelastic intramedullary nailing for tibio talo calcaneal arthrodesis." Expert Review of Medical Devices 8.2 (2011): 159+. Academic OneFile Web. 2 May 2016 33. Rho, Jae Young, Richard B. Ashman, and Ch arles H. Turner. "Young's modulus of trabecular and cortical bone material: ultrasonic and microtensile measurements." Journal of biomechanics 26.2 (1993): 111 119. 34. EMG and Ground Reaction Forces between Barefoot and Shod Gait in Participants with BMC Musculoskeletal Disorders 11 (2010): 24. PMC Web. 2 May 2016. 35. Hsu, Andrew R., J. Kent Ellington, and Samuel B. Adams. "Tibiotalocalc aneal Arthrodesis Using a Nitinol Intramedullary Hindfoot Nail." Foot & ankle specialist (2015): 1938640015598838. 36. Generating Journal of the Mechanical Behavior of Biomedical Materials (2016) 37. Foot Ankle Int March 2016 37: 294 299, first published on October 15, 2015 doi:1 0.1177/1071100715611891 38. Yakacki, Christopher M., et al. "Compression forces of internal and external ankle fixation devices with simulated bone resorption." Foot & ankle international 31.1 (2010): 76 85. 39. ASM Material Data Sheet Titanium Ti 6Al 4V (Grade 5), Annealed http://asm.matweb.com/search/SpecificMaterial.asp?bassnum=MTP641 40. L. Fortington and J. Geertzen, "Short and Long Term Mortality Rates after a Lower Limb Amputation", Sciencedirect.com 2016. [Online]. Available: http://www.sciencedirect.com/sc ience/article/pii/S1078588413002256. [Accessed: 25 Nov 2016]. 41. M. Erduran, "Biomechanical effects of the distance from the fracture zone to the interlocking fixation screw of intramedullary nail", Jbiomech.com 2016. [Online]. Available: http://www.jbiome ch.com/article/S0021 9290(11)00119 9/abstract. [Accessed: 25 Nov 2016]. 42. A. Nihal, "Ankle Arthrodesis", Sciencedirect.com 2016. [Online]. Available: http://www.sciencedirect.com/science/article/pii/S1268773107000744. [Accessed: 25 Nov 2016]. 43. C. Yakacki K. Gall, D. Dirschl and D. Pacaccio, "Pseudoelastic intramedullary nailing for tibio talo calcaneal arthrodesis", Tandfonline.com 2016. [Online]. Available: http://www.tandfonline.com/doi/abs/10.1586/erd.10.93. [Accessed: 25 Nov 2016].

PAGE 29

22 44. https://patients.stryker.com/assets/ankle_fusion 97bd33028a61e413fcfa119bb288b664902eef5dff0ab03c175081f786922f6f.jpg [Accessed 25 Nov 2016] 45. l femoral strength Clinical orthopaedics and related research 219 ,August 2005. [Accessed 25 Nov 2016] Online: http://journals.lww.com/corr/Abstract/2005/10000/Predicting_the_Strength_of_Femoral_Sha fts_with_and.30.aspx 46. element analysis of a pseudoelastic compression generating intramedullary ankle arthrodesis Journal of the Mechanical Behavior of Miomedical Materials, Septemb er 2016 Volume 62 Pages 83 92 [Online] http://www.sciencedirect.com/science/article/pii/S1751616116301102 47. Wolff, Julius. The law of bone remodellin g Springer Science & Business Media, 2012