<%BANNER%>

Tensile and Viscoelastic Properties of Two and Four Strand Anterior Tibialis and Peroneus Longus Grafts as a Substitute ...

Permanent Link: http://ufdc.ufl.edu/UFE0043043/00001

Material Information

Title: Tensile and Viscoelastic Properties of Two and Four Strand Anterior Tibialis and Peroneus Longus Grafts as a Substitute for Anterior Cruciate Ligament Recnstruction
Physical Description: 1 online resource (64 p.)
Language: english
Creator: SHIELDS,ERIC P
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ANTERIOR -- CRUCIATE -- LIGAMENT -- LONGUS -- PERONEUS -- TIBIALIS
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The tearing of the anterior cruciate ligament (ACL) is one of the most common sports related injuries, occurring over 250,000 times a year in the United States. If a revision of the ACL is not performed, abnormal knee kinematics may occur, further damaging the knee. The purpose of this study was to examine the tensile and viscoelastic properties of two common ACL replacement grafts as a function of surgical size and number of strands. All test samples were sized and grouped according to a novel sizing technique, mimicking standard surgical procedure. The comparison properties derived from this test include the relaxation rate, recovery rate, ultimate tensile force (UTF), ultimate tensile stress (UTS), strain at failure, and Young?s modulus. For the first evaluation, the groups for comparison included two-strand tendons with loop diameter (LD) sizes that ranged from 7.00 mm to 7.75 mm (Group 1) and 8.00 mm to 8.75 mm (Group 2). Four groups were then defined: two LD groups created from AT tendons (AT1, AT2) and two LD groups created from PL tendons (PL1, PL2). A one-way ANOVA for UTF determined the AT2 group had a 493.0 N higher UTF compared to AT1 and an 871.8 N higher UTF compared to PL1. Additionally, the PL2 group had a 744.7 N higher UTF compared to the PL1 group. A Browne-Forsythe for cross-sectional area (CSA) determined the AT2 group had a 9.02 mm2 higher CSA compared to the AT1 group and a 7.50 mm2 higher CSA compared to the PL1 group. Additionally, a Browne-Forsythe for CSA determined the PL2 group had a 9.48 mm2 higher CSA compared to the AT1 group and a 7.96 mm2 higher CSA compared to PL1 group. For the second evaluation, the AT two-strand graft (AT3) was compared to the PL four-strand graft (PL4). The AT3 graft had an LD size ranging from 9.00 mm to 9.75 mm. The PL4 graft had an LD size ranging from 10.00 mm to 10.75 mm. A two-sample t-test determined the PL4 group had a 661 N higher UTF and a 14.49 mm2 larger CSA than the AT3 group. All sample groups tested in this study exceeded the maximum forces of the native ACL. No practical difference was found between the AT and PL across similar sizes, displaying no difference between these two types of tendons. Additionally, only UTF and CSA were practically different as the LD size increased. This data supports both the AT and PL as suitable ACL replacement grafts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by ERIC P SHIELDS.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Banks, Scott A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043043:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043043/00001

Material Information

Title: Tensile and Viscoelastic Properties of Two and Four Strand Anterior Tibialis and Peroneus Longus Grafts as a Substitute for Anterior Cruciate Ligament Recnstruction
Physical Description: 1 online resource (64 p.)
Language: english
Creator: SHIELDS,ERIC P
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ANTERIOR -- CRUCIATE -- LIGAMENT -- LONGUS -- PERONEUS -- TIBIALIS
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The tearing of the anterior cruciate ligament (ACL) is one of the most common sports related injuries, occurring over 250,000 times a year in the United States. If a revision of the ACL is not performed, abnormal knee kinematics may occur, further damaging the knee. The purpose of this study was to examine the tensile and viscoelastic properties of two common ACL replacement grafts as a function of surgical size and number of strands. All test samples were sized and grouped according to a novel sizing technique, mimicking standard surgical procedure. The comparison properties derived from this test include the relaxation rate, recovery rate, ultimate tensile force (UTF), ultimate tensile stress (UTS), strain at failure, and Young?s modulus. For the first evaluation, the groups for comparison included two-strand tendons with loop diameter (LD) sizes that ranged from 7.00 mm to 7.75 mm (Group 1) and 8.00 mm to 8.75 mm (Group 2). Four groups were then defined: two LD groups created from AT tendons (AT1, AT2) and two LD groups created from PL tendons (PL1, PL2). A one-way ANOVA for UTF determined the AT2 group had a 493.0 N higher UTF compared to AT1 and an 871.8 N higher UTF compared to PL1. Additionally, the PL2 group had a 744.7 N higher UTF compared to the PL1 group. A Browne-Forsythe for cross-sectional area (CSA) determined the AT2 group had a 9.02 mm2 higher CSA compared to the AT1 group and a 7.50 mm2 higher CSA compared to the PL1 group. Additionally, a Browne-Forsythe for CSA determined the PL2 group had a 9.48 mm2 higher CSA compared to the AT1 group and a 7.96 mm2 higher CSA compared to PL1 group. For the second evaluation, the AT two-strand graft (AT3) was compared to the PL four-strand graft (PL4). The AT3 graft had an LD size ranging from 9.00 mm to 9.75 mm. The PL4 graft had an LD size ranging from 10.00 mm to 10.75 mm. A two-sample t-test determined the PL4 group had a 661 N higher UTF and a 14.49 mm2 larger CSA than the AT3 group. All sample groups tested in this study exceeded the maximum forces of the native ACL. No practical difference was found between the AT and PL across similar sizes, displaying no difference between these two types of tendons. Additionally, only UTF and CSA were practically different as the LD size increased. This data supports both the AT and PL as suitable ACL replacement grafts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by ERIC P SHIELDS.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Banks, Scott A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043043:00001


This item has the following downloads:


Full Text

PAGE 1

TENSILE AND VISCOELASTIC PROPER TIES OF TWO AND FOUR STRAND ANTERIOR TIBIALIS AND PERONEUS LO NGUS GRAFTS AS A SUBSTITUTE FOR ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION By ERIC P. SHIELDS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011 1

PAGE 2

2011 Eric P. Shields 2

PAGE 3

To Theodore J. Dorsa, my champion 3

PAGE 4

ACKNOWLEDGMENTS I would like to thank RTI Biologics, Inc. for providing both t he time and resources necessary to complete this study. Pedro Pedr oso, your dedication to the orthopaedic field is infectious. You inspired me to a ccomplish goals I never thought achievable. I would also like to thank my committee chair, Dr. Scott Banks, sparking my interest in research while taking his So lid Biomechanics course. Additionally, I would like to thank my committee members, Dr. Malisa Sarntinoranont and Dr. Benjamin Fregly, for their contribut ions to this study. Finally, I would like to thank my parents. My father instilled his work ethic and my mother, her compassion. They have provided every opportunity for me to grow, and for that I am thankful. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES ............................................................................................................ 8LIST OF FI GURES .......................................................................................................... 9LIST OF ABBR EVIATION S ........................................................................................... 10ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRODUCT ION ........................................................................................................ 13Anterior Cruciate Liga ment Replac ement ............................................................... 13Knee Anat omy ........................................................................................................ 13Anterior Cruciate Ligament Reconstructi on ............................................................ 14Multiple Strand Graf t Replacem ent ......................................................................... 15Allograft ................................................................................................................... 16Soft Tissue Grafts with Bone Blocks ................................................................ 17Achilles tendon ........................................................................................... 17Bone-patellar tendon-bone ........................................................................ 17Pure Soft Tissue Grafts without Bone Bl ocks ................................................... 18Anterior ti bialis ........................................................................................... 18Peroneus l ongus ........................................................................................ 19Posterior tibialis .......................................................................................... 19Semitendinosus and gracilis ...................................................................... 19Allograft Recovery Process ..................................................................................... 20Graft Sizing ............................................................................................................. 21Relaxation Data ...................................................................................................... 21Introduction into Pr esent St udy ............................................................................... 222 MATERIALS AN D METHOD S ................................................................................... 26Tissue Allocation ..................................................................................................... 26Sizing Te st .............................................................................................................. 26Tendon Preparat ion .......................................................................................... 26Test Meth od ..................................................................................................... 27Mechanical Test ...................................................................................................... 28Tendon Preparat ion .......................................................................................... 28Test Fixture ...................................................................................................... 28Test Meth od ..................................................................................................... 29Sizing Dam age Test ............................................................................................... 30Statistical Methods .................................................................................................. 31 5

PAGE 6

Sample Size Ca lculations ................................................................................. 31Statistical A nalysis ............................................................................................ 323 RESULT S ................................................................................................................... 3 7Mechanical Test ...................................................................................................... 37Biomechanical Comparison of Two-St rand AT versus Tw o-Strand PL ................... 37Viscoelastic Portion .......................................................................................... 37First relaxation phase properti es ................................................................ 37Recovery phase propertie s ........................................................................ 38Second relaxation phase properties ........................................................... 39Cyclic Loadi ng Portio n ...................................................................................... 40Average Youngs modulus during the cyclic portion ................................... 40Load to Failure Portion ..................................................................................... 40Ultimate tensile force ................................................................................. 40Ultimate tensile stress ................................................................................ 40Cross-sectional area .................................................................................. 41Youngs modulus during the failure curve .................................................. 41Strain at failure ........................................................................................... 41Extension at failure .................................................................................... 41Biomechanical Comparison of Two-St rand AT versus F our-Strand PL .................. 42Viscoelastic Portion .......................................................................................... 42First relaxation phase properti es ................................................................ 42Recovery phase propertie s ........................................................................ 42Second relaxation phase properties ........................................................... 43Cyclic Loadi ng Portio n ...................................................................................... 43Average Youngs modulus during the cyclic portion ................................... 43Load to Failure Portion ..................................................................................... 43Ultimate tensile force ................................................................................. 43Ultimate tensile stress ................................................................................ 44Cross-sectional area .................................................................................. 44Youngs modulus during the failure curve .................................................. 44Strain at failure ........................................................................................... 44Extension at failure .................................................................................... 44Biomechanical Comparison of Weak Pull versus St rong Pull ................................. 44Load to Failure Portion ..................................................................................... 45Ultimate tensile force ................................................................................. 45Ultimate tensile stress ................................................................................ 45Cross-sectional area .................................................................................. 45Youngs modulus during the failure curve .................................................. 45Strain at failure ........................................................................................... 454 DISCUSS ION ............................................................................................................. 515 TEST COMPARISONS: LIMITAT IONS AND FUTURE RESEARCH ......................... 57Limitations ............................................................................................................... 57 6

PAGE 7

Future Res earch ..................................................................................................... 58REFERENCE LIST........................................................................................................ 60BIOGRAPHICAL SKETCH ............................................................................................ 64 7

PAGE 8

LIST OF TABLES Table page 3-1 Statistical analysis for the two-st rand versus two-str and evaluation ................... 463-2 Statistical analysis for the two-st rand versus four-s trand evaluation .................. 473-3 Statistical analysis for the sizing dam age test .................................................... 473-4 Mechanical results for the sizing dam age test .................................................... 473-5 Two-strand versus two-strand first relaxation result s .......................................... 483-6 Two-strand versus two-st rand recovery result s .................................................. 483-7 Two-strand versus two-strand second relaxati on result s .................................... 483-8 Two-strand versus two-strand cyclic loadi ng result s ........................................... 483-9 Two-strand versus two-strand failure curv e result s ............................................ 483-10 Two-strand versus two-strand failure curve strain and ext ension results ........... 493-11 Two-strand AT versus four-strand PL first relaxati on results .............................. 493-12 Two-strand AT versus four-s trand PL recovery results ....................................... 493-13 Two-strand AT versus four-str and PL second relaxa tion results ........................ 493-14 Two-strand AT versus four-s trand PL cyclic loading results ............................... 493-15 Two-strand AT versus four-str and PL failure cu rve results ................................. 493-16 Two-strand AT versus four-strand PL failure curve strain and extension result s ................................................................................................................. 504-1 Theoretical CSA compared to actual m easured CS A ......................................... 56 8

PAGE 9

LIST OF FIGURES Figure page 1-1 Average UTS values for commo n ACL replacement grafts ................................ 231-2 AT tendon, PL tendon, and ACL compar ison ..................................................... 241-3 Sizing block for determining LD size ................................................................... 252-1 Sizing test fixture ................................................................................................ 342-2 Mechanical test grip ............................................................................................ 342-3 Mechanical te st fixt ure ........................................................................................ 352-4 Mechanical test fixture with top and bottom ice holder s ..................................... 352-5 Test method schematic for mechanical test ....................................................... 36 9

PAGE 10

LIST OF ABBREVIATIONS AATB American Association of Tissue Banks ACL anterior cruciate ligament AT anterior tibialis tendon AT1 2-strand AT tendon with a LD size between 7.00 and 7.75 AT2 2-strand AT tendon with a LD size between 8.00 and 8.75 AT3 2-strand AT tendon with a LD size between 9.00 and 9.75 CSA cross-sectional area E Youngs modulus LD loop diameter PL peroneus longus tendon PL1 2-strand PL tendon with a LD size between 7.00 and 7.75 PL2 2-strand PL tendon with a LD size between 8.00 and 8.75 PL4 4-strand PL tendon with a LD size between 10.00 and 10.75 UTF ultimate tensile force UTS ultimate tensile stress 10

PAGE 11

Abstract of Thesis Pres ented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Master of Science TENSILE AND VISCOELASTIC PROPER TIES OF TWO AND FOUR STRAND ANTERIOR TIBIALIS AND PERONEUS LO NGUS GRAFTS AS A SUBSTITUTE FOR ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION By Eric P. Shields May 2011 Chair: Scott Banks Major: Mechanical Engineering The tearing of the anterio r cruciate ligament (ACL) is one of the most common sports related injuries, occurring over 250,000 times a year in the United States. If a revision of the ACL is not performed, abno rmal knee kinematics ma y occur, further damaging the knee. The purpose of this study was to examine the tensile and viscoelastic properties of two common ACL replacement grafts as a function of surgical size and number of strands. All test sample s were sized and grouped according to a novel sizing technique, mimicking standard surgical procedure. The comparison properties derived from this test include the relaxation rate, recove ry rate, ultimate tensile force (UTF), ultimate tensile stress (UTS ), strain at failure, and Youngs modulus. For the first evaluation, the groups for comparison included two-strand tendons with loop diameter (LD) size s that ranged from 7.00 mm to 7.75 mm (Group 1) and 8.00 mm to 8.75 mm (Group 2). F our groups were then defined: two LD groups created from AT tendons (AT1, AT2) and two LD groups created from PL tendons (PL1, PL2). A oneway ANOVA for UTF determined the AT2 gr oup had a 493.0 N higher UTF compared to AT1 and an 871.8 N higher UTF compared to PL1. Additionally, the PL2 group had a 744.7 N higher UTF compared to the PL1 gr oup. A Browne-Forsythe for cross-sectional 11

PAGE 12

12 area (CSA) determined the AT2 group had a 9.02 mm2 higher CSA compared to the AT1 group and a 7.50 mm2 higher CSA compared to the PL1 group. Additionally, a Browne-Forsythe for CSA determined the PL2 group had a 9.48 mm2 higher CSA compared to the AT1 group and a 7.96 mm2 higher CSA compared to PL1 group. For the second evaluation, the AT two-st rand graft (AT3) was compared to the PL four-strand graft (PL4). The AT3 graft had an LD size ranging fr om 9.00 mm to 9.75 mm. The PL4 graft had an LD size ranging from 10.00 mm to 10.75 mm. A two-sample t-test determined the PL4 group had a 661 N higher UTF and a 14.49 mm2 larger CSA than the AT3 group. All sample groups tested in this study ex ceeded the maximum fo rces of the native ACL. No practical difference was found bet ween the AT and PL across similar sizes, displaying no difference between these two ty pes of tendons. Additionally, only UTF and CSA were practically different as the LD size increased. This data supports both the AT and PL as suitable ACL replacement grafts.

PAGE 13

CHAPTER 1 INTRODUCTION Anterior Cruciate Ligament Replacement The tearing of the anterio r cruciate ligament (ACL) is one of the most common sports related injuries, occurring over 250, 000 times a year in the United States.1 If no revision of the ACL is performed, abnormal knee kinematics and instability may occur, resulting in greater damage to the knee.2 Since mid-substance ACL tears do not have the ability to self-heal,3 approximately 100,000 to 250,000 ACL reconstructive surgeries are performed in the Un ited States each year.4 Each year, nearly $1 billion is spent for ACL reconstructions in the United States.4 In order to restor e proper knee function, reconstructive surgery with an autograft or allograft tissue implant with known mechanical characteristics is necessary. Rec ent research has suggested that the use of multiple strands of tissue during reconstructive surgery may have a clinical advantage over the use of a single strand.5-7 For this study, the characterization of the anterior tibialis (AT) and peroneus longus (PL) tendon as a direct function of surgical size and number of strands was investi gated to determine any statistical difference in mechanical properties. Knee Anatomy The knee joint consists of four ligaments: the ACL, the posterior cruciate ligament (PCL), the medial collateral ligament (MCL), and the lateral collateral ligament (LCL). Both of the cruciate ligament s are intraarticular. The AC L extends from the lateral femoral condyle and runs anterio-medially and distally to the head of the tibia.8 It is responsible for restricting anter ior translation of the tibia wit h respect to the femur. The PCL extends from the medial femoral condy le to the posterior tibia and restricts 13

PAGE 14

posterior translation of the tibia with respec t to the femur. The LCL extends from the lateral femoral epicondyle to the head of the fibula. The MCL extends from the medial femoral epicondyle to the posteromedial tibi al crest. The two collateral ligaments work together to restrict va rus and valgus angulation.9 Anterior Cruciate Ligament Reconstruction The ACL plays a critical role in knee join t stability. The lengt h of the ACL ranges from 22 to 41 mm while the width of the AC L ranges from 7 to 12 mm. The ACL has an irregular, non-uniform cross-sectional shape that increases from the lateral femoral condyle to the head of the tibia. Named after t heir tibial insertions, the ACL is composed of two bundles: the anteromedial (AM) bun dle and the posterolateral (PL) bundle.8 Each bundle experiences different loads at different degrees of knee flexion. As the knee is flexed from 20 to 90 the AM bundle in situ force increases while the PL bundle in situ force decreases. Conversely, the PL bundle in situ force increases while the AM bundle in situ force decreases when the knee extends.8,10 The ACL displays complex mechanical charac teristics. As a viscoelastic material, it displays time-dependent behavior. When the AC L is held at a constant strain, it will exhibit a decrease in stress known as stress relaxation. Additionally the ACL will exhibit creep, an increase in strain when a constant stress is applied. Once the stress or strain has been removed, the ACL will display a period of recovery. It is important to fully understand these characteristics in order to determine a suitable re placement graft. Failure of the ACL occurs from sudden moti ons, such as pivoting or landing from a jump, normally associated with sports-related activities.11 The mechanical strength of the native ACL has been extensively res earched. For younger don ors (16-26 years), 14

PAGE 15

Noyes et al.12 determined the ultimate tensile force (UTF) to be 1725 269 N with an ultimate tensile stress (UTS) of 37.8 3.8 MPa and an elastic modulus of 111 MPa. For older donors (48-86 years), the mechanical pr operties decreased with a UTF of 622 N, a UTS of 13.3 MPa, and an elastic modulus of 65.3 MPa for the same study. Woo et al.13 determined the UTF of the native ACL to be 2,160 N for younger donors (22-35 years) and 658 N for older donor s (60-97 years). With t he dramatic decrease in mechanical properties as age increases, t he demand for a proper ACL reconstruction technique will continue to grow as an agi ng population continues to pursue active lifestyles. Multiple Strand Graft Replacement As previously stated, the ACL is co mposed of two bundles, with the AM bundle carrying more of the load during flexion and the PL bundle playing a more active role during extension. Due to the short lengt h of the ACL, it cannot be used as a replacement graft for ACL reconstruction. One method to restore knee kinematics includes using a single bundle of either allograft or aut ograft tissue. This technique restores natural kinematics dur ing flexion; however, recent evidence shows that a single bundle technique may fail to restore stability during a rotary load.5 In order to restore more natural knee kinematics, the use of a double bundle technique has been investigated.6,7 During ACL reconstruction, a bundl e can be composed of a one-strand graft, a two-strand graft, or a looped graft. A looped gra ft is achieved when a single tendon is folded in half to creat e a two-strand graft. For th is reason, research into multiple strand grafts to obtain a ccurate mechanical strength data has been performed.14-19 15

PAGE 16

Allograft The preferred method for ACL reconstr uction can be performed with either an autograft or allograft. The benefits of usi ng allograft tissue include: no donor-site morbidity from graft recovery, no size limitat ion, significant decrease in operative time, decreased pain due to smaller incisions, and no trauma to host tissue. During revision surgery, allograft tissue may be necessary to achieve the required graft size and to reduce further damage of native tissue. Additiona lly, allografts are t he preferred method for multiple strand reconstruction techniques since autograft tissue supply is often limited.20 In 2004, more than 1 million bone and tiss ue allografts were used in orthopaedic sports medicine surgeries from the 86 tissue banks composing the American Association of Tissue Banks (AATB).21 In 2006, this number grew to 1.5 million distributed throughout the Unit ed States. According to a 2006 American Orthopaedic Society for Sports Medicine survey, 86% of the surveyed populat ion reported using allograft tissue.22 An estimated 20% of all ACL reconstruction procedures are performed with allograft tissue.21 A concern with the usage of a llograft tissue for ACL recons truction is the risk of disease transmission. With donor screening in place, Buck et al.23 estimated the risk of receiving tissue from a donor infected with HIV was one in 1,667,000 compared to the risk of HIV transmission from a unit of blood at one in 200,000-800,000.24 To further reduce the risk of infection, sterilization techniques, such as gamma irradiation and chemical processing, can be performed.22 Although the threat of infection is still present, donor screening and sterilization techniques are further reducing this concern. 16

PAGE 17

Two main types of allograft tissue may be used: soft tissue gr aft with bone blocks attached for fixation or pure soft tissue with no bone blocks. Soft tissue grafts with bone blocks require less time to incorporate into host tissue;25 however, pure soft tissue grafts have mechanically superior properti es as a result of the abili ty to use multiple strand techniques.23 Figure 1-1 compares the average UTS values for common ACL replacement grafts. It is important to note that test ing procedure has an impact on mechanical properties. Factors such as strain rate,26-30 hydration,31-33 and temperature34, 35 have been shown to affect the mechanical properties. Soft Tissue Grafts with Bone Blocks The Soft tissue grafts with bone block options include the calcaneusAchilles tendon allograft and the bone-pate llar tendon-bone allograft. Achilles tendon Located on the posterior side of the ti bia, the Achilles tendon connects the plantaris, gastrocnemius, and so leus to the calcaneus bone. For a strain rate of 1 mm per second, Wren et al.36 determined the UTF and UTS of the Achilles tendon to be 4,617 1,107 N and 71 17 MPa, respectively. With an increase in strain rate to 10 mm per second, the UTF and UTS increased to 5,579 1,143 N and 86 24 MPa, respectively. The Achilles tendon is not available as an autograft option. Bone-patellar tendon-bone Considered the gold-standard for ACL reconstruction, t he patellar tendon, also known as the patellar ligament, connects the patella to the tibia. Flahiff et al.37 determined the midsubstance UTF of the bo ne-patellar tendon-bone allograft to be 3,424 668 N. With a strain rate of 10% per se cond, the UTS was calculated to be 78.4 18.5 MPa with an average crosssectional area (CSA) of 45.0 1.5 mm2. For this 17

PAGE 18

study, the entire patellar t endon was mechanically tested. When the patellar tendon is used as an autograft, only the cent ral third of this tendon is utiliz ed. With a strain rate of 100%, Noyes et al.38 determined the UTF and UTS of the central third of the patellar tendon to be 2,900 260N and 58.3 6.1 MPa, respectively. Due to the bone attachment at each end of the graft, the patellar tendon can only be used as a single strand replacement graft. Pure Soft Tissue Grafts without Bone Blocks Pure soft tissue graft options include the anterior tibialis, peroneus longus, posterior tibialis, semitendinosus, and gracilis. Anterior tibialis Found on the lateral surface of the tibia, the AT muscle dorsiflexs and inverts the foot in conjunction with the peroneus brevis and posterior tibialis. The AT tendon passes from the AT muscle to the medial cuneiform and first metatarsal, running lateral to the tibia. It is both an antagonist and synergist of the posterior tibialis. Figure 1-2 shows a single AT compared to a native ACL. For a single-strand configurat ion with a strain rate of 6 mm/min, the UTF and CSA were 776.87 174 N and 24 4 mm2, respectively.14 Haut Donahue et al.15 determined the UTF of the AT in a singl e loop configuration to be 4,122 893N at a strain rate of 1.5 mm per second. Additiona lly, the CSA and UTS were 48.2 11.8 mm2 and 89.8 19.4 MPa. Further analysis, per formed by Pearsall et al.,16 showed the UTF and CSA in a double strand configurat ion was 3,412 N and 37.9 mm2 with a strain rate of 1 mm per second. 18

PAGE 19

Peroneus longus The PL muscle, found on the lateral surface of the fibula, plantar flexes and everts the foot in conjuncti on with the peroneus brevis and posterior tibialis. The PL tendon passes from the PL muscle to the medial cuneiform and first metatarsal, running posterior to the lateral malle olus. For the single-strand conf iguration, Bohnsack et al. 39 determined the UTF to be 1,342 135 N for a CSA of 22.4 5.3 mm2 at a strain rate of 10 mm per minute. The double-st rand configuration, perform ed by Pearsall et al.,16 determined the UTF to be 2, 483 N with a CSA of 36.6 mm2 at a strain rate of 1 mm per second. Figure 1-2 shows a single PL compared to a native ACL. Posterior tibialis The posterior tibialis muscle is responsible for the inversion of the foot and plantar flexion of the ankle. The posterior tibialis tendon has been mechanically tested in a single strand,14 double strand,16 and a single loop configuration.15 Almqvist et al.14 determined the UTF and CSA of a single strand to be 888.8 259 N and 23.92 4 mm2, respectively, at a rate of 6 mm/min. W hen tested as a double strand, the UTF and CSA were determined to be 3,391N and 47.7 mm2, respectively, at a strain rate of 1 mm per second.16 For a single loop configuration, Haut Donahue et al.15 determined the UTF and CSA to be 3,594 1,330 N and 41.9 17.3 mm2, respectively, with a strain rate of 1.5 mm per second. Semitendinosus and gracilis The hamstring tendons, semitendinosus and gr acilis, have been tested individually as a single and double-strand, as well was combined as a doubled looped configuration.19 The semitendinosus had a UTF and CSA of 1,060 227 N and 10.8 19

PAGE 20

2.2 mm2. The gracilis had a UTF and CSA of 837 138 N and 7.4 1.1 mm2. The UTS for the semitendinosus and gracilis was 99.3 14.9 and 113.1 18.1 MPa, respectively. When used as an autograft, a double looped gracilis and semitendinosus is the preferred method. When tested in a doubl e loop configuration, the gracilissemitendinosus graft has a UTF and CSA of 2,914 644 N and 41.6 6.5 mm2, respectively. Allograft Recovery Process A number of freeze-thaw cycles occur during the donor recovery process. According to AATB and RTI Biologics In c. standard operating procedures, the donor tissue site is recovered post mortem. Duri ng the recovery phase, the skin is removed from the ankle to the hip wit h a midline incision on the inside of the leg. Next, the femoral head is disarticulated from the acetabul ar cup. This structure, known as a leg en block, is frozen at a temperature of less than -40 C. While frozen, the leg en block is then shipped to an authorized processing site. Next, the processing phase requires the thawing of the leg en block. Once the leg en block has thawed, the tendons are obtained and excess muscle is removed. Normally, the tendons do not immediately undergo the sterilization process. Instead, they are again fr ozen until all tissue products from a specific donor are created. Once achieved, all products from a given donor including tendons are again thawed before the sterilization process. Following the sterilization technique, the tendons are again frozen and shipped to nearby hospitals. The last thaw cycle occurs prior to implantation. 20

PAGE 21

Graft Sizing The looped diameter (LD) sizing process is a similar process for both autograft and allograft replacement gr afts. After the replacement autograft tendon has been recovered, the excess muscle is removed. Allografts have already undergone this process during the processing procedure. Nex t, the length of the replacement graft is measured. A graft must have su fficient length to account for both the femoral and tibial tunnels as well as the ACL replacement. Ea ch end of the tendon is then whip stitched. Following this step, the graft is looped over a su ture at the midline of the graft, folding the graft in half. The graft is then tensioned on a tension t able with the suture from the midline on one side and the two whip stitched ends on the other. Next, each end of the now two-strand graft is whip stitched in order to prevent complications during fixation. The area of whipstitch is approximately 3 cm, representing t he length of both the femoral and tibial bone tunnel. Once this has occurred, the tendon is pulled through a LD sizing block to determine the proper diamet er for the femoral and tibial bone tunnels. Figure 1-3 displays this step. The LD sizing bl ock allows for proper sizing of the graft to the nearest 0.5 cm. Following this step, the graft is normally wrapped in a saline soaked surgical sponge or placed in a bowl of saline. Relaxation Data Ligaments and tendons are viscoelastic ma terials, displaying time-dependent behavior. One phenomenon of visc oelastic materials is st ress relaxation. Under a constant strain, the stress in the material will decrease. Once this strain is removed, the material will undergo a recovery phase. Duenwald et al.40 showed the stress relaxation and recovery rate of the por cine digital flexor tendon were well approximated by the power law. However, the tendon stress rela xation and recovery did not occur at the 21

PAGE 22

same rate. Recovery of a viscoelastic mate rial at an undefined rate makes modeling the behavior in which loads are applied without full recovery extremely difficult to predict.41 In order to accurately predict the mechani cal behavior of a viscoelastic material, the stress relaxation and recovery rates must be well defined. For this reason, stress relaxation and recovery rates of the ACL replacement graft are important for a successful ACL reconstruction. Introduction into Present Study Characterization of the AT and PL tendon as two-strand and four-strand grafts was performed in the present study. Figure 1-2 displays these two tendons as well as a native ACL. All test samples were siz ed according to a novel sizing technique, mimicking standard surgical procedure. Fo r the two-strand versus two-strand evaluation, the groups for comparison includ ed two-strand tendons with LD sizes that ranged from 7.00 mm to 7.75 mm (Group 1) and 8.00 mm to 8.75 mm (Group 2).Four groups were then defined: two LD groups created from AT tendons (AT1, AT2) and two LD groups created from PL tendons (PL1, PL2) For the two-strand versus four-strand evaluation, the AT two-str and graft (AT3) was compared to the PL four-strand graft (PL4). The AT3 group had an LD size ranging from 9.00 mm to 9.75 mm. The PL4 group had an LD size ranging from 10.00 mm to 10.75 mm. The purpose of this study was to examine t he tensile and viscoelastic properties of two common ACL replacement tendons as a f unction of surgical size and number of strands. The properties derived from this test include the relaxation rate, recovery rate, UTF, UTS, strain at failure, ext ension at failure, and Youngs modulus. 22

PAGE 23

0 20 40 60 80 100 120 140 UTS (MPa)Figure 1-1. Average UTS values for common ACL replacement grafts. 23

PAGE 24

Figure 1-2. Comparison of human tissue. A) AT tendon. B) PL tendon. C)ACL. Photo courtesy of the author. 24

PAGE 25

25 Figure 1-3. Sizing block for determining LD size. Photo courtesy of the author.

PAGE 26

CHAPTER 2 MATERIALS AND METHODS Tissue Allocation Aseptic donor tissue was obtained through RT I Biologics, Inc. For the two-strand versus two-strand evaluation, 108 AT tendons (50.49 14.94 yrs, 47 males, 61 females) and 114 PL tendons (45.61 16.64 yrs, 78 males, 36 females) were utilized. For the four-strand versus two-strand eval uation, an additional 66 PL tendons (52.17 15.44 yrs, 18 males, 48 females) formi ng 33 four-strand samples and 26 AT tendons (51.15 15.16 yrs, 26 males, 0 females) were utilized. In orde r to confirm that the novel sizing technique was not damaging the tendons, 28 PL samples from 7 donors (49.43 10.60 yrs, 6 males, 1 female) were utilized. All donor tissue was recovered according to the AATB standards for tissue banking. After the recovery process was complete, all donor tissue was frozen at -80 C until the day of the sizing test. Following the sizing test, all donor tissue was frozen at -80 C until the day of the mechanical test. Sizing Test Tendon Preparation On the day of the sizing test, each donor tissue was thawed at room temperature in an airtight bag to reduce the risk of dehy dration for a minimum of 30 minutes. Once the tendon had completely thawed, it was placed in a 0.9% saline bath at room temperature for 30 minutes to mimic standard rehydration techniques in the operating room. After this 30 minute rehydration step was completed, the midline for each tendon was measured at exactly 10 cm from the distal end of each tendon. 26

PAGE 27

Test Method An electromechanical testing system (Instron 5865, Instron Corporation, Norwood, CA) was utilized to perform the sizi ng test. Bluehill 2 software recorded the time, extension, and force applied to the sample during the test. The sizing test included a 150 mm pull through a custom sizing block at a rate of 2.5 mm per second. With a diameter range from 5 mm to 13 mm in 0.25 mm increments, the sizing block identified the LD of the graft. Each tendon was looped around the suture at the midline with the midline being the first portion of the tendon to enter the sizing block. Figure 2-1 illustrates the sizing test configuration. For each tendon, the force experienced as the tendon was pulled through the sizing block was measured and the maximum force was recorded. For a given LD, a surgeon will drill the necessary bone tunnel size for the ACL reconstruction process. This measured force is the force necessary for the surgeon to pull the graft through that particular sized bone tunnel. If the force is too small, the graft will not sit securely before the fixation is set. If this forc e is too large, the potential for damaging the tendon is high. In order to determine the acceptable r ange for the maximum force experienced while sizing the tendon, three orthopaedic surgeons from The Orthopedic Institute (Gainesville, FL) pulled several different si zed tendons through the custom sizing block. Based on the force readings and their comments, a force range of 70 N to 200 N was determined to be the correct force necessary to size a tendon. If the maximum force reached at least 70 N, the official size for that tendon was defined by the specific sizer and the tendon was stored at -80 C. If the maximum force did not reach 70 N, the tendon was tested again at the next smallest size. 27

PAGE 28

For the four-stranded grafts, each tendon wa s measured individually first and stored at -80C. Fo llowing the second thaw ing and rehydration process, two tendons were looped around the suture at each midl ine and underwent another sizing test. The LD size was determined when the maximum load reached at least 70 N. Mechanical Test Tendon Preparation On the day of the sizing test, each donor tissue was thawed at room temperature in an airtight bag to reduce the risk of dehydration. Once the tendon had completely thawed, the CSA of the tendon was measured at six equally spaced locations by measuring both the major and minor ax is. The first three major and minor measurements on the distal side of the midline determined the CSA for the distal end of the tendon while the second three major and mi nor measurements on the proximal side of the midline determined the CSA for the proximal end of the tendon. If the ratio of major axis to minor axis was greater than 2. 65 for either side, a rectangular shape was assumed. Otherwise, an elli ptical shape was assumed. All measurements were performed using Mitutoyo IP67 digital caliper s. Once these measurements were made, the tendon was loaded on the mechanical testing fixture. Test Fixture The two-strand and four-strand mechanical testing fixture was designed to have each strand of tendon tested at 37C. Whether a two-strand or a four-strand, the graft was looped around a 6 mm cross pin located on the top grip Figure 2-2 displays the cross pin location with respect to the top grip The free ends of each graft were clamped in the bottom grip exactly 35 mm below the top grip. This 35 mm gage length was used to symbolize the native ACL length, which ranges between 22 mm to 41 mm.8 A 28

PAGE 29

pretension of approximately 50 N was applied in order to allow each graft to slide across the 6 mm cross pin in order to reach an equ ilibrium force for each strand. In order to keep the 35 mm gage length at body temper ature, a warm-water jacket encapsulated the test sample between the top and botto m grip. Water was circulated through the jacket tubes once the test sample was loaded. The top grip was t hen clamped. In order to prevent slipping or teari ng at the tendon grip interface, the top and bottom grips were frozen. Dry ice was inserted into the top and bottom ice holders, which surround both the top and bottom grips, respectively. Th is method of freezing each grip has been successfully utilized in previous studies.15,16,19 Figure 2-3 displays the mechanical grip configuration without the t op and bottom ice holders, while Figure 2-4 displays the complete mechanical test fixtur e. In order to completely fr eeze each grip, the test began approximately 18 minutes after the dry ice was added. Test Method A servohydraulic testing system (MTS 858 Bionix Test System, MTS, Eden Prairie, Minnesota) was utilized to perform the mechanical test. Multiworks software recorded the time, extension, and force appli ed to the sample. Using this software, a testing scheme was programmed (Figure 2-5) A 90 N load was applied for 60 seconds in order to precondition the tendon. Following t he preconditioning, a 6% strain relaxation step was applied at a rate of 2 mm per second and remained for 100 seconds. Next, a 3% strain recovery step was applied at 2 mm per second and remained for 100 seconds. Following the recovery step, a second 6% strain relaxation step was applied at a rate of 2 mm per second and held for 100 seconds. These three strain phases were considered the viscoelastic portion of the mechanical test. Following the viscoelastic portion a cyclic load consisting of a sinusoida l wave from 50 N to 250 N at a frequency 29

PAGE 30

of 1 Hz for 100 cycles was applie d to the test sample. The final portion of the test consisted of a load-to-failure. This was done at a strain rate of 100% per second until destructive failure occurred. The properties derived from this test included the relaxation rate from the two relaxation phases, the recove ry rate from the recovery phase, Youngs modulus during the cyclic loading, UTF, UTS, strain at failure, extension at failure, and Youngs modulus during the l oad-to-failure portion. This testing protocol was designed to mimic a normal exercise immediately following an ACL reconstruction. The 90 N load for 60 seconds was found to be the normal pretension applied during the surgical procedure. Following this step, the two 6% strain phases mimicked a 100 second stretch sequence with a 3% strain recovery for 100 seconds between the two phases. Recovery at 3% ensured the tendon did not become slack. Additionally, a 6% strain wa s used in order to ensure adequate recovery at a nonzero strain without causing damage.40 Once the stretch sequence had completed, the cyclic loading portion char acterized the normal jogging motion for a human. The load-to-failure portion mimics the rate with which mid-substance ACL failure occurs.26 Sizing Damage Test The purpose of the sizing damage test wa s to determine if mechanical damage to the tendon occurred during the novel sizing technique. For each donor, both PL tendons were obtained. Each tendon was cut in half. T he distal side of the first tendon from a donor was sized at a maximum force less than 20 N. The proximal side of that same tendon was sized at a maximum force gr eater than 70 N. Fo r the second tendon belonging to the same donor, the proximal side of the te ndon was sized at a maximum force less than 20 N and the distal side of the tendon was sized at a maximum force 30

PAGE 31

greater than 70 N. Tendons with a maximum fo rce of less than 20 N were classified in the Weak Pull group, while tendons with a maximum force greater than 70 N were classified in the Strong Pull group. A fter the test sample was sized, it was mechanically tested as a single strand graft with the midline for the sizing test being the center of the test sample. The viscoelastic po rtion for this test me thod included three 90 N holds rather than three strain phases. Additionally, the load-to-failure rate during the final portion occurred at a rate of 2.4% per second. The cyclic loading portion remained the same. Statistical Methods Sample Size Calculations Minitab 15 was utilized to perform the sample size calculations. For all sample size calculations, the standard deviation values for each parameter were obtained from previous data and a power of 0.9 was assum ed. For the two-strand versus two-strand evaluation, a four level oneway ANOVA was performed. These four levels included the AT1, AT2, PL1, and PL2 groups. For the twostrand versus four-strand evaluation, a two sample t-test was performed. The AT 3 group was compared to the PL4. In order to determine practical differenc e values, previous literature was used. According to Noyes et al.,38 the ACL experiences approximately 454 N of force during normal life activities. For this reason, a prac tical difference for the UTF was set at 450 N. In order to determine the practical differ ence for the UTS, the practical difference for UTF was divided by 24 mm2, the average CSA for a single strand AT determined by Almqvist et al.14 This calculation determined a practical difference for UTS at 18.75 MPa. Duthon et al.8 determined the elongation of the AM bundle of the ACL to be 4 mm when flexed to 90. For a 35 mm ACL, an elongation of 4 mm would equal a strain of 31

PAGE 32

11.4%, which was used as the practical diffe rence for failure strain. During normal life activities, the knee will not be flexed to 90. However, the knee is flexed to 90 during more strenuous exercise. Noyes et al.38 determined the ACL experienced approximately 1000 N during strenuous exercise. In order to determine the practical difference for Youngs modulus, Hookes law was calcul ated using 1000 N as the force, 24 mm2 as the CSA, and 11.4% as the strain. This prac tical difference value for Youngs modulus was calculated to be 365 MPa. Sample size calculations were taken fo r each parameter. The largest calculated sample size was determined to be the necessa ry sample size for each evaluation. With the sample size set, a practical differenc e of one standard deviation was set as a practical difference for CSA and the viscoelasti c properties, as no previous data outlines what might make a practical difference for these values. Statistical Analysis For the two-strand versus two-strand eval uation, a Bartletts f-test was performed to determine homogeneity of variance. If homogenous variances were found, a one-way ANOVA was performed to determine if there were any differences between groups. For a parameter in which a diffe rence was found, Tukeys mult iple comparison adjusted ttest for pairwise comparisons was conducted to determine which groups were statically different from each other. If heterogeneous variances were found, a Browne-Forsythe ftest was performed to determine if there were any differences. For the two-strand versus four-strand ev aluation, an f-test was performed to determine homogeneity of variance. Next, a two-sample t-test was performed to determine if there was a differenc e between the AT3 and PL4 groups. 32

PAGE 33

For the sizing damage test, a paired t-test was performed to determine if there was a difference between the weak pull and strong pull groups. 33

PAGE 34

Figure 2-1. Sizing test fixture. Photo courtesy of the author. Figure 2-2. Mechanical test grip Photo courtesy of the author. 34

PAGE 35

Figure 2-3. Mechanical test fixture. Photo c ourtesy of the author. Figure 2-4. Mechanical test fixture with top and bottom ic e holders. Photo courtesy of the author. 35

PAGE 36

36 0 1000 2000 3000 4000 50000 100 200 300 400 500Force (N)Time (s)Viscoelastic Portion Cyclic Loading Portion Load-to-Failure Portion Precondition PortionFigure 2-5. Test method sc hematic for mechanical test.

PAGE 37

CHAPTER 3 RESULTS Mechanical Test A statistical comparison of the mechanical results for t he load-to-failure test was performed for two separate evaluations. For the first evaluation, both the PL and AT were compared as two-strand grafts. The groups for the comparison included the AT1, AT2, PL1, and PL2. For the second evaluat ion, the PL4 group wa s compared to the AT3 group. For numerical results, the m ean followed by standar d deviation is the format. Biomechanical Comparison of TwoStrand AT versus Two-Strand PL All p-values for this evaluation can be found in Table 3-1. A statistical comparison of the mechanical results of the two-strand grafts determined a statistical difference for the UTF and CSA values. A statistical differ ence was found for the UTS values between groups and the Youngs modulus during loading and unloading; however, these differences were not of practical importance. Additionally, a statistical but not practical difference was found for the initial stress, the final stress, and the calculated A value for both of the relaxation phases as well as the recovery phase. Finally, a statistical but not practical difference for the n value during the recovery phase was found. The following values were recorded for each parameter. Viscoelastic Portion First relaxation phase properties The initial stress values during the firs t relaxation phase were 46.52 8.59 MPa, 43.63 6.69 MPa, 40.37 6.32 MPa, and 39. 284 7.61 MPa for AT1, AT2, PL1, and PL2, respectively. The final stress values during the same phase were 34.63 6.87 37

PAGE 38

MPa, 31.99 5.21 MPa, 29.47 5.66 MPa, and 28.84 6.08 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA determined the AT1 group had a 6.15 MPa higher initial stress than t he PL1 group and a 7.24 MPa higher initial stress than the PL2 group during the first relaxation phase. A dditionally, a one-way ANOVA determined the AT1 group had a 5.16 MPa higher final stress than the PL1 group and a 5.79 MPa higher final stress t han the PL2 group. When set to a power law equation, the A va lues from the relaxation curve were 725.1 126.6 MPa, 674.0 98.6 MPa, 624. 7 102.5 MPa, and 608.5 117.1 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA determined the AT1 group had a 100.5 MPa higher A value than the PL1 group and a 116.6 MPa higher A value than the PL2 group. The n values from the relaxation curve during the same phase were -0.056 0.010, -0.056 0.019, -0.061 0.013, and -0.059 0.011 fo r AT1, AT2, PL1, and PL2, respectively. A Brown-Forsythe f-test for the n values determined no statistical difference between any groups (p=0.72264). Recovery phase properties The initial stress values during the reco very phase were 3.82 2.30 MPa, 2.63 1.67 MPa, 2.86 1.90 MPa, and 2.26 1.90 MPa for AT 1, AT2, PL1, and PL2, respectively. The final stress values duri ng this phase were 7.54 2.92 MPa, 6.35 2.01 MPa, 6.22 2.29 MPa, and 5.58 2.41 MPa for AT 1, AT2, PL1, and PL2, respectively. A one-way ANOVA determi ned the AT1 group had a 1.57 MPa higher initial stress than the PL2 group during the recovery phase. Additionally, a one-way ANOVA determined the AT1 group had a 1.96 MP a higher final stress than the PL2 group. 38

PAGE 39

The A values for the power law curve we re 172.9 85.5 MPa, 133.7 59.0 MPa, 131.4 66.5 MPa, and 109.3 67.6 MPa for AT 1, AT2, PL1, and PL2, respectively. A one-way ANOVA for the A values determined the AT1 group had a 63.66 MPa higher A value than the PL2 group. The n values for this same curve were 0.0825 0.0383, 0.100 0.042, 0.086 0.040, and 0.107 0.051 for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA for the n values determined the AT1 group had a 0.0249 higher n value than the PL2 group. Second relaxation phase properties The initial stress values during the second relaxation phase were 40.98 7.41 MPa, 38.49 5.79 MPa, 35.29 6.13 MPa, and 34.81 6.70 MPa for AT1, AT2, PL1, and PL2, respectively. The final stress values during this phase were 34.78 6.98 MPa, 32.38 5.28 MPa, 29.79 5.61 MPa, and 29. 31 6.11 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA determined the AT1 group had a 5.69 MPa higher initial stress than the PL1 gr oup and a 6.17 MPa higher initial stress than the PL2 group during the second relaxation phase. Additi onally, a one-way ANOVA for the final stress determined the AT1 group had a 4.99 MPa higher final stress than the PL1 group and a 5.47 MPa higher stress than the PL2 group. The A values for this phase were 632.4 119.1 MPa, 587.6 91.1 MPa, 539.9 96.8 MPa, and 529.4 107.0 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA determined the AT1 group had a 92.5 MPa higher A value than the PL1 group and a 102.9 MPa higher A value than the PL2 group. The n values during the second stress relaxation phase were -0.0208 0.0057, -0.0218 0.0043, -0.0221 0.0057, and -0.0210 0.0052 for AT1, AT2, PL1, and PL2, respectively. A Brown-Forsythe f-test for the n values determined no statistical difference between any groups (p=0.73462). 39

PAGE 40

Cyclic Loading Portion Average Youngs modulus during the cyclic portion The average Youngs modulus values for the loading portion across the 100 cycles were 991.1 232.1 MPa, 914.3 205.0 MPa, 890.0 203.5 MPa, and 877.3 215.2 MPa for AT1, AT2, PL1, and PL2, respecti vely. The average Youngs modulus values for the unloading portion across the 100 cycles were 1,037.3 217.1 MPa, 958.7 190.5 MPa, 937.5 192.4 MPa, and 924.4 199.8 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA for the aver age Youngs modulus fo r both the loading and unloading portion during cyclic loading det ermined the AT1 had a 113.8 MPa higher Youngs modulus during the loaded phase and 112.9 MPa higher Youngs modulus during the unloaded phase compared to the PL2. Load to Failure Portion Ultimate tensile force The average UTF values were 3,879.9 512.2 N, 4,390.8 616.4 N, 3,519.1 453.4 N, and 4,263.7 524.2 N for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA for UTF determined the AT2 group had a 493.0 N higher UTF compared to AT1 and an 871.8N higher UTF compared to PL1. Additionally, the PL2 group had a 744.7 N higher UTF compared to the PL1 group. Ultimate tensile stress The average UTS values were 86.6 13.1 MPa, 82.0 14.5 MPa, 75.8 12.6 MPa, and 78.3 11.5 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA determined the AT1 group had a 10.82 MP a higher UTS than the PL1 group. 40

PAGE 41

Cross-sectional area The average CSA values were 45.39 4.75 mm2, 54.42 7.50 mm2, 46.91 4.96 mm2, and 54.88 4.40 mm2 for AT1, AT2, PL1, and PL2, respectively. A BrowneForsythe for CSA determined the AT2 group had a 9.02 mm2 higher CSA compared to the AT1 group and a 7.50 mm2 higher CSA compared to t he PL1 group. Additionally, a Browne-Forsythe for CSA determined the PL2 group had a 9.48 mm2 higher CSA compared to the AT1 group and a 7.96 mm2 higher CSA compared to PL1 group. Youngs modulus during the failure curve The average Youngs modulus during the failure curve values were 1,132.8 145.0 MPa, 1,084.9 163.3 MP a, 977.9 143.1 MPa, and 1,015.9 137.0 MPa for AT1, AT2, PL1, and PL2, respectively. A one-way ANOVA for Youngs modulus determined no statistical di fference between any groups. Strain at failure The average strain at failure values were 12.89 1.76 %, 13. 45 2.94 %, 14.25 3.16 %, and 13.85 2.52 % for AT1, AT2, PL1, and PL2, respectively. A BrowneForsythe for strain at failure determined no st atistical difference between any groups (p = 0.543). Extension at failure The extension at failure values were 3.52 0.61 mm, 3.67 0.94 mm, 3.92 1.04 mm N, and 3.81 0.83 mm for AT1, AT2, PL1, and PL2, respectively. A BrowneForsythe for the extension at failure dete rmined no statistical difference between any groups (p=0.543). 41

PAGE 42

Biomechanical Comparison of TwoStrand AT versus Four-Strand PL A statistical comparison of the mechanical results of the AT3 group and the PL4 group determined a statistical difference for the UTF and CSA values. A statistical difference was found for the UTS and the second relaxation phase n values; however, these differences were not of practical importance. The following values were recorded for each parameter. Viscoelastic Portion First relaxation phase properties The initial stress values during the first relaxation phase were 36.77 8.0 MPa and 36.95 7.1 MPa for AT3 and PL4, respectively. The final stress values during this phase were 26.86 6.41 MPa and 27.09 5.53 MPa for AT3 and PL4, respectively. The A values for the power law curve fit were 568 126 MPa, and 578 110 MPa for AT3 and PL4, respectively. The n values were -0.0538 0.0267 and -0.06006 0.00642 for AT3 and PL4, respectively. No statis tical difference was found for any of the parameters between these two samples. Recovery phase properties The initial stress values during the reco very phase were 1.71 1.37 MPa and 2.41 1.78 MPa for AT3 and PL4, respectively. T he final stress values for this same phase were 4.46 2.24 MPa and 5.35 2.29 MPa for AT3 and PL4, respectively. The A values were 92.1 55.1 MPa and 103.3 72.3 MPa for AT3 and PL4, respectively. The n values were 0.136 0.143 and 0.0996 0.0438 for AT3 and PL4, respectively. No statistical difference was found for any of the parameters between these two samples. 42

PAGE 43

Second relaxation phase properties The initial stress values during the second relaxation phase were 32.94 7.26 MPa and 32.65 6.06 MPa for AT3 and PL4, respectively. The final stress values during the second relaxation phase were 27. 58 6.38 MPa and 27.40 5.42 MPa for AT3 and PL4, respectively. The A values during the second relaxation phase were 494 111 MPa and 500 94.3 MPa for AT3 and PL4, respectively. The n values during the second relaxation phase were -0.02 0.00345 and -0.02258 0.00426 for AT3 and PL4, respectively. A two-sa mple t-test for the n val ues determined the PL4 had a 0.00258 higher absolute n value than the AT3. For the initial st ress, the final stress, and the A value, no statistical differ ence was found between these groups. Cyclic Loading Portion Average Youngs modulus during the cyclic portion The average Youngs modulus values for the loading portion during cyclic loading were 731.0 189.0 MPa and 771.0 174.0 MPa for AT3 and PL4, respectively. Additionally, the average Y oungs modulus values for th e unloading portion across the 100 cycles were 789.0 190 MPa and 801.0 169.0 MPa for AT3 and PL4, respectively. A two-sample t-test for the average Youngs modul us for the loading portion during cyclic loading and unloading determined no statis tical difference (p=0.410 and p=0.799, respectively). Load to Failure Portion Ultimate tensile force The average UTF values were 4,882 615 N and 5,543 616 N for AT3 and PL4, respectively. A two-sample t-test determined the PL4 group had a 661 N higher UTF than the AT3 group (p<<0.05). 43

PAGE 44

Ultimate tensile stress The average UTS values were 80.5 13.9 MPa and 73.6 12.3 MPa for AT3 and PL4, respectively. A two-sample t-te st determined the AT3 group had a 6.88 MPa higher UTS than the PL4 group (p=0.048). Cross-sectional area The average CSA values were 61.64 8.51 mm2 and 76.13 6.47 mm2 for AT3 and PL4, respectively. A two-sample t-te st determined the PL4 group had a 14.49 mm2 larger CSA than the AT3 group (p<<0.05). Youngs modulus during the failure curve The average Youngs modulus during the failure curve values were 1,041 146 MPa and 948 115 MPa for AT3 and PL4, respecti vely. A two-sample t-test determined the Youngs Modulus during the failure curv e for the AT3 group was 93.5 MPa higher than the PL4 group (p= 0.008). Strain at failure The average strain at failure values were 14.54 2.80 % and 13.34 2.19% for AT3 and PL4, respectively. A two-sample ttest for strain at failure determined no statistical difference (p=0.070). Extension at failure The average extension at failure values were 3.977 0.905 mm and 3.687 0.715 mm for AT3 and PL4, respectively. A two-sa mple t-test for extension at failure determined no statistical difference (p=0.175). Biomechanical Comparison of Weak Pull versus Strong Pull A statistical comparison of the mechanical results of the sizing damage test showed no statistical difference between eith er sizing groups for UTF, UTS, CSA, 44

PAGE 45

Youngs Modulus, and strain at failure. All p-values for this evaluation can be found in Table 3-3. Average values for each par ameter can be found in Table 3-4. Load to Failure Portion Ultimate tensile force The average UTF values were 2,384 408 N and 2,331 517 N for the strong pull group and weak pull group, respectively. Ultimate tensile stress The average UTS values were 107.9 35.2 MPa and 95.8 28.1 MPa for the strong pull group and weak pull group, respectively. Cross-sectional area The average CSA values were 26.67 10.45 mm2 and 24.80 9.95 mm2 for the strong pull group and weak pull group, respectively. Youngs modulus during the failure curve The average Youngs modulus during the failure curve values were 1260 461 MPa and 1107 395 MPa for the strong pull gr oup and weak pull group, respectively. Strain at failure The average strain at failure values were 13.84 2.78 % and 13.71 2.86% for the strong pull group and weak pull group, respectively. 45

PAGE 46

Table 3-1. Statistical analysis for the two-strand versus two-strand evaluation Parameter Bartlett's Test ANOVA/BrowneForsythe 1st Ralaxation Phase Initial Stress (MPa) > 0.1 < 0.001 1st Relaxation Phase Final Stress (MPa) > 0.1 < 0.001 1st Relaxation Phase A (MPa) > 0.1 < 0.001 1nd Relaxation Phase n < 0.001 > 0.1 Recovery Phase Initial Stress (MPa) > 0.1 < 0.001 Recovery Phase Final Stress (MPa) > 0.05 < 0.001 Recovery Phase A (MPa) > 0.1 < 0.001 Recovery Phase n > 0.1 < 0.05 2nd Relaxation Phase Initial Stress (MPa) > 0.1 < 0.001 2nd Relaxation Phase Final St ress (MPa) > 0.1 < 0.001 2nd Relaxation Phase A (MPa) > 0.1 < 0.001 2nd Relaxation Phase n < 0.001 > 0.1 Average Young's Modulus During Loading (MPa) > 0.1 < 0.05 Average Young's Modulus During Unloading (MPa) > 0.1 < 0.05 Ultimate Tensile Force (N) > 0.1 < 0.001 Ultimate Tensile Stress (MPa) > 0.1 < 0.001 Cross-sectional Area (mm2) < 0.001 < 0.001 Strain at Failure (%) < 0.001 > 0.1 Extension at Failure (mm) < 0.001 > 0.1 Young's Modulus During Failure (MPa) > 0.1 < 0.001 46

PAGE 47

Table 3-2. Statistical analysis for the two-strand versus four-strand evaluation Parameter Bartlett's Test ANOVA/BrowneForsythe 1st Relaxation Phase Initial Stress (MPa) >0.1 >0.1 1st Relaxation Phase Final Stress (MPa) >0.1 >0.1 1st Relaxation Phase A (MPa) >0.1 >0.1 1nd Relaxation Phase n <0.001 >0.1 Recovery Phase Initial Stress (MPa) >0.1 >0.1 Recovery Phase Final St ress (MPa) >0.1 >0.1 Recovery Phase A (MPa) >0.1 >0.1 Recovery Phase n <0.001 >0.1 2nd Relaxation Phase Initial Stress (MPa) >0.1 >0.1 2nd Relaxation Phase Final Stress (MPa) >0.1 >0.1 2nd Relaxation Phase A (MPa) >0.1 >0.1 2nd Relaxation Phase n >0.1 <0.05 Average Young's Modulus During Loading (MPa) >0.1 >0.1 Average Young's Modulus During Unloading (MPa) >0.1 >0.1 UTF (N) >0.1 <0.001 UTS (MPa) >0.1 <0.05 CSA (mm2) >0.1 <0.001 Strain at Failure (%) >0.1 <0.1 Extension at Failure (mm) >0.1 >0.1 Young's Modulus During Failure (MPa) >0.1 <0.01 Table 3-3. Statistical analysis for the sizing damage test Parameter Paired ttest Ultimate Tensile Force (N) 0.71 Ultimate Tensile Stress (MPa) 0.455 Cross-sectional Area (mm2) 0.710 Strain at Failure (%) 0.915 Young's Modulus During Failure (MPa) 0.469 Table 3-4. Mechanical result s for the sizing damage test Tendon Group UTF (N) UTS (MPa) CSA (mm2) Strain at Failure (%) Young's Modulus During Failure (MPa) Strong Pull 2384 107.85.21 24.80 9.95 13.84.78 1260 Weak Pull 2331 95.80.07 26.67. 45 13.71.85 1107 47

PAGE 48

Table 3-5. Two-strand versus twostrand first relaxation results Group (Loop Diameter Size) Initial Stress (MPa) Final Stress (MPa) A (MPa) n AT1 (7.00-7.75) 46.52.59 34. 63.87 725.1.6 -0.056.010 AT2 (8.00-8.75) 43.63.69 31. 99.21 674.0.6 -0.056.019 PL1 (7.00-7.75) 40.37.32 29. 47.66 624.7.5 -0.061.013 PL2 (8.00-8.75) 39.28.81 28. 84.08 608.5.1 -0.059.011 Table 3-6. Two-strand versus two-strand recovery results Group (Loop Diameter Size) Initial Stress (MPa) Final Stress (MPa) A (MPa) n AT1 (7.00-7.75) 3.82.30 7.54.92 172.9.5 0.083.038 AT2 (8.00-8.75) 2.63.67 6.35.01 133.7.0 0.100.042 PL1 (7.00-7.75) 2.86.90 6.22.29 131.4.5 0.086.040 PL2 (8.00-8.75) 2.26.90 5.58.41 109.3.6 0.107.051 Table 3-7. Two-strand versus tw o-strand second relaxation results Group (Loop Diameter Size) Initial Stress (MPa) Final Stress (MPa) A (MPa) n AT1 (7.00-7.75) 40.98.41 34.78.98 632.4 119.1 -0.021.006 AT2 (8.00-8.75) 38.49.79 32. 38.28 587.6.1 -0.022.004 PL1 (7.00-7.75) 35.29.13 29. 79.61 539.9.8 -0.022.006 PL2 (8.00-8.75) 34.81.70 29.31.11 529.4 107.0 -0.021.005 Table 3-8. Two-strand versus tw o-strand cyclic loading results Group (Loop Diameter Size) Average Young's Modulus During Loading (MPa) Average Young's Modulus During Unloading (MPa) AT1 (7.00-7.75) 991. 1.1 1037.3.1 AT2 (8.00-8.75) 914.3.0 958.7.5 PL1 (7.00-7.75) 890.0.5 937.5.4 PL2 (8.00-8.75) 877.3.2 924.4.8 Table 3-9. Two-strand versus tw o-strand failure curve results Group (Loop Diameter Size) UTF (N) UTS (MPa) CSA (mm2) Young's Modulus During Failure (MPa) AT1 (7.00-7.75) 3880 86. 6.1 45.39. 75 1132.8.0 AT2 (8.00-8.75) 4391 82. 0.5 54.42. 50 1084.9.3 PL1 (7.00-7.75) 3519 75. 8.6 46.91.96 977.9.1 PL2 (8.00-8.75) 4264 78. 3.5 54.88. 40 1015.9.0 48

PAGE 49

Table 3-10. Two-strand versus two-strand fa ilure curve strain and extension results Group (Loop Diameter Size) Strain at Failure (%) Extension at Failure (mm) AT1 (7.00-7.75) 12.89.76 3.52.61 AT2 (8.00-8.75) 13.45.94 3.67.94 PL1 (7.00-7.75) 14.25.16 3.92.04 PL2 (8.00-8.75) 13.85.52 3.81.83 Table 3-11. Two-strand AT versus fou r-strand PL first re laxation results Group (Loop Diameter Size) Initial Stress (MPa) Final Stress (MPa) A (MPa) n AT3 (9.00-9.75) 36.77.00 26. 86.41 567.9.9 -0.054.027 PL4 (10.00-10.75) 36.95.10 27. 09.53 577.9.4 -0.060.006 Table 3-12. Two-strand AT versus f our-strand PL reco very results Group (Loop Diameter Size) Initial Stress (MPa) Final Stress (MPa) A (MPa) n AT (9.00-9.75) 1.71.37 4.46.24 92.1.1 0.136.143 PL (10.00-10.75) 2.41.78 5.35.29 103.3.3 0.100.044 Table 3-13. Two-strand AT versus four-s trand PL second re laxation results Group (Loop Diameter Size) Initial Stress (MPa) Final Stress (MPa) A (MPa) n AT3 (9.00-9.75) 32.94.26 27. 58.38 493.8.9 -0.020.003 PL4 (10.00-10.75) 32.65.06 27. 40.52 500.0.3 -0.023.004 Table 3-14. Two-strand AT versus f our-strand PL cyclic loading results Group (Loop Diameter Size) Average Young's Modulus During Loading (MPa) Average Young's Modulus During Unloading (MPa) AT3 (9.00-9.75) 731.0 789.0 PL4 (10.00-10.75) 771.0 801.0 Table 3-15. Two-strand AT versus fou r-strand PL failure curve results Group (Loop Diameter Size) UTF (N) UTS (MPa) CSA (mm2) Young's Modulus During Failure (MPa) AT3 (9.00-9.75) 4882 80.5 13.9 61.64.51 1041.3.0 PL4 (10.00-10.75) 5543 73. 6.3 76.13.47 947.8.4 49

PAGE 50

50 Table 3-16. Two-strand AT versus four-str and PL failure curve strain and extension results Group (Loop Diameter Size) Strain at Failure (%) Extension at Failure (mm) AT3 (9.00-9.75) 14.54.80 3.98.91 PL4 (10.00-10.75) 13.45.94 3.69.72

PAGE 51

CHAPTER 4 DISCUSSION This study showed limited difference between the PL and AT tendons as a function of LD size and number of strands. The two-strand versus two-strand evaluation showed a direct comparison of the AT and PL across two LD sizes. The four-strand versus two-strand evaluation performed a dire ct comparison of the two-strand AT graft with a 9.00 to 9.75 mm LD versus a four-s trand PL graft with a 10.00 to 10.75 mm LD size. For both evaluations, no practical diffe rence was found between the UTS, strain, stress relaxation properties, recovery properti es, and Youngs modulus at various strain rates. The similarities in the mechanical characteristics of these ACL replacement graft options may be a result of both being lowe r-leg tendons, playing an active role in ambulatory motion and stability. Tensile stress is the amount of tensile force applied to a body per unit area. The UTS of a material is the maximum allowa ble stress a material can withstand before destructive failure occurs. The micros tructure compositio n for a tendon type, responsible for the tendons me chanical characteristics, shoul d be relatively consistent independent of size. Since UT S scales the UTF as a function of CSA, there should not be a variation in UTS for different LD size s for the same tendon type. The data in this study supports this theory. Additionally, no practical difference between PL and AT tendons was observed. Strain is the amount of elongation that occurs when a load is applied. Ultimate strain is the strain at which failure occurs A replacement implant for ACL reconstruction that exhibits too little ultima te strain would fail due to the elongation of the graft during flexion of the knee. The strain at failure for all groups in both evaluations was much 51

PAGE 52

greater than the physiological strains ex perienced by the knee during all activities.40 Additionally, no practical difference in stra in at failure was found between any groups for both evaluations. In order to provide ambulat ory motion, muscles in the legs contract. Tendons are responsible for transmitting this contraction force to the bone insertion site. If the tendon exhibits stress relaxation, more energy is necessary to co ntract the muscle further in order to maintain the same bone displacemen t. For this reason, a material with a slow relaxation rate is preferred. Additionally, a fast recovery rate of lower leg tendons is expected, as quick, repetitive loading is common during such activities as walking and running. Viscoelastic properties showed no practical difference between any groups for both evaluations. The stress relaxation and reco very curves were well estimated by the power law. Both stress relaxa tion rates proceeded at a much slower absolute rate when compared to the recovery rate for all groups The third relaxation rate was the slowest relaxation rate of all three phases. This may show that additional crimp is present at a higher strain that was not removed duri ng the loading at 50 N during the 18 minute freeze and at 100 N for 60 seconds. A diffe rence of approximately 5 MPa was found during a comparison of the initial stress after the 1st relaxation phase versus the 2nd relaxation phase, while a difference of less than 1 MPa was found during a comparison of the final stress after the 1st relaxation phase versus the 2nd relaxation phase. This may show that each human tendon relaxes to a given stress at a specific strain, independent of previous work. 52

PAGE 53

There were no practical differences in Youngs modulus at each strain rate between groups for both evaluations. Howe ver, a difference between the Youngs modulus during the cyclic loading and failure curve was observed. For a viscoelastic material, parameters such as temperature,34,35 hydration state,31-33 and strain rate impact material properties26-30 impact the mechanical properties of the test sample. Temperature and hydration level during testing were held c onstant for all test samples; however, the strain rate of the cyclic loading differed from the strain rate of the load to failure. The cyclic loading was designed to rep licate normal jogging. For this reason, the sinusoidal loading pattern was applied, with t he fastest strain rate observed near the midline of the sinusoidal wave and the sl owest strain rate near both the maximum and minimum forces. With a frequency of 1 Hz, t he average strain rate was approximately 1% per second for each loading and unloading cycle. A strain rate of 1% per second maintained a linear stress strain curve for each loading and unloading cycling. For the failure curve, a strain rate of 100% per second was used. Lee et al.26 determined a 100% per second strain rate to be the correct strain rate in or der to achieve a midsubstance ACL failure, which was replicated in this study. For both the two-strand versus two-str and evaluation and the four-strand versus two-strand evaluation, the only practical differences were the CSA and UTF. The first practical difference for both evaluations was CSA. With a larger LD size, there is an increase in the area available for the LD t endon to pass. In Table 4-1, the theoretical CSA was calculated utilizing the average LD size of all the tendons in each group as the diameter for the calculation. The theoretical CSA calculat ions support the practical difference in CSA between LD sizes. For the two-strand versus two-strand evaluations, 53

PAGE 54

no difference between the ac tual CSA and theoretical CSA was observed for each group. The difference between actual CSA and theoretical CSA increased for both tendons in the four-strand versus two-strand evaluation. For the PL4 group, the actual CSA values were smaller than the theoretical CSA values, showing that these grafts did not pack as efficiently through the LD sizing block as the two-strand grafts. However, this may be a result of size since both of the AT3 and PL4 had a small increase in the CSA difference. If this is true, it might indicate that the use of a sm aller LD graft may have the ability to pack more area into a given bone tunnel. Further analysis of larger LD tendons is necessary to make such conclusions. The second practical difference was the UTF for both evaluations. For both evaluations, the tendon groups with a large LD size displayed superior failure loads. For each tendon type, the microstr ucture should be relatively consistent independent of size. Therefore, each tendon type should maintain similar material properties across sizes, such as UTS and Youngs modulus. As previously mentioned, the UTS is equal to the UTF divided by the CSA. If the UTS re mained the same and the CSA is increased, there will be an increase in the UTF, which wa s confirmed in this study. Although there is a practical difference between UTF acro ss tendon LD size, all tendon groups tested at least twice as strong as the native ACL. This supports that all tendon graft options have adequate mechanical strength to serve as replacement grafts. As expected, the force range used to sort the tendons into groups did not have any impact on the mechanical properties. Each orthopaedic surgeon was given an opportunity to pull tendons through the LD sizing block. After each tendon, the surgeon was asked if this was an acceptable amount of force to pull th rough the bone tunnel 54

PAGE 55

during an ACL reconstruction. The findings in this study support the common surgical method for sizing ACL replacement implants. 55

PAGE 56

56 Table 4-1. Theoretical CSA co mpared to actual measured CSA Group Average LD (mm) Average LD CSA (mm2) Measured CSA (mm2) Difference between LD and Measured CSA (mm2) LD Avg Pull Force (N) AT1 7.53 44.53 45.39 -0.86 121.89 AT2 8.43 55.81 54.42 1.39 109.22 PL1 7.47 43.83 46.91 -3.09 102.93 PL2 8.39 55.29 54.88 0.41 98.95 AT3 9.3 67.93 61.6 6.33 112.46 PL4 10.5 86.59 76.13 10.46 95.54

PAGE 57

CHAPTER 5 TEST COMPARISONS: LIMITATIONS AND FUTURE RESEARCH Limitations Equivalent grip methods similar to the gr ip method used in this current study have been adopted by several researchers.15,16,19 This grip method allowed for no slippage or inadvertent destruction of the tendon sample dur ing testing. Since each end is frozen at nearly -60 C, the use of the warm water jacket is necessary to keep the tendon sample at a physiologically rele vant temperature of 37 C. With the application of the heat on the external surface of the t endon, non-uniform temperature di stribution may have occurred between the external surface and the center of each str and. Since temperature has been well documented to affect the viscoelastic properties, comparing the current data to previous data performed at room temperat ure may explain the slight variation in values. Another factor to consider when usi ng the current studys grip method was hydration. Betsch and Baer found that dehydrated tendons di splayed a higher stiffness when compared to hydrated tendons.42 Additionally, the amount of load relaxation decreased as the water content decreas ed in strips of patellar tendons.43 Also, tensile loading has been shown to increase the apparent diffusion coefficient, further reducing the hydration of the tendon test sample.32 Once the test sample was loaded, no additional hydration was provided. Dehydr ation during the 18 minute freezing and the tensile nature of the te st procedure may have an impact on the mechanical characteristics The additional saline soak cycle for the four-strand grafts was examined. It was necessary to size the tendons individually firs t in order to know what LD size tendons 57

PAGE 58

combined to form the given four-strand LD. The second thirty minute saline soak was utilized in order to ensure the four-strand tendon had the equivalent hydration. Post experimental data estimated that each thirty minute saline soak cycle caused approximately a 5% swell. A 5% swell reduction assumption for the CSA for the fourstrand tendons did not change any of the statistical results for the four-strand versus two-strand evaluation. The final limitation of the cu rrent study was the inability to compare the relaxation and recovery data to human ACL data. The current study grip method displayed consistent gripping of pure soft tissue gr afts. Due to the short length and bone ends of the ACL, several revisions of the current gripping method did not display consistent proper failure modes. In order to determine practical differences between replacement graft groups, native ACL data is necessary. Future Research Future research includes more testing of two-strand and four-strand AT and PL for sizes not included in this study. Additionally other pure soft tissue grafts used for ACL reconstruction procedures, such as semitendinosus and gracilis, will be included in future research. In order to achieve more re levant practical differences, the native ACL will be mechanically tested in a similar fashion. In order to predict the behavior of the ACL and soft ti ssue replacement grafts, a robust model requires a better understanding of the viscoelastic properties of each tendon. A testing procedure for human tendons with more relaxation and recovery phases at different strain levels, such as that performed by Duenwald et al.40 on porcine digital flexor tendons, woul d provide a better understandi ng of the relaxation and recovery rates at different strain levels. 58

PAGE 59

59 The difference between actual CSA and theo retical CSA increased for both tendon groups in the four-strand versus two-strand evaluation. Further research with fourstrand and larger LD two-strand grafts is necessary to determine any practical difference between this difference growth as a function of LD size. This correlation may show that small LD tendons have a faster bone tunnel healing rate compared to large LD tendons due to their ability to pack more ma terial into the bone tunnel. If true, this research may show that a small LD graft has a more rapid healing process while still having the appropriate mechanica l strength, making it a bette r ACL replacement option. Additionally, this would suggest the use of tw o femoral and two tibial small tunnels for a four-strand reconstruction superior to a si ngle large femoral and tibial tunnel.

PAGE 60

REFERENCES LIST 1. Sun K, Tian SQ, Zhang JH, et al. Ante rior cruciate ligament reconstruction with bone-patellar tendon-bone autograft versus allograft. Arthroscopy 2009;25:750759. 2. Woo SL, Abramowitch SD, Kilger R, et al. Biomechanics of knee ligaments: injury, healing, and repair. J Biomech 2006;39:1-20. 3. Woo SL, Wu C, Dede O, et al. Biom echanics and anterior cruciate ligament reconstruction. J Orthop Surg Res 2006;1:2. 4. Cooper MT, Kaeding C. Comparison of the hospital cost of autograft versus allograft soft-tissue anterior cruciate ligament reconstructions. Arthroscopy 2010;26:1478-1482. 5. Loh JC, Fukuda Y, Tsuda E, et al. Knee stability and graft function following anterior cruciate ligament reconstruction: comparison between 11 oclock and 10 oclock femoral tunnel placement. Arthroscopy 2003;19:297-304. 6. Branch TP, Siebold R, Freedberg HI, et al. Double-bundle ACL reconstruction demonstrated superior clinical stability to single-bundle ACL reconstruction; a matched-pairs analysis of instrumented test s of tibial anteri or translation and internal rotation laxity. Knee Surg Sports Traumatol Arthrosc 2011;19:432-440. 7. Seon JK, Gadikota HR, Wu JL, et al. Comparison of singleand double-bundle anterior cruciate ligament reconstructions in restoration of knee kinematics and anterior cruciate ligament forces Am J Sports Med 2010;38:1359-1367. 8. Duthon VB, Barea C, Abra ssart S, et al. Anatomy of the anterior cruciate ligament. Knee Surg Sports Tr aumatol Arthrosc 2006;14:204-213. 9. Bowman KF Jr, Sekiya JK. Anatomy and bi omechanics of the posterior cruciate ligament, medial and lateral side of the knee. Sports Med Arthrosc 2010;18:222229. 10. Gabriel MT, Wong EK, Woo SL, et al. Distri bution of in situ forces in the anterior cruciate ligament in response to rota tor loads. J Orthop Res 2004;22:85-89. 11. Russell KA, Palmieri RM, Zinder SM, et al. Sex differences in valgus knee angle during a single-leg drop jump. J Athl Train 2006;41:166-171. 60

PAGE 61

12. Noyes FR, Grood ES. The strength of the anterior cr uciate ligament in humans and Rhesus monkeys. J Bone Joing Surg Am 1976;58:1074-1082. 13. Woo SL, Hollis JM, Adams DJ, et al Tensile properties of the human femuranterior cruciate ligament-tibia complex. The effects of specimen age and orientation. Am J S ports Med 1991;19:217-225. 14. Almqvist KF, Jan H, Vercruysse C, et al. The tibialis tendon as a valuable anterior cruciate ligament allograft subs titute; biomechanical properties. Knee Surg Sports Traumatol Arthrosc 2007;15:1326-1330. 15. Haut Donahue TL, Howell SM, Hull ML et al. A biomechanical evaluation of anterior and posterior tibial is tendons as suitable single-loop anterior cruciate ligament grafts. Arth roscopy 2002.;18:589-597. 16. Pearsall AW IV, Hollis JM, Russell GV Jr, et al. A biomechanical comparison of three lower extremity tendons for ligam entous reconstruction about the knee. Arthroscopy 2003;19:1901-1906. 17. Conner CS, Morris RP, Va llurupalli S, et al. Tensio ning of anterior cruciate ligament hamstring grafts; compari ng equal tension versus equal stress. Arthroscopy 2008;24:1323-1329. 18. Hher J, Scheffler SU, Withrow JD, et al. Mechanical behavior of two hamstring graft constructs for reconstruction of t he anterior cruciate ligament. J Orthop Res 2000;18:456-461. 19. Hamner DL, Brown CH Jr, Steiner ME et al. Hamstring tendon grafts for reconstruction of the anterior cruciate ligament: biomechanical evaluation of the use of multiple strands and tensioni ng techniques. J Bone Joint Surg Am 1999;81:549-557. 20. Royalty RN, Junkin DM Jr, Johns on DL. Anatomic double-bundle revision anterior cruciate ligament surgery using fr esh-frozen allograft tissue. Clin Sports Med 2009;28:311-326. 21. Cohen SB, Sekiya JK. Allograft sa fety in anterior cruciate ligament reconstruction. Clin Spor ts Med 2007;26:597-605. 61

PAGE 62

22. Clark JC, Rueff DE, Indelic ato, et al. Primary ACL reco nstruction using allograft tissue. Clin Sports Med 2009;28:223-244. 23. Buck BE, Malinin TI, Brown MD. Bone transplantation and human immunodeficiency virus. An estimate of risk of acquired immunodeficiency syndrome (AIDS). Clin Orthop 1989;240:129-135. 24. Asselmeier MA, Caspari RB, Bottenfield S. A review of allograft processing and sterilization techniques and their role in transmission of the human immunodeficiency virus. Am J Sports Med 1993;21:170-175. 25. Rodeo SA, Arnoczky SP, Torzilli PA, et al. Tendon-healing in a bone tunnel. A biomechanical and histological study in the dog. J Bone Joint Surg Am 1993;75:1795-1803. 26. Lee M, Hyman W. Modeli ng of failure mode in knee ligaments depending on the strain rate. BMC Muscu loskelet Disord 2002;3:3. 27. Haut RC. Age-dependence influe nce of strain rate on the tensile failure of rat-tail tendons. J Biomech Eng 1983;105:296-299. 28. Lewis G, Shaw KM. Tensile properties of human tendo Achillis: effect of donor age and strain rate. J Foot Ankle Surg 1997;36:435-445. 29. Pioletti DP, Rakotomanana LR, Leyvraz PF Strain rate effect on the mechanical behavior of the anterior cruciate ligam ent-bone complex. Med Eng Phys 1999;21: 95-100. 30. Noyes FR, DeLucas JL, Torvik PJ. Biomechanics of anterior cruciate ligament failure; an analysis of strain-rate sensit ivity and mechanisms of failures in primates. J Bone Joint Su rg Am 1974;56:236-253. 31. Haut TL, Haut RC. The state of tiss ue hydration determines the strain-ratesensitive stiffness of human patel lar tendon. J Biomech 1997;30:79-81. 32. Han S, Gemmell SJ, Jelmer KG, et al. Changes in ADC caused by tensile loading of rabbit achilles tendon: evi dence for water transport. J Magn Reson 2000;144:217-227. 62

PAGE 63

63 33. Haut RC, Powlison AC. The effects of test environment and cyclic stretching on the failure properties of human patellar tendons. J Orthop Res 1990;8:532-540. 34. Hasberry S, Pearcy MJ. Temperature dependence on the tensile properties of interspinous ligaments of the s heep. J Biomed Eng 1986;8:62-66. 35. Bass CR, Planchak CJ, Salzar RS et al. The temperature-dependent viscoelasticity of porcine lumbar sp ine ligaments. Spine 2007;32:E436-E442. 36. Wren TA, Yerby SA, Beaupr GS, et al. Mechanical proper ties of the human achilles tendon. Clin Biomech 2001;16:245-251. 37. Flahiff CM, Brooks AT, Hollis JM, et al Biomechanical analysis of patellar tendon allografts as a function of donor a ge. Am J Sports Med 1995;23:354-358. 38. Noyes FR, Butler DL, Grood ES, et al. Biomechanical analysis of human ligament grafts used in knee-ligament repai rs and reconstructions. J Bone Joint Surg Am 1984;66:344-352. 39. Bohnsack M, Srie B, Kirsch IL, et al. Biomechanical properties of commonly used autogenous transplants in the surgical treatment of chronic lateral ankle instability. Foot An kle Int 2002;23:661-664. 40. Duenwald SE, Vanderby R Jr, Lakes RS. Viscoelastic relaxation and recovery of tendon. Ann Biomed E ng 2009;37:1131-1140. 41. Duenwald SE, Vanderby R Jr, Lakes RS. Stress relaxation and recovery in tendon and ligament: experiment and modeling. Biorheology 2010;47:1-14. 42. Betsch DF, Baer E. Structure and mechanical properties of rat tail tendon. Biorheology 1980;17:83-94. 43. Atkinson TS, Ewers BJ, Haut RC. The tensile and stress relaxation responses of human patellar tendon varies with specim en cross-sectional area. J Biomech 1999;32:907-914.

PAGE 64

BIOGRAPHICAL SKETCH Eric Shields was born on July 20, 1986 in Laf ayette, LA. After relo cating to Slidell, LA, he attended Jesuit High School in New Or leans, LA. He received a National Merit scholarship to attend the University of Florida and he earned his B.S. in Mechanical Engineering in 2009. Eric began working at RTI Biologics in June of 2009 while pursuing his masters degree at the University of Florida. He received his M.S. in Mechanical Engineering from t he University of Florida in the spring of 2011. Through the guidance of Pedro Pedroso and Dr. Scott Banks, the current study was achieved. 64