<%BANNER%>

Biomechanical Characterization of Extensor Digitorum Longus, Flexor Digitorum Longus, Flexor Hallucis Longus, and Perone...

Permanent Link: http://ufdc.ufl.edu/UFE0041304/00001

Material Information

Title: Biomechanical Characterization of Extensor Digitorum Longus, Flexor Digitorum Longus, Flexor Hallucis Longus, and Peroneus Brevis Tendons as Viable ACL Replacements and Marathon Simulation to Determine Physiological Responses to Cyclic Loading
Physical Description: 1 online resource (76 p.)
Language: english
Creator: Myrick, Meridith
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: anterior, brevis, cruciate, digitorum, extensor, flexor, hallucis, ligament, longus, marathon, peroneus, simulation, tendons, tensile, testing, tibialis
Biomedical Engineering -- Dissertations, Academic -- UF
Genre: Biomedical 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 incidence of anterior cruciate ligament (ACL) replacement surgeries in the US is estimated to reach nearly 350,000 per year. Utilization of allograft tissue as a regenerative approach to tissue engineering can aid in restoration of native ACL biomechanics, while also preventing donor site defects associated with autologous grafts. A limited supply of allograft tissue motivates further investigation of alternative graft sources; therefore, biomechanical characterization of extensor digitorum longus (EDL), flexor digitorum longus (FDL), flexor hallucis longus (FHL), and peroneus brevis (PB) was performed in order to determine their efficacy as potential ACL replacements in a double strand configuration. Values achieved in testing anterior tibialis (AT) were used as a comparison because of its acceptance and current use as an ACL allograft. Tensile testing with a load-to-failure protocol was performed on all five tendons. Average ultimate tensile force (UTF) values for single strands of EDL, FDL, FHL, and PB were 1136.5 plus or minus 291.8 N (n=36), 954.1 plus or minus 270.5 N (n=36), 1156.3 plus or minus 348.0 N (n=36), and 1285.9 plus or minus 300.0 N (n=44). Single strand AT had an average UTF of 2122.3 plus or minus 574.7 N (n= 412) and an average ultimate tensile stress (UTS) of 91.05 plus or minus 28.04 MPa. Average UTS values were 134.59 plus or minus 36.0 MPa, 99.45 plus or minus 25.9 MPa, 101.98 plus or minus 28.9 MPa, and 87.85 plus or minus 22.5 MPa for EDL, FDL, FHL, and PB. With this data, it was shown that EDL has a statistically greater Young s modulus of elasticity value and greater UTS value than AT (p > 0.05). It was concluded that EDL, FDL, FHL, and PB provide viable sources as allograft tissue for ACL replacement if implemented as double strand grafts. To further characterize allograft tendons, AT was used to perform physiologic cyclic loading like that experienced by the native ACL during a simulated marathon. The average Young s modulus of elasticity over 11,712 cycles was 382.0 plus or minus 133.16 MPa and damage accumulation was observed in 7 of the 20 specimens. Average dynamic creep of twenty samples was 0.078 plus or minus 0.039 mm/mm, and a trend of increased strain was associated with decreased Young s modulus values. Negative hysteresis trends indicated decreased cyclic energy dissipation; however, no correlation was found between damage accumulation and hysteresis value. All tendons performed without failure, and a validation test confirmed that even with evidence of damage accumulation, the tendon was able to maintain its tensile strength.
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 Meridith Myrick.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Banks, Scott A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041304:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041304/00001

Material Information

Title: Biomechanical Characterization of Extensor Digitorum Longus, Flexor Digitorum Longus, Flexor Hallucis Longus, and Peroneus Brevis Tendons as Viable ACL Replacements and Marathon Simulation to Determine Physiological Responses to Cyclic Loading
Physical Description: 1 online resource (76 p.)
Language: english
Creator: Myrick, Meridith
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: anterior, brevis, cruciate, digitorum, extensor, flexor, hallucis, ligament, longus, marathon, peroneus, simulation, tendons, tensile, testing, tibialis
Biomedical Engineering -- Dissertations, Academic -- UF
Genre: Biomedical 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 incidence of anterior cruciate ligament (ACL) replacement surgeries in the US is estimated to reach nearly 350,000 per year. Utilization of allograft tissue as a regenerative approach to tissue engineering can aid in restoration of native ACL biomechanics, while also preventing donor site defects associated with autologous grafts. A limited supply of allograft tissue motivates further investigation of alternative graft sources; therefore, biomechanical characterization of extensor digitorum longus (EDL), flexor digitorum longus (FDL), flexor hallucis longus (FHL), and peroneus brevis (PB) was performed in order to determine their efficacy as potential ACL replacements in a double strand configuration. Values achieved in testing anterior tibialis (AT) were used as a comparison because of its acceptance and current use as an ACL allograft. Tensile testing with a load-to-failure protocol was performed on all five tendons. Average ultimate tensile force (UTF) values for single strands of EDL, FDL, FHL, and PB were 1136.5 plus or minus 291.8 N (n=36), 954.1 plus or minus 270.5 N (n=36), 1156.3 plus or minus 348.0 N (n=36), and 1285.9 plus or minus 300.0 N (n=44). Single strand AT had an average UTF of 2122.3 plus or minus 574.7 N (n= 412) and an average ultimate tensile stress (UTS) of 91.05 plus or minus 28.04 MPa. Average UTS values were 134.59 plus or minus 36.0 MPa, 99.45 plus or minus 25.9 MPa, 101.98 plus or minus 28.9 MPa, and 87.85 plus or minus 22.5 MPa for EDL, FDL, FHL, and PB. With this data, it was shown that EDL has a statistically greater Young s modulus of elasticity value and greater UTS value than AT (p > 0.05). It was concluded that EDL, FDL, FHL, and PB provide viable sources as allograft tissue for ACL replacement if implemented as double strand grafts. To further characterize allograft tendons, AT was used to perform physiologic cyclic loading like that experienced by the native ACL during a simulated marathon. The average Young s modulus of elasticity over 11,712 cycles was 382.0 plus or minus 133.16 MPa and damage accumulation was observed in 7 of the 20 specimens. Average dynamic creep of twenty samples was 0.078 plus or minus 0.039 mm/mm, and a trend of increased strain was associated with decreased Young s modulus values. Negative hysteresis trends indicated decreased cyclic energy dissipation; however, no correlation was found between damage accumulation and hysteresis value. All tendons performed without failure, and a validation test confirmed that even with evidence of damage accumulation, the tendon was able to maintain its tensile strength.
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 Meridith Myrick.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Banks, Scott A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041304:00001


This item has the following downloads:


Full Text

PAGE 1

1 BIOMECHANICAL CHARAC TERIZATION OF EXTENS OR DIGITORUM LONGUS FLEXOR DIGITORUM LON GUS, FLEXOR HALLUCIS LONGUS, AND PERONEUS BREVIS TENDONS AS VI ABLE ACL REPLACEMENT S AND MARATHON SIMULATION TO DETERMINE PHYSIOL OGICAL RESPONSES TO CYCLIC LOADING By MERIDITH R MYRICK A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Meridith R Myrick

PAGE 3

3 To Pedro D. Pedroso, who pursues excellence with a passion and encourages others to join in this endeavor

PAGE 4

4 ACKNOWLEDGMENTS I thank RTI Biologics Inc. for providing the time, resources, and guidance necessary to complete my thesis. I have been bless ed to work with a brilliant and caring team of people who have helped me to grow in so many ways. I give a special thanks and acknowledgement to Pedro Pedroso for encouraging me to pursue my thesis and for always providing guidance and expertise. I am grateful to Arunas Zhukauskas for sharing his four little tendons with me and for sharing his wisdom and experience. I could not have completed my thesis without the cooperation and guidance of my committee chair, Dr. Scott Banks, and my committee members, Dr Malisa Sarntinoranont and Dr. Benjamin Fregly, and to them I am grateful. I would also like to thank my patient and wonderful fianc, who knows more about tendons than any electrical engineer should. He is my champion. God has blessed me with a family who loves me and supports me through all of my endeavors, and with this security comes the confidence to pursue new challenges. Lastly, or rather firstly, I would like to thank Ms. Rebecca Austin for the spark. I have been fortified beyond measure.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES ................................................................................................................ 7 LIST OF FIGURES .............................................................................................................. 8 LIST OF ABBREVIATIONS ................................................................................................ 9 ABSTRACT ........................................................................................................................ 10 CHAPTER 1 INTRODUCTION ........................................................................................................ 12 Anterior Cruciate Ligament Replacement .................................................................. 12 Graft Utilization ........................................................................................................... 14 Allograft Tendons ................................................................................................. 14 Characteristics of the ACL and its Replacement Grafts ..................................... 15 EDL, FDL, FHL, PB, and AT Anatomy ................................................................ 18 2 MATERIALS AND METHODS I ................................................................................. 24 Tissue Allocation ......................................................................................................... 24 Loadto -Failure Test ................................................................................................... 24 Tendon Preparation ............................................................................................. 24 Gripping Mechanism ............................................................................................ 25 Test Method ......................................................................................................... 26 Statistical Methods ...................................................................................................... 27 Power and Sample Size ...................................................................................... 27 Comparative Analysis .......................................................................................... 28 3 MATERIALS AND METHODS II ................................................................................ 31 Marathon Test for Cyclic Loading .............................................................................. 31 Tendon Preparati on ............................................................................................. 31 Gripping Mechanism ........................................................................................... 31 Test Method ......................................................................................................... 32 Statistical Methods ...................................................................................................... 34 4 RESULTS I .................................................................................................................. 36 Loadto -Failure Test ................................................................................................... 36 Cross -Sectional Area o f EDL, FDL, FHL, PB, and AT ........................................ 36 Biomechanical Results of EDL, FDL, FHL, and PB ............................................ 37

PAGE 6

6 Assumed Double-Strand Comparison ....................................................................... 39 5 RESULTS II ................................................................................................................. 46 Marathon Test for Cyclic Loading Test ...................................................................... 46 Biomechanical Results of AT under Dynamic Loading ............................................. 46 6 DISCUSSION I ............................................................................................................ 52 Loadto -Failure Test ................................................................................................... 52 Biomechanical Attributes ............................................................................................ 52 7 DISCUSSION II ........................................................................................................... 55 Marathon Simulation Test .......................................................................................... 55 Biomechanical Attributes ............................................................................................ 55 8 TEST COMPARISONS: ATTRIBUTES, LIMITATIONS, AND FUTURE WORK ...... 59 Viscoelastic Properties ............................................................................................... 59 Future Applications ..................................................................................................... 64 APPENDIX A AVERAGE MARATHON TEST RESULTS ................................................................ 65 B MARATHON SIMULATION REGRESSION LINE EQUATIONS .............................. 66 C MARATHON SIMULATION STRAIN VS. MODULUS ............................................... 70 D TENTH CYCLE PERFORMANCE COMPARISON ................................................... 71 LIST OF REFERENCES ................................................................................................... 72 BIOGRAPHICAL SKETCH ................................................................................................ 76

PAGE 7

7 LIST OF TABLES Table page 2 -1 Practical difference values and standard deviation values for parameters in order to determine appropriate sample size at a power of 0.9. ............................ 30 4 -1 Single strand average biomechanical results for EDL, FDL, FHL, and PB .......... 41 4 -2 Average biomechanical results of double EDL, FDL, FHL, and PB and single AT ........................................................................................................................... 41 4 -3 Statistical analysis of all cohort groupings. ........................................................... 41 A-1 Average load and unload cycle for all marathon t endons .................................... 65 A-2 Average load cycle for all marathon t endons ....................................................... 65 A-3 Average unload cycle for all marathon t endons ................................................... 65 B-1 Regression lines for rate of hysteresis ................................................................. 66 B-2 Regression lines for Youngs modulus ................................................................. 67 B-3 Re gression lines for strain .................................................................................... 68 B-4 Regression lines for extension .............................................................................. 69 D -1 Tendon comparison for each test method ............................................................ 71

PAGE 8

8 LIST OF FIGURES Figure page 1 -1 Average UTF values of accepted allograft tendons for ACL replacement. .......... 21 1 -2 Average UTS values of accepted allograft tendons for ACL replacement ........... 22 1 -3 FDL, PB, EDL, FHL, and AT tendon comparison. ................................................ 23 2 -1 CryoGrip design ..................................................................................................... 29 2 -2 Loadto -failure test protocol. .................................................................................. 29 3 -1 Marathon test set -up with the MTS 858 Bionix testing s ystem. ............................ 35 4 -1 Average cross -sectional area (CSA) of each tendon. ........................................... 42 4 -2 Biomechanical UTF results for EDL, FDL, FHL, and PB. ..................................... 42 4 -3 Biomechanical UTS results for single strand AT, EDL, FDL, FHL, and PB. ........ 43 4 -4 Biomechanical Youngs modulus results for single strand AT, EDL, FDL, FHL, and PB. .......................................................................................................... 43 4 -5 Youngs modulus averaged for each tendon over 100 cycles. ............................. 44 4 -6 ......... 44 4 -7 Biomechanical UText results for single strand AT, EDL, FDL, FHL, and PB. ...... 45 4 -8 Biomechanical UTF results for double strand EDL, FDL, FHL, and PB compared to single strand AT. ............................................................................... 45 5 -1 Dynamic creep ........................................................................................................ 49 5 -2 Strain. Load cycles exhibit higher strain than unload cycles. ............................... 49 5 -3 Extension at maximum, average, and minimum loads. ........................................ 50 5 -4 Youngs modulus. Average load and unload cycles plotted. ............................... 50 5 -5 Cycle 2 stress -strain curve displaying hysteresis as the area between the curves. .................................................................................................................... 51 5 -6 Cycle 6088 stress -strain curve displaying hysteresis as the area between the curves. .................................................................................................................... 51 C -1 Average strain vs. average Youngs modulus ....................................................... 70

PAGE 9

9 LIST OF ABBREVIATION S CDC Centers for Disease Control and Prevention ACL Anterior cruciate ligament EDL E xtensor digitorum longus FDL F lexor digitorum longus FHL F lexor hallucis longus PB P eroneus brevis AT Anterior tibialis U TF U ltimate tensile force UTS U ltimate tensile stress U ltimate tensile strain UText U ltimate tensile extension E Youngs modulus CSA C ross -sectional area

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science BIOMECHANICAL CHARAC TERIZATION OF EXTENS OR DIGITORUM LONGUS FLEXOR DIGITORUM LON GUS, FLEXOR HALLUCIS LONGUS, AND PERONEUS BREVIS TENDONS AS VI ABLE ACL REPLACEMENT S AND MARATHON SIMULATION TO DETERMINE PHYSIOL OGICAL RESPONSES TO CYCLIC LOADING By Meridith Myrick December 2009 Chair: Scott Banks Major: Biomedical Engineering The incidence of anterior cruciate ligament ( ACL ) replacement surg eries in the US is estimated to reach nearly 350,000 per year. Utilization of allo graft tissue as a regenerative approach to tissue engineering can aid in restoration of native ACL biomechanics, while also preventing donor site defects associated with autologous grafts. A limited supply o f allograft tissue motivates further investigat ion of alternative graft sources; therefore, biomechanical characterization of ex tensor digitorum longus (EDL), flexor digitorum longus (FDL), flexor hallucis longus (FHL), and peroneus brevis (PB) was performed in order to determine their efficacy as pote ntial ACL replacements in a double strand configuration. Values achieved in testing anterior tibialis (AT) were used as a comparison because of its acceptance and current use as an ACL allograft. Tensile testing with a load -to -failure protocol was perfor med on all five tendons. Average ultimate tensile force (UTF) values for single strands of EDL, FDL, FHL, and PB were 1136.5 291.8 N (n=36), 954.1 270.5 N (n=36), 1156.3 348.0 N (n=36), and 1285.9 300.0 N (n=44). Single strand AT had an average U TF of 2122.3 574.7

PAGE 11

11 N (n= 412) and an average ultimate tensile stress (UTS) of 91.05 28.04 MPa. Average UTS values were 134.59 36.0 MPa, 99.45 25.9 MPa, 101.98 28.9 MPa, and 87.85 22.5 MPa for EDL, FDL, FHL, and PB. With this data, it was show n that EDL has a statistically greater Youngs modulus of elasticity value and greater UTS value than AT (p>0.05). It was concluded that EDL, FDL, FHL, and PB provide viable sources as allograft tissue for ACL replacement if implemented as double strand g rafts. To further characterize allograft tendons, AT was used to perform physiologic cyclic loading like that experienced by the native ACL during a simulated marathon. The average Youngs modulus of elasticity over 11,712 cycles was 382.0 133.16 MPa a nd damage accumulation was observed in 7 of the 20 specimens Average dynamic creep of twenty samples was 0.078 0.039 mm/mm, an d a trend of increased strain was associated with decreased Youngs modulus values. Negative hysteresis trends indicated decre ased cyclic energy dissipation; however, no correlation was found between damage accumulation and hysteresis value All tendons performed without failure, and a validation test confirmed that even with evidence of damage accumulation, the tendon was able to maintain its tensile strength.

PAGE 12

12 CHAPTER 1 INTRODUCTION Anterior Cruciate Ligament Replacement The use of allograft tissue for reparative surgeries is an increasing trend, but resources are limited to the number of available donors. Anterior cruciate l igament (ACL) reconstructions are estimated to reach nearly 350,000 in the United States each year and this trend is growing by about 5% yearly1,2. Nearly 2.3 million orthopaedic soft tissue procedures were performed in the US in 20082, and the CDC estim ated that 1.5 million bone and tissue allografts are distributed yearly by American Association of Tissue Banks accredited tissue banks in the United States3,4. With a growing demand for tissues reaching 1.5 million3, and a limited number of donors, an in vestigation into alternative viable tissues for ACL replacement was conducted. Characterization included the following tendons: extensor digitorum longus (EDL), flexor digitorum longus (FDL) flexor hallucis longus (FHL) peroneus brevis, and anterior tibialis (AT) tendon. Alternative source s of allograft material better utilize the gift of donor tissue and provide an efficient and economically attractive source of additional graft material To the authors knowledge, no previous biomechanical evaluat ion has been performed on the FHL, EDL, PB, or FDL as viable ACL replacements. These grafts were tested to determine if, when doubled, they would perform comparable to the native ACL. According to different implant techniques, double or even quadruple st rands could be applied as allografts. The AT tendon also was tested to serve as a comparison graft because of its popularity as allograft tissue and the available m echanical property values reported in literature. To further characterize allograft tendons, AT was used to

PAGE 13

13 perform physiologic cyclic loading like that experienced by the native ACL during a simulated marathon. Sports -related injuries are the most common sources of knee i njuries, and these injuries most often manifest in tears of the ACL, whi ch greatly reduces the stability of the knee joint2. In 2008, 334,500 ACL reconstructions were performed, and it is predicted that 423,700 reconstructions will be performed annually by the year 20132. The total cost of knee ligament reconstruction grafts and devices in 2008 totaled about $435 million, and this area of orthopaedics will only continue to grow as an ageing population continues to pursue active lifestyles and incur more spor ts -related injuries2,5. In addition, women are more likely to sustai n ACL injuries than men, and their increased participation in sports will further increase the need for knee ligament reconstructions5. Ligament damage in the knee can occur when a sudden motion, hyperextension, or flexion causes damage to the ligament2. Sports requiring cutting motions are most associated with this type of ligament damage and include footb all, basketball, and volleyball2, 5. In addition to the abrupt rupture or tear of the ACL due to this motion, it has been established that tendons could sustain damage due to the failure of a cyclically loaded tissue under higher stresses or through time dependent mechanisms in cycling, or by a combination of the two6. This type of motion could be observed during a run in which repetitive loading is ex perienced by the ACL To address both mechanisms of tendon damage, in vitro tendon testing was performed to determine tendon response under both ultimate tensile loads in a load -to -failure test protocol and under repetitive cyclic loading which simulated marathon forces and frequency.

PAGE 14

14 Graft Utilization Since the first ACL reconstruction in 1917, many new grafts and many new techniques have been implemented7. The use of allografts is a growing trend for multiple reasons and graft selection and fixation te chnique can be specialized based on patient needs and activity level. To this end, surgeons may select tendons with bone blocks or pure soft tissue (with the necessary fixation hardware) and t hey may choose the double strand technique or the single strand technique. To achieve the double strand orientation, s ome surgeons will loop the tendon, folding it in half and subsequently making two strands out of one tendon. In other cases, a double strand configuration is achieved by applying two single strand tendons as one graft. While there are merits to both autografts and allografts, the advantages of using a llograft tissue include: no donor site morbidity, reduced operating time for surgeons, unaltered patellofemoral tracking and thigh muscle function, unlimited available sizes, and smaller surgical incisions4,8,9. Additionally, a llograft tissue may be a necessary selectio n for revision ACL surgery if the patient formerly received an autologous graft. Allograft Tendons Multiple graft sources are util ized for allograft reconstruction, the two most common tendon grafts being the patellar tendon and a combinati on of gracilis and semitendinosus tendons10,11. Other common graft sources for ACL replacement include achilles tendon, quadriceps tendon, anteri or tibialis, and posterior tibialis1,4 ,9,12. Depending on the size and the location of the harvested tendon, the graft may be implanted as a single strand or as a double strand. T he number of strands applied depends on surgeon preference and/or whether b one blocks are present with the graft tissue.

PAGE 15

15 The double strand technique is suggested to more closely resemble the mechanics of the native ACL, and thus provide greater stability. H owever, depending on the technique of implantation utilized, four tunne ls may be necessary instead of the standard two7 ,13. The rate of graft healing within bone tunnels is a result of the size and number of the tunnels that must be produced for proper graft fixation13. Grafts composed of all soft tissue, therefore containi ng no bone blocks, have been shown to sustain loads adequate for native ACL activity, while also minimizing the necessary bone tunnel size for fixation9. While the merits of allograft tendons are numerous with regards to reconstructive purposes, a concern lies with the issue of sterility. To mitigate this concern, tissues can be sterilized to remove or inactivate organisms. The primary method s of st erilization for tendon tissue are through gamma irradiation, cryoprotectant incubation, and/or through propr ietary chemical processing4. According to the FDA, achievement of sterility is attained for biological medical devices at an assurance level of 103 (1 in 1000 chance of a living microbe existing in the tissue) 4. Greater regulations have been implemented since May 2005 regarding Current Good Tissue Practices in an effort to prevent disease transmission t hrough reconstructive surgeries3. Characteristics of the ACL and its Replacement Grafts In vitro testing of tendons, like the tests performed in the authors present study, have been performed in order to determine the efficacy of different tendons as replacements for the ACL10 ,14,15. Challenges exist in testing tendons, and the mitigating solutions to these challenges can affect final mechanical resul ts. Factors to take into consideration include hydration, temperature, gripping method, preconditioning, and tendon type and number (i.e. single, double, looped). A lite rature review is provided in

PAGE 16

16 which the biomechanics of different tendons are describe d. Ultimate tensile force (UTF) and ultimate tensile stress (UTS) graft comparisons are displayed in Figure 11 and Figure 12, respectively. ACL The ACL is consists of the anteromedial bundle (AM) and the posterolateral bundle (PM), which extends fro m the lateral femoral condyle within the intercondylar notch to an insertion at the anterior part of the central tibial plateau7. The length of the ACL ranges from about 22 mm to about 41 mm and its width averages about 7 mm to 12 mm13. The cross-section al area is not uniform and increases from the femur to the tibia13,16. In a pull to -failure test, Noyes et al. reported a UTF of the native ACL of younger donors as 1725 269 N with a maximum stress of 37.8 3.8 MPa 14,17,18. While the ACL is recorded t o fail at loads exceeding about 1725 N, it is estimated that the native ACL experiences only approximately 250 N during ambulation, and only approximately 454 N for most activities14,17,18. Obvious variations in the results for Youngs modulus of elas ticity values amongst different tendons, as well as tendons of the same type, indicate that determining this parameter is sensitive to many factors10,14,15,19,20. Youngs modulus is a material property; therefore, its values are expected to be relatively constant, or at least consistent. Tendons and ligaments, however, are viscoelastic in nature, and environmental factors such as temperature and hydration directly affect this pr operty21,22. Noyes et al. calc ulated the Youngs modulus of the ACL for adult s aged 4886 years old at a value of 65.3 24 MPa and a value of 111 26 MPa for the human ACL of donors aged 16 26 years old1 7. Butler et al. later tested the AM and PL bundles

PAGE 17

17 of the ACL separately and found Youngs modulus values of 283 114 MPa an d 154 120 MPa, respectively23,24 Bone patellar -tendonbone The centr al portion of the bone -patellar -tendon bone (BPTB) graft displayed a UTF of 2900 260 N, which is about 168% the UT F of the native ACL18. With an average area of 50.5 2.8 mm, thi s results in a UTS of 58.3 6.1 MPa18. Of the grafts listed for comparison, BPTB provides the greatest UTF value as a single one -stranded alternative to the ACL, and its cross -sectional area indicates the reason that it is applied as a single stranded gr aft. Hamstring and semitendinosus tendons. The hamstring tendons, semitendinosus and gracilis, are tested as single strands or as double strands, with either a combination of the two tendons, or two of the same tendons. When the double stranded tendons are tested, an important factor in determining the combined tensile properties is that tension is applied equally to both tendons1 0. Biomechanical observation of the double -strand approach displays the additive effect of applying multiple strands and this physical law is observed through previous work1 0,15. One stranded semitendinosus tendon with an average area of 10.8 2.2 mm h as a UTF of 1060 227N and a UTS of 99.3 14.9 MPa1 0. One stranded gracilis tendon with an average area of 7.4 1.1 mm has a lower UTF of 837 138 N and a higher UTS of 113.1 18.1 MPa1 0. Anterior and posterior tibialis tendons Anterior and posterior tibialis tendons perform similarly, with no significant difference in the values reported for UTF ( p= 0.53)1 5,20. Pearsall et al. reported a UTF of 3412 N for t he t wo -stranded AT tendon20, and Taut et al. reported a UTF of 4122 893 N for the looped double strand AT

PAGE 18

18 tendon1 5. Additionally, Pearsall et al. reported a UTF value of 3391 N for t he twostranded posteri or tibialis tendons15, and Taut et al. reported a UTF value 3594 1330 N for the looped double strand posterior tibialis tendon20. UTS values range from about 85 MPa to 108 MPa for both tendon types by each investigator of the double strand methods15,20. When doubled, the area of these tendons is comparable to that of the native ACL (p= 0.35), and their UTF values are about 200% that of the native ACL 15,18,20. Youngs modulus values f or AT varied in magnitude according to different studies. Haut et a l. recorded a value of 847 301 MPa for the looped double strand method15, yet Pearsall et al. reported Youngs modulus results as low as 0.243 MPa20. An interesting in vivo study was performed by Maganaris et al. in which the Youngs modulus was determ ined at different forces25. At an in vivo force of 71 12 N, a Youngs modulus was recorded as 450 60 MPa, and at an in vivo force of 530 59 N, Youngs modulus values were recorded at 1200 150 MPa25. The difference in Youngs modulus at these diff erent forces could be a result of straightening of the fiber crimp as the applied force increased. EDL, FDL, FHL, PB and AT Anatomy The tendons EDL, FDL, FH L, and PB contribute to the muscles of the ankle and foot. These muscle/tendon complexes steady the tibia and fibula perpendicularly on the talus, thus helping to maintain an upright posture and helping to give strength to the ankle -joint. Figure 1 3 displays these tendons The EDL tendon lies on the lateral portion of the front of the leg, and t he AT tendon lies on its medial side and the peronaei lie on its lateral side. The EDL divides into four strands and runs forward on the dorsum of the foot to the four lesser toes

PAGE 19

19 With this conformation, EDL acts to extend the phalanges of the toes and flex the foot upon the leg. Measurement of the EDL tendon from a donor legenbloc of the tibia/fibula portion was about 250 mm. The FDL tendon is located on the tibial side of the leg, and the muscle that it supports curls the second, third, fourth and fifth toes. As it passes into the sole of the foot, it crosses below the FHL tendon and is joined in a tendinous slip. It then divides into four tendons which are inserted into the phalanges of the lesser toes. In this way, i t is the antagonist muscl e/tendon complex of the EDL Measurement of the FDL tendon from a donor leg en bloc of the tibia/fibula portion was about 230 mm. The FHL tendon is located on the posterior compartment of the fibular side of the leg and occupies nearly the whole length of the posterior surface of the muscle. It runs along the sole of the foot and is inserted into th e last phalanx of the great toe where it is situated above and crosses from lateral to medial side of the FDL tendon, where it is connected by a fibrous slip. FDL and FHL are direct flexors of the phalanges, and extend the foot upon the leg to assist the Gastroc nemius and Soleus in extending the foot. This allows the actions of walking or standing on tiptoe. Measurement of the FHL tendon from a donor leg en bloc of the tibia/fibula portion was about 180 mm. The PB muscle lies under the peroneus longus muscle and its tendon is shorter and smaller than the other tendons, EDL, FDL, and FHL The fibers of the muscle pass downward and end in the PB tendon, wh ich runs behind the lateral malleolus. The p eroneus longus and PB tendons proceed within the same compartment via a common mucous sheath. PB tendon then passes down the outside of the lower leg and everts the foot, attaching the fifth metata rsal to the out side of the foot. The p eroneus longus

PAGE 20

20 and PB thus extend the foot upon the leg, acting against the AT and p eroneus tertius, which are flexors of the foot. Measurement of the PB tendon from a donor leg en bloc of the tibia/fibula portion was about 180 mm. Finally, the AT tendon located on the lateral side of the tibia and is inserted into the base of the first metatarsal bone. It is attached to the most medial muscle of the anterior compartment of the leg and is a direct flexor of the foot at the ankle joint, acting with peroneus tertius. When acting in conjunction with posterior tibialis, it raises the medial border of the foot, inverting it and s tabilizing the ankle as the foot hits the ground in walking. The five tendons were tested for their average Youngs modulus value, average UTF average UT S, average ultimate tensile strain (UT and average ultimate tensile extension (UText) The aim of this analysis was to determine the potential use of the EDL, FDL, FHL, and PB as viable ACL replacement tendons, therefore, maximizing utilization of donor tissue. To further predict allograf t performance, a marathon simulation utilizing AT was performed and the properties dynamic creep, strain, extension, hysteresis, and Youngs modulus were calculated over 11,712 cycles to display tendon response to long-term cyclic loading.

PAGE 21

21 Figure 11. Average UTF values of accepted allograft tendons for ACL replacement.

PAGE 22

22 Figure 12. Average UTS values of accepted allograft tendons for ACL replacement

PAGE 23

23 Figure 13. FDL, PB, EDL, FHL, and AT tendon comparison.

PAGE 24

24 CHAPTER 2 MATERIALS AND METHODS I Ti ssue Allocation All donor tissue was obtained through RTI Biologics, Inc. A minimum of 36 samples accounted for each tendon type. A possibility of two samples were retrieved from one tendon, one distal and one proximal in orientation. Therefore, four samples coul d originate from a single donor in the event that tendons were recovered from both legs of a donor. All tendons were recovered according to criteria established by the FDA and the American Association of Tissue Banks The tendons were stored in -70 C until the day of testing, and at the time of testing, they were thawed in 0.9% saline at room temperature for a minimum of thirty minutes. Load-to Failure Test Tendon Preparation Thirty -six EDL samples were acquired, constituting ten donors aged 32 years to 76 years with an average age of 54.1 years old. Thirty -six FDL samples were acquired, constituting eleven donors aged 29 years to 82 years with an average age of 63.64 years old. Thirty -six FHL samples were acquired, constituting eleven donors aged 29 years to 82 years with an average age of 57.9 years old. Forty -four PB samples were acquired, constituting fourteen donors aged 22 years to 70 years with an average age of 49.1 years old. EDL, FDL, FHL, and PB tendons were all aseptically process ed. Four -hundred twelve AT samples were acquired, aged 15 to 94 years old with an average of 57.5 years. One hundred thirteen donors contributed to this sample size, and these ten dons were all sterilized through the BioCleanse sterilization process. AT tendons contributing to the authors comparison study were also applied to other

PAGE 25

25 research aims, which explains the selection for a large sample size and the reasoning for sterilized tendons Additionally, the fact that the AT tendons were sterilized did n ot alter the biomechanical results achieved. Sch mizzi et al performed a study comparing tendons processed through the BioCleanse sterilization process to those aseptically processed and no significant differences were found between the two groups after t he first cycle was performed and no significant differences were found in tendon failure loads26. A length of approximately 35 mm was measured in the center of each sample with about 20 mm remaining on each end for gripping. The gage length of 35 mm was distinguished to closely match the length of the ACL between tibia and femur insertions, which typically ranges between 22 mm to 41 mm13. Measurements were recorded using Mitutoyo IP67 digital calipers. Major and minor dimensions were measured in three locations along the sample and averaged to calculate an average cross -sectional area2 7. Major and minor dimensions were calculated for gage repeatability and reproducibility, and acceptable levels were found at 6.14% and 11.93% contribution in measurement variation. Gripping Mechanism Gripping of the tendons was achieved using custom designed CryoGrips provided by RTI Biologics, Inc. Studies of the factors associated with this gripping method have been reported, and many researchers have adopted similar gripping methods10,15,20. No slippage of the tendon occurred with the application of the CryoGrip. Both ends of the tendon were clamped between grooved surfaces such that the tendon snakes around the grooves for a better hold. Dry ice was packed in a cooler that fit around each grip in order to freeze the tendon end to the gripper as one system. A warm -water jacket was

PAGE 26

26 folded over the exposed tendon, and water was circulated within the tubing of this jacket such that the tendon was maintained at a temperature of about 37C (body temperature). This configuration allowed the ends of the tendon to be frozen to the grips while maintaining a temperature of about 37C along the mid-substance of the tendon that was to be tested. Th e necessary freeze timing depended upon on the thickne ss of the tendon. The AT tendons were allowed fifteen minutes to completel y freeze to the grips, and the EDL, FDL, FHL, and PB were allowed eleven minutes for freezing. Figure 2-1 illustrates the CryoGrip configuration. Test Method Testing was performed using an MTS 858 Bionix testing system, and Multiworks software was utilized to record the time, extension, and force applied to the sample throughout the test. The tensile test included t hree operations, which involved three 90N holds for one minute each, one hundred cycles at loads of 50N to 250N, and a load-to -failure of the tendon at 50 mm/min Figure 2-2 summarizes this testing sequence. The 90 N hold serves as a preconditioning element for the tendon, simulating what many surgeons practice before implanting a tendon for reconstructive surgery7. This 90 N hold helps to straighten out the crimp in the collagen fibers associated with the toeregion of the stress -strain curve28,29. Thus straightening the crimp results i n a linear stress-strain curve28,29. The load profile during cycling replicates the forces experienced by the ACL during normal ambulation18. Finally, the load to -failure provides ultimate tensile properties of the tendon in the case of maximum utility o f the tendon. Failure mode was recorded, with mid -substance failures being the ideal, and most frequent,

PAGE 27

27 mode of failure. The properties that were derived from this test included UTF UTS, UText, and Youngs modulus. Statistical Methods Power an d Sample Size Sample size was calculated using MiniTab Statistical software. A power of 0.9 was selected for all sample size calculations. Sample sizes were determined using oneway ANOVA calculations with five levels for the EDL, FDL, FHL, PB, and AT t endons All standard deviation values were obtained from a dataset of 245 AT tendons that had already been tested by the author. Practical differences were deduced from findings in literature. Noyes et al. stated that the ACL will experience a maximum o f about 450 N during normal activity18. Blythe et al., however, suggested that a value of 700 N allows a greater factor of safety; therefore this value was applied in the authors study in the event that athletic events, etc. should require forces exceedi ng 450 N11. The UTS practical difference was calculated from the adopted 700 N practical difference and divided by the average cross -sectional area of the 245 AT tendons. The practical difference recorded for strain was based on an in vivo determination by Duthon et al. in which he found the natural elongation of the native ACL in different flexion orientations30. Based on the stress and strain parameters, Youngs modulus was calculated. With these constants, sample sizes for each parameter were calcula ted and the largest sample size was applied. This ensured adequate resolution for calculations and comparisons. Table 2-1 displays the practical differences and standard deviations utilized for calculating sample size.

PAGE 28

28 Comparative Analysis Comparisons o f variation between the five tendon cohorts utilized Bartletts test. For results displaying equal variances (p way ANOVA was applied to determine if any significant differences were detected. If so, Tukeys 95% confidence interval of pair w ise comparisons was conducted to determine the tendon cohorts that were statistically different. If Bartletts test indicated unequal variance, a BrowneForsythe f -test was executed. If significant differences were detected, Welchs t -test was utilized t o determine the tendon cohorts that were statistically different.

PAGE 29

29 Figure 21. CryoGrip design. Coolers contain dry ice which freezes the ends of the tendon to the grip. A warm water jacket surrounds the tested tendon portion in order to maintain a tissue temperature near 37C for testing. Permission for image use provided by Pedro Pedroso. Figure 22. Load -to -failure test protocol.

PAGE 30

30 Table 2 1. Practical difference values and standard deviation values for parameters in order to determine appropriate sample size at a power of 0.9. Parameter Practical Difference Standard Deviation UTF 700 N 606.84 UTS 26.82 MP a 28.3 0 UT 0.14 mm/mm 0.034 E avg 191.57 M P a 0.38

PAGE 31

31 CHAPTER 3 MATERIALS AND METHODS II Marathon Test for Cyclic Loading Tendon P reparation Twenty samples were acquired from twelve donors for the marathon simulation. Only proximal ends were tested due to the grips for this test, and measurements were acquired in the same manner as for that of the load -to -failure test method. Tendon length between the grips for this test ranged from 16.08 mm to 35.7 mm with an average of 26.85 mm. This discrepancy was also due to the gripping method adopted for the marathon test; however, in dynamic testing, Youngs modulus is not dependent on specimen length29. Also, AT tendons prepared for the marathon test were aseptically proces sed and were not sterilized through the BioCleanse sterilization process Again, this does not alter the viability of comparison amongst tendon samples according to Schimizzi et al26. Gripping Mechanism The marathon simulation test was performed in a saline bath; therefore, the standard freezing method was not possible. Soft tissue grips with rough surfaces were initially used to grip the tendons; however, yield was not consistent with this method and tendons were being torn at the interface of the gripper. A grooved appliance designed to fit over the grip surface was next attempted, where the tendon would snake around the grooves. This method provided more consistent results, but the tendon sample was subjected to micro-tearing at the interface where it emerged from the grips. To mi tigate this problem, a foam sheet was cut in the shape of the grooved appliance and served as

PAGE 32

32 a protective barrier against rubbing at each end. This did not interfere with the response of the tendon to the applied forces. Only the proximal end of the t endon was tested in the marathon test because of this gripping system. With the necessary tightness of the grip to prevent slippage and the snaking of the tendon, distal ends, which are more elliptical in shape, could not sustain their shape and failed at the grip interface. No problems were found with testing the more proximal ends of the tendons. Test Method The marathon test was also performed using the MTS 858 Bionix testing system, and Multiworks software recorded the time, extension, and force ap plied to the sa mple throughout the test. P arameters were selected for this marathon test to best simulate the impact that an ACL experiences during a 26.2 mi run for a competitive runner in vitro. Load values were calculated based on a runner with a mass of 55 kg, which was reported as the average mass of a top-class marathon runner by Billat et al31. Results from the top ten male and top ten female finishers of the Boston M arathon of 2009 were averaged to find the speed of the runners, and thus the stri de length. A stride length of 3.60 m was calculated based on the reported race frequency of 1.37 Hz by De Zee et al3 2 and average Boston marathon running times of 2 h and 22.6 min33. Race day preparation was also factored into the test protocol, which b oth simulated the race experience, and also served to precondition the tendon before exerting marathon forces and frequencies on the tendon. Ambulation with forces of 50 N to 250 N w as first applied with a frequency of 0.67 Hz for 100 cycles as a warm -up to the marathon. Next, three 90 N holds of 30 s duration were applied, thus further preconditioning the tendon and simulating a stretch sequence by the runner. Finally, the marathon proceeded with

PAGE 33

33 forces of 250 N to 600 N at a frequency of 1 Hz for the f irst 1000 cycles. The frequency was then increased to 1.37 Hz for the remaining 10,712 cycles of the marathon. The entire marathon test, including preconditioning, lasted about 2.5 hr. Maximum forces experienced by the ACL during running are estimated to be about 2.2 times bodyweight34. For a mass of 55 kg, the estimated load at peak impact of the ACL is 1191.3 N. This load was then divided in half to represent the load that would be experienced by one strand of a double-stranded graft. A picture of th is test set -up is provided in Figure 3-1 and a plot describing the test protocol is provided in Figure 3-2. To prevent desiccation of the tendon over the 2.5 hr period, the tendon was tested in a saline bath of 0.1% to 0.5% saline (salinity measured with an Accumex probe). Ideally, this salinity would have been maintained at physiological levels, which would have been closer to 0.6% salinity levels35. Because tendons are viscoelastic materials, temperature is an important factor. Water temperature for th e marathon test was therefore maintained at the physiological level of 37C by using an immersion heater within the bath. Calculations derived from the data collected included strain, strain rate, Youngs modulus, extension, hysteresis, and dynamic creep. Strain was calculated as the change in length (extension) of the tendon over the original length of the tendon, and Youngs modulus was calculated based on Hookes law, where the slope of the linear portion of the stress -strain curve represented stiffness for each load and unload cycle component. The area between the load and unload curves of the stress -strain plot was calculated for hysteresis. Dynamic creep was found by calculating the difference

PAGE 34

34 between the strain of the peak of the last cycle and the strain of the peak of the first cycle. Statistical Methods Marathon Test for Cyclic Loading. A sample size of 20 AT tendons was determined in order to achieve a representative response. Because 10 female and 10 male race results were used to calculate the test protocol, a sample size of 20 was selected. To analyze phenomenon within this group, Bartletts test for equal variance was used, and a t test with a 95% confidence interval was utilized. The marathon test method, however, was not designed as a comparative analysis.

PAGE 35

35 Figure 31. Marathon test set up with the MTS 858 Bionix testing system. Figure 32. Marathon test protocol summary.

PAGE 36

36 CHAPTER 4 RESULTS I Load-to Failure Test Statistical comparisons of biomechanical results were first conducted to compare EDL, FDL, FHL, and PB separately as single strands. An independent evaluation of the four tendons, excluding AT, provided information concerning properties that cannot be simply doubled and also helps to determine which, if any, of t he four smaller tendons possess specific qualities that are of special interest. Analysis was then conducted for all five tendons as single strands, and finally, a statistical comparison of double strand EDL, FDL, FHL, and PB compared to the single strand AT was conducted. Table 4 -1 provides single strand biomechanical results of all tendons and Table 4-2 provides a hypothetical comparison of double strand EDL, FDL, FHL, and PB to the single strand AT. All p val ue results are listed in Table 4-3. Cross -Sectional Area of EDL, FDL, FHL, PB, and AT The cross -sctional area s of the EDL, FDL, FHL, and PB were measured as single strands; however, according to the results of Hamner et al. the doubling of these values provides an approximation of the propert ies achieved for double stranded testing10. Average cross -sectional areas were measured at 8.85 mm, 9.88 mm, 11.97 mm, and 15.98 mm for the tendons EDL, FDL, FHL, and PB, respectively. Unequal variance was fo und between the four tendons (p<< 0.05); t herefore, a BrownForsythe f -test was administered. The only tendons that were found to be statistically similar were EDL and FDL (p> 0.05). The average cross -sectional area of single strand AT tendons was measured at 26.10 mm, which is statistically gr eater than the assumed double strand

PAGE 37

37 configuration of EDL, FDL, FHL, and PB (p << 0.05). Figure 41 displays average cross-sectional areas for each tendon. Biomechanical R esults of EDL, FDL, FHL, and PB UTF. Average UTF values were 1136.5 291.8 N, 954. 1 270.5 N, 1156. 3 348.0 N, and 1285.9 300.0 N for the single strand tendons EDL, FDL, FHL, and PB, respectively (Fig 3-2). The average UTF value for AT was 2122.28 574.68 N. Analysis of EDL, FDL, FHL, and PB showed that the UTF values had equal variance with a Bartletts Test p value of 0.496. One way ANOVA for UTF showed a significant difference in the maximum loads (p<<0.05), with FDL displaying a different and lower UTF value according to Tukey s 95% simultaneous confidence intervals. Removal of FDL from the four tendon cohort gave a pvalue of 0.069, indicating no difference in UTF for EDL, FHL, and PB. Figure 42 displays average UTF values for EDL, FDL, FHL, and PB. When including AT into this analysis, unequal variance was detected and Bro wn -Forsythe analysis indicated that AT was statistically greater than EDL, FDL, FHL, and PB as single strand tendons (p<<0.05). UTS. Average UTS values were 134.59 36.0 MPa, 99.45 25.9 MPa, 101.98 28.9 MPa, and 87.85 22.5 MPa for the tendons EDL, FDL, FHL, and PB, respectively. The average UTS value for AT was 91.05 28.04 MPa. Comparing EDL, FDL, FHL, and PB, variances were found to be unequal according to Bartletts test (p=0.027), and the BrownForsythe f -test indicated a difference between at least one pair of groups (p=0.00064). A Welchs t -test was then computed to find that EDL was significantly greater than all other tendons. Analysis of all five tendons gave the same results with EDL being statistically greater than all other tendons according to Tukeys 95%

PAGE 38

38 simultaneous confidence interval. Figure 4-3 plots average UTS values for all five tendons. Youngs modulus Average Youngs modulus values were 1820.4 MPa, 1453.6 MPa, 1442.7 MPa, and 1220.3 MPa for the tendons EDL, FDL, FHL, and PB, respectively. The average Youngs modulus value for AT was 1387.2 487.5 MPa. Equal variances were found for the EDL, FDL, FHL, and PB comparison (p=0.403) but there was a significant difference in Youngs modulus values for EDL compared to the o ther tendons according to Tukey 95% simultaneous confidence interval (p<<0.05). When Youngs modulus of elasticity was statistically analyzed against all five tendons, no difference in variance was detected (p=0.075). However, one way ANOVA indicated tha t there was a significant difference amongst the five tendons (p<<0.05 ). Analysis with a Tukey 95% simultaneous confidence interval showed that EDL, again is significantly greater t han all other tendons. Figure 44 displays the average Youngs modulus values for all five tendons. Additionally, Figure 45 displays average Youngs modulus values across 100 cycles of the test protocol. Strain. Average strain values were 0.107 mm/mm, 0.0960 mm/mm, 0.103 mm/mm, and 0.113 mm/mm for the tendons EDL, FDL, FHL and PB, respectively. These tendons displayed equal variance values (p=0.654), but oneway ANOVA indicated a difference between FDL and PB. A Tukey 95% simultaneous confidence interval showed that PB had a significantly greater strain than FDL. Strain was calculated for AT as 0.11 0.033 mm/mm. Statistical analysis displayed that there was unequal variance amongst the tendon cohorts a nd further analysis showed a statistical difference amongst the tendons (p=0.033) with PB having

PAGE 39

39 a greater strain than FDL and AT having a greater strain than both FDL and FHL according to Welchs t -test. Figure 4-6 displays average strain values for each tendon. Extension: Average extension values during the loadto -failure were 3.75 mm, 3.35 mm, 3.60 mm, and 3.96 mm fo r the tendons EDL, FDL, FHL, and PB, respectively. Equal variance (p=0.677) allowed for a oneway ANOVA test, which indicated that at least one value was different (p=0.015). Further analysis applying a Tukey 95% simultaneous confidence interval showed t hat PB had significantly greater extension than FDL. Average extension was recorded for AT as 4.05 1.18 mm. Comparison of extension against all tendon cohorts did not display equal variance acco rding to Bartletts test (p<<0.05 ), therefore, a BrownFo rsythe f -test was again executed to detect for any differences within the tendon cohorts and it detected that there was a difference between at leas t one pair of tendons (p=0.031). Using a Welchs t -test, it was determined that PB ha d a statistically greater extension than FDL and AT had a statistically greater extension than FDL and FHL. Figure 47 displays average extension values for each tendon. Assumed Double Strand Comparison Single strand UTF values for EDL, FDL, FHL, and PB were doubled for each sample to conjecture how these tendons would compare to a single strand AT, which is currently an accepted allograft tissue. This double strand comparison mimics the current double strand gracilis and semitendinosus allograft combination. The average single strand AT tendon is 2122.2 N, and the assumed double UTF for the four tendons was calculated as 2273.0 N, 1908.2 N, 2312.6 N, and 2571.7 N for the EDL, FDL, FHL, and PB tendons, respectively. Equal variance produced a p value

PAGE 40

40 of 0.574 and one way A NOVA utilizing a Tukey 95% simultaneous confidence interval indicated that a doubled PB is significantly higher than the single strand AT (p<<0.05). Figure 48 displays UTF values with double strand EDL, FDL, FHL, and PB, and single strand AT.

PAGE 41

41 Table 4 1. Single strand average biomechanical results for EDL, FDL, FHL, and PB Tendon UTF (N) UTS (MPa) E (MP a) (mm/mm) n EDL 1136.5 291.79 134.6 35.98 1 82 0.4 0.45 0.11 0.024 36 FDL 954.1 270.46 99.5 25.94 1 45 3.6 0.38 0.096 0.023 36 FHL 1156.3 347.99 102.0 28.88 1442.7 0.47 0.10 0.022 36 PB 1285.9 300.03 87.9 22.46 1 22 0.3 0.37 0.11 0.03 44 *E represents Youngs modulus. Table 4 2. Average biomechanical results of double EDL, FDL, FHL, and PB and single AT Double Strand Max Load (N) UTS (MPa) CSA (mm ) double EDL 2272.98 128.42 17.70 double FDL 1908.16 96.57 19.76 d ouble FHL 2312.56 96.60 23.94 double PB 2571.72 80.47 31.96 single AT 2102.68 86.09 26.10 Table 4 3. Statistical analysis of all cohort groupings. Cohort Parameter Bartlett's test ANOVA/Browne Forsythe Tukey's 95% CI/Welch's t test 4 tendons UTF 0.496 0.00 FDL different from EDL, FHL, and PB UTS 0.027 0.00064 EDL > all E 0.40 << 0.05 EDL > all 0.65 0.015 PB > FDL Ext 0.68 0.015 PB > FDL 5 tendons UTS 0.057 << 0.05 EDL > all E 0.075 << 0.05 EDL > all << 0.05 0.033 PB>FDL, AT>FDL, FHL Ext << 0.05 0.031 PB>FDL, AT>FDL,FHL Double Strand UTF 0.57 << 0.05 dbl PB>single AT UTS 0.017 0.0020 single AT> dbl PB *4 tendons: EDL, FDL, FHL, and PB. tendons: EDL, FDL, FHL, PB, and AT. Double strand: double EDL, FDL, FHL, and PB compa red to single AT.

PAGE 42

42 P B F H L F D L E D L A T 2 5 2 0 1 5 1 0 5 0 T e n d o nC S A ( m m ^ 2 )1 5 9 8 4 1 1 1 9 7 4 3 9 8 7 8 2 1 8 8 5 0 1 1 2 4 6 1 4 1I n t e r v a l P l o t o f CS A ( m m ^ 2 )9 5 % C I f o r t h e M e a n Figure 41. Average cross -sectional area (CSA) of each tendon. P B F H L F D L E D L 1 4 0 0 1 2 0 0 1 0 0 0 8 0 0 6 0 0 4 0 0 2 0 0 0 T e n d o nU T F ( N )I n t e r v a l P l o t o f U T F ( N )9 5 % C I f o r t h e M e a n 1 1 3 6 4 9 9 5 4 0 8 1 1 5 6 2 8 1 2 8 5 8 6 Figure 42. Biomechanical UTF results for EDL, FDL, FHL, and PB.

PAGE 43

43 P B F H L F D L E D L A T 1 6 0 1 4 0 1 2 0 1 0 0 8 0 6 0 4 0 2 0 0 T e n d o nU T S ( M p a )8 7 8 5 1 0 1 9 8 9 9 4 5 1 3 4 5 9 9 1 3 2I n t e r v a l P l o t o f U T S ( M p a )9 5 % C I f o r t h e M e a n Figure 4 3. Biomechanical UTS results for single strand AT, EDL, FDL, FHL, and PB. P B F H L F D L E D L A T 2 0 1 5 1 0 0 5 0 0 T e n d o nE ( G P a )1 2 2 0 1 4 4 3 1 4 5 4 1 8 2 0 1 3 9 19 5 % C I f o r t h e M e a nI n t e r v a l P l o t o f Y o u n g s M o d u l u s ( M P a ) Figure 44. Biomechanical Youngs modulus results for single strand AT, EDL, FDL, FHL, and PB.

PAGE 44

44 Figure 45. Youngs modulus averaged for each tendon over 100 cycles. P B F H L F D L E D L A T 0 1 2 0 1 0 0 0 8 0 0 6 0 0 4 0 0 2 0 0 0 T e n d o nS t r a i n ( m m / m m )0 1 1 3 0 1 0 3 0 0 9 6 0 0 1 0 7 0 1 1 5I n t e r v a l P l o t o f S t r a i n ( m m / m m )9 5 % C I f o r t h e M e a n Figure 46.

PAGE 45

45 P B F H L F D L E D L A T 4 3 2 1 0 T e n d o nE x t ( m m )3 9 6 3 6 0 3 3 5 3 7 5 4 0 4I n t e r v a l P l o t o f E x t ( m m )9 5 % C I f o r t h e M e a n Figure 47. Biomechanical UText results for single strand AT, EDL, FDL, FHL, and PB. P B F H L F D L E D L A T 3 0 0 0 2 5 0 0 2 0 0 0 1 5 0 0 1 0 0 0 5 0 0 0 T e n d o nM a x L o a d ( N )2 5 7 1 7 2 2 3 1 2 5 7 1 9 0 8 1 5 2 2 7 2 9 8 2 1 2 2 1 69 5 % C I f o r t h e M e a nI n t e r v a l P l o t o f U T F ( N ) Figure 48. Biomechanical UTF re sults for double strand EDL, FDL, FHL, and PB compared to single strand AT.

PAGE 46

4 6 CHAPTER 5 RESULTS II Marathon Test for Cyclic Loading Test Average AT cross -sectional area values for the marathon test ranged from 20.49 mm to 43.84 mm, and the average cross -sectional area was 31.18 mm. Density measurements were taken at room temperature and an average of 0.95 g/ml was calculated, with a range of values between 0.802 g/ml and 1.163 g/ml. This is very close to the density of water, which is about 1g/ml at room temperature. Biomechanical Results of AT under Dynamic Loading Properties averaged over data derived from the entire marathon test were calculated, such as dynamic creep, average strain, average extension, average Youngs modulus, and average stress. These values were also determined after 1000 cycles, a half marathon, and a full marathon, and this inf ormation is located in Appendix A which also includes a table relating the load and unload values at these three time points. The parameters of strai n, extension, Youngs modulus, and hysteresis were also plotted and observed for changes over time as the test progressed and the tendon was subjected to elements of fatigue. Regression line equations of each tendon sample for these param eters are included in Appendix B Dynamic creep Dynamic creep is the difference of the strain of the last cycle peak and the strain from the first cycle peak. The average dynamic creep of all twenty marathon test s was 0.0780 0.039 mm/mm, with a range of 0.0152 mm/m m to 0.151 mm/mm. The average dynamic creep of tendons exhibiting tertiary creep (loss of stiffness) was 0.063 mm/mm (n=7), and the average dynamic creep of tendons

PAGE 47

47 exhibiting primary creep (no loss in stiffness) was 0.086 mm/mm (n=13). Figure 5 1 shows the dynamic creep of a sample specimen. Strain. Strain values remained in the physiological range with a minimum average strain of 0.106 mm/mm and a maximum average strain of 0.465 m m/mm The average strain over the course of the marathons was 0.279 0.118 mm/mm. Figure 52 indicates a positive slope for increasing strain over time. All of the samples in the marathon exhibited positive slopes indicating increasing strain as the marathon test progressed. However, strain rate was found to decrease as the test progressed. Extension. Average extension of the tendon over the duration of the marathon was 6.88 1.93 mm, with a range of 3.45 mm to 11.38 mm. Extension over the duration of the 11,712 cycles was logarithmic with a positive slope, indicat ing that the tendon was still extending at the end of the cycles and had not reached a state of equilibrium. Figure 53 shows the extension at minimum load, maximum load, and average load. Youngs modulus of elasticity. The average Youngs modulus f or the duration of the marathon was 382.04 133.16 MPa. The range for all twenty marathons was 168.485 MPa to 626.897 MPa. Seven of the marathons had a final negative slope, indicating that by the end of the cycling, the stiffness of the tendon was decr easing. The remaining thirteen tendons increased in Youngs modulus values as the cycle number increased, indicating stress hardening of the tendon. The frequency increased from 1 Hz to 1.37 Hz at cycle number 1000, and an evident increase in Young s modulus occur r ed at this increase in frequency and then the tendon returned to its previous Youngs modulus trend. At cycle 1000, the Youngs modulus of Fig 5 4 is

PAGE 48

48 361.35 MPa and increases to a maximum Youngs modulus value of 370.31 MPa at cycle 1003. The Youngs modulus value then continues to decrease and falls below its 1000 cycle value of 361.35 MPa at about cycle 9300. The gradual decrease of Youngs modulus after the increased frequency could be tendon recovery, or it could indicate fatigue of the te ndon over time. This result is consi stent, however, with Schecht man et als conclusion that frequency does not cause any discernable trends on the dynamic pa rameters of tendons in loading36. Hysteresis: Measurements were calculated to determine the hyst eresis by taking into account the load and unload areas of the stress -strain curve. Rebound resilience accounts for the energy not dissipated by the effects of hysteresis. Finally, the rate of hysteresis was also determined. The average hysteresis during the course of the marathon was 0.0515 0.0124; therefore, the average rebound resilience 0.949 0.0124. The range for hysteresis was 0.0304 to 0.0733 and the range for rebound resilience was accordingly 0.927 to 0.970. Hysteresis was found to decreas e over time with a minimal negative slope as the tendon became stiffer and the area between the stress-strain curve decrease (Figure 5 5 and Figure 5 6 ). The average rate of hysteresis was calculated as 2.35E -5 s1, and steady state of hysteresis was assum ed after analyzing that hysteresis values were greater than 98% percent consistent from mid marathon to the end of the marathon.

PAGE 49

49 Figure 51. Dynamic creep. The peak strain of the first cycle and the peak of the last cycle are plotted to show dynamic c reep. Figure 52. Strain. Load cycles exhibit higher strain than unload cycles.

PAGE 50

50 Figure 53. Extension at maximum, average, and minimum loads. Figure 54. Youngs modulus. Average load and unload cycles plotted.

PAGE 51

51 Figure 55. Cycle 2 stress -str ain curve displaying hysteresis as the area between the curves. Figure 56. Cycle 6088 stress -strain curve displaying hysteresis as the area between the curves.

PAGE 52

52 CHAPTER 6 DISCUSSION I Load-to Failure Test Biomechanical testing of the EDL, FDL, FHL, and PB provided results that are positive and promising for future use as allograft material. The inclusion of the four tendons EDL, FDL, FHL, and PB to the retinue of currently accepted allografts provides up to four times more grafts per donor. The four tendons were described as double strand assuming matching tendon type; however, combinations of the tendons are also capable of being paired. AT as has been reported previously, also displayed efficacious biomechanical results as an ACL replacement graf t, both as a single strand and as a double strand. The mechanical performance inherent in any particular tendon results from the load patterns it experiences and the subsequent fibril size that develops as a response to these stresses3 7. Biomechanical A ttributes Comparing single strand UTF values amongst the four tendons resulted in a significant difference between FDL EDL, FHL, and PB with FDL having a lower value of 954.1 N. This value, however, is still higher than the determined practical differ ence of 700 N. Even as a single strand, this tendon would sustain the estimated forces experienced by the ACL during normal activity11,14,17,18. When doubled, EDL, FHL, and PB all displayed a higher UTF than the single stranded AT. Statistically, doubled EDL, FDL, FHL, and PB are all viable ACL replacements when compared to the currently accepted single strand AT tendon Additionally, a ll four of the smaller tendons displayed greater average UTF values than the gracilis tendon, as reported by both

PAGE 53

53 Hamner et al. and Noyes et al.10,18, and all but the FDL tendon are within the range of UTF values reported in these two studies for the semitendinosus tendon10. UTS values provide a more normalized description of the strength of the tendon because they take cross-sectional area into account. This parameter displays that FDL, FHL, and PB were not different from AT in UTS v alues and EDL was in fact significantly greater than AT T he cross-sectional area of EDL is the smallest of all of the tendons, b eing only about 34% that of the AT tendon and the FDL, FHL, and PB tendons are about 38%, 46%, and 61% that of the AT cross-sectional area, respectively. Given the small size of the four tendons in relation to that of the AT tendon, this indicates that fibril distribution of the smaller tendons is more dense3 7. Implant site morbidity is reduced when smaller tunnels are required for implantation, as would be the case in utilizing the smaller tendons EDL, FDL, FHL, and PB, which provide the same or greater mech anical strength as other allografts. The muscle to which each tendon is associated and its function determine the strain necessary to complete each task, whether responsible for fine movements, extension, or flexion. A greater strain value would prevent t he tendon from impeding fine movement and would signify greater elongations during loading38. The similarity of strain between all five tendons is expected, given the location of the tendons in the lower leg and their function to aid in steadying the tibia and fibula perpendicularly on the talus. While there were statistically different strain values amongst different tendons, these differences were practical ly insignificant (no greater than 0.14 mm/mm). As would be expected, increased strain values displ ayed a relationship to decreased values of Youngs modulus.

PAGE 54

54 Youngs modulus indicates the degree to which the tendon store s and reuse s elastic energy according to its function38; therefore, stiffer tendons exhibit greater Youngs modulus values and do no t utilize elastic energy with the same efficiency as a tendon with a lower Youngs modulus value3 9. The stiffest tendon was EDL with a statistically greater Youngs modulus than any of the other tendons. This tendon branches into four strands as it appr oaches the digits of the foot and may need to be stiffer in order to maintain its integrity as it separates and terminates at each digit. Youngs modulus values for EDL were reported by Schechtman et al. at values of about 1090 MPa to 1356 MPa3 6, which is lower th an the 1,820 MPa, value achieved with the load-to -failure test FHL and FDL originate from the fibular and tibial sides of the leg, but in the sole, the FHL passes from the lateral to the medial side and is situated above FDL, where the two tendons are conn ected by a fibrous slip. This orientation exposes both tendons to the same mechanical stresses, and explains the similarity i n Youngs modulus response to cycling. PB was not statistically different from the tendons, excluding EDL, but it did provide the lowest Youngs modulus value of all of the tendons at 1220 MPa. The function of the peroneus muscles is to steady the foot upon the leg, and thus a lower Youngs modulus makes the tendon more compliant in this task

PAGE 55

55 CHAPTER 7 DISCUSSION II Marathon Simulation Test The marathon simulation replicated the forces experienced by the ACL during a long -term, high intensity run. The average stress subjected during the marathon was about 14 MPa, and this stress represented the forces experienced b y one strand of a double strand implant. If the full impact of a 2.2 times body weight load34 had been administered to a double strand AT tendon, the stresses would have remained the same because doubled forces would have also been matched with doubled cr oss sectional area. All twenty of the tendons performed the marathon test without failure and the data that was collected from each specimen was useful in understanding the mechanisms that occurred as the tendon experienced longterm cyclic loading. Bio mechanical Attributes Responses of strain, extension, Youngs modulus, and hysteresis all possessed logarithmic trends, and each property displayed a spike at the 1000 cycle in response to the increased frequency. The stiffness of the tendon, or Youngs modulus, is the best predictor of the state of damage accumulation in the tendon over time. Increasing stiffness, as seen in 13 of the 20 samples, indicated stress hardening as the tendon responded to the cyclic loading and stored more energy. Seven tend ons, however, displayed negative slopes at the end of the marathon, thus indicating damage accumulation as stiffness was not maintained. While some tendons did display damage due to cyclic loading, they were still able to withstand c yclic loading without failure, indicating that the fibril network was not compromised beyond a state that could perhaps be reparable in vivo at the end of the marathon. Schechtman et al. observed

PAGE 56

56 human tendons under the microscope after performing cyclic fatigue testing, and h e reported that partially fatigued tendons displayed well aligned fibers forming a well organized microstructure3 6. The strain sustained by the AT tendons during the marathon simulation increased over time, but the strain rate was negative and practicall y steady -state. While increasing strain values would eventually lead to failure of the tendons in vitro, the incidence of a constant strain rate indicates that the tendons were able to sustain cyclic loading for the entire marathon. If cycling had been administered beyond the marathon duration, observation of increased strain and increased strain rate could act to predict failure. Wren et al. cyclically tested Achilles tendons to failure and reported that rapidly increasing strain rates were usually obse rved prior to tendon failure in cyclic loading6. Again, this verifies that the AT marathon tendons were sustaining the marathon loading pattern with only signs of fatigue and no signs of failure. No correlations were found between the average strains exp erienced by the tendons and the presence or absence of damage accumulation. Appendix C provides a plot of average strain and average Youngs modulus values. T endons that experienced damage are indicated on this plot Dynamic creep of the AT tendon during the marathon test averaged at 0.078 0.039 mm/mm. A relationship between the overall average Youngs modulus value and the dynamic creep value indicated that stiffer tendons produced less dynamic creep and more compliant tendons allow ed the tendon t o experience more dynamic creep. Of the twenty AT tendons tested in the marathon test, seven displayed tertiary creep, as indicated by a decrease in Youngs modulus over time. The average dynamic creep of the tendons that did not exhibit tertiary creep (i.e. primary creep) was 0.086 mm/mm

PAGE 57

57 (n=13) and the tendons that did exhibit tertiary creep had dynamic creep values of 0.063 mm/mm (n=7); therefore, the dynamic creep value is not an indicator of damage as much as the display of a loss o f stiffness in the tendon. De Zee et al. modeled the response of an Achilles tendon to predict its dynamic creep during a marathon by conducting cyclic loading of 1600 cycles under the prescribed conditions experienced in running3 2. He assumed that 1600 cycles provided a r epresentative tendon response because it was found that most of the dynamic creep occurred within this first period before reaching an asymptote3 2. According to the results of the authors present marathon test, greater than 1600 cycles are required to pr edict tendon response to long -term cyclic loading because of the effects of damage accumulation. The viscoelastic property of hysteresis was examined to determine the energy dissipation of the tendon over time. Hysteresis and hysteresis rate both decreased over time, but this decrease was of 105 to 103 magnitude, indicating that hysteresis basically reached a steady state (less than 1.2% change from midmarathon to the end of the marathon). Both hysteresis and Youngs modulus values are derived from the same stress-strain relationship, but Youngs modulus appears to increase or decrease with time, and the hysteresis response remained constant. Further analysis of modulus and hysteresis revealed that, over time, the stiffening of the tendon reached a poi nt where no energy was dissipated and the ten don basically acted as an elastic solid At the beginning of the marathon test, individual load and unload cycles plotted for stress strain relationships displayed a hysteresis loop with a large amount of hyste resis, as indicated by the area between the curves. However, cycles analyzed at the end of the marathon test displayed load and unload stress -strain curves with little or no hysteresis.

PAGE 58

58 During an in an vivo cycling test with human subjects, Kubo et al. f ound no significant correlations between hysteresis and stiffness, and he therefore deduced that the stiffness of the tendon did not affect the ability of the tendon to reuse elastic energy and recover lost energy due to internal damping39. The authors present study produced a similar result where little correlation was found between stiffness and hysteresis for the overall marathon, but correlation was detected for part of the marathon up to the point when the tendon started to display purely elastic pr operties. The variation that was detected between different test specimens was expected because many factors contribute to tendon performance. Increased age is associated with decreased Youngs modulus values17, but this relationship was not directly in dicated amongst the 20 specimen samples. Instead, in vivo mechanical loading of the tendon by the donor is thought to contribute to its performance in vitro40. Pike et al. studied high and low stressed sheep tendons and found that tendons which are subjected to high-stress during growth are more resistant to damage due to cyclic fatigue3 7. Reeves et al. performed a clinical study by incorporating strength training to fourteen elderly individuals for 14 weeks and found that exercise training increased p atella tendon stiffness by 64% and reduced hysteresis by 28%41. These changes in stiffness and hysteresis are most likely due to increased packing density of collagen fibrils and changes in the collagen crimp structure41. If the activity level of the donors utilized in the marathon simulation test were known, it would be suspected that the tendons displaying damage accumulation belonged to less active individuals and the robust test tendons belonged to active individuals.

PAGE 59

59 CHAPTER 8 TEST COMPARISONS: ATT RIBUTES, LIMITATIONS, AND FUTURE WORK Viscoelastic Properties Viscoelastic properties of the AT tendon were quantified with both test methods, load-to -failure and marathon simulation, and the factors that affected viscoelasticity were also thus quantified. Utilization of the same tendon type, AT, for both test methods allowed for direct comparison of environmental effects and greater discernment concerning tendon properties and sensitivities, with Youngs modulus being the mode of comparison. The average Youngs modulus calculated for AT during the 100 cycles of the load to -failure test was 1387.2 487.5 MPa. However, the average Youngs modulus value calculated for AT in the simulated marath on test was 382.04 133.16 MPa. Due to the variations in the Youngs modulus values reported in literature, it is apparent that test method imposes critical factors that influence the stiffness of the tendon15,20,25. To validate the responses calculated for each test, a load -to -failure test was performed with one of the marathon AT tendon samples. The marathon test was nondestructive, and after testing, this sample was frozen at 70C until further load -to -failure testing. The average Youngs modulus value for this sample during the marathon test was of 505.6 MPa but interestingly, when this same tendon was tested under different environmental conditions in the load-to -failure test, its Youngs modulus was 1566.4 MPa. The UTF found for this tendon was 2613.6 N, which indicates that any damage which the tendon mi ght have sustained did not affect its strength or stiffness. With this comparison, it was deduced that parameters of the test method were responsible for this variation in Youngs modulus. The factors that were not consistent

PAGE 60

60 for each test included prec onditioning parameters, frequency, applied force, temperature, and hydration of the tendon. To evaluate the potential effects of force and Youngs modulus, the first ten cycles of the marathon test at a load profile of 50 N to 250 N and a frequency of 0.67 Hz was compared to the load profile of 250 N to 600 N and frequency of 1 Hz. The resulting Youngs modulus values were 343.7 MPa and 534.2 MPa, respectively. Thi s indicates that the tendon became stiffer in response to higher forces and an increased fr equency. Also, the load profile of 250 N to 600 N occurred later in the test, so the tendon fibers might have aligned (preconditioned) at this point to work in concert. The first ten cycles of the loadto -failure test (1 Hz frequency and loads of 50 N t o 250 N) for both AT and FDL were next compared to these marathon cycle Youngs modulus values and the load-to -failure test consistently displayed gr eater Youngs modulus values. Appen dix D displays a table with summary results of 10th cycle performance o f tendons in the different tests. Hydration of the tendon was first investigated to determine the mechanisms experienced by the tendon in both an immersed state and in a state exposed to air. Tendon hydration effects have been well documented22,42 45, but hydration of the tendon is not often a focus of the test method because of difficulties that exist in gripping the tendon without slippage. The tendon in the marathon test was fully immersed, while the tendon in the load -to -failure test was frozen to its grips with warm tubing wrapped around the mid-substance of the sample for about 11 to 15 minutes and then it was tested for 5 minutes (~20 min without active hydration but with tubing protecting against evaporation). According to Han et al, changes in the apparent diffusion coefficient have a mean time of approximately 15 minutes; therefore, it is possible that sufficient

PAGE 61

61 time was allowed for extrusion of fluid from the load -to -failure tendon before it was tested44. Immersion in saline causes swelling of the tendon, which increases tendon viscoelasticity; however, long-term testing with only a saline drip or atomizer does not maintain physiological hydration of the inner collagen fibrils6, 44. The tendons tested in the marathon test were subjected to i mmersion in 0.10% to 0.53% saline solution. No statistical difference was found in Youngs modulus values for salinity. Although it would be expected that fluid of less salinity would cause the Youngs modulus value to decrease as the tendon swelled, the average Youngs modulus value for tendons < 0.5% salinity was actually higher than tendons at A difference in Youngs modulus value was found for AT tested with the loadto failure test design (dehydrated) and with AT t ested in an immersion bath of the marathon design (hydrated), so the lack of statistical difference in Youngs modulus values in the varying salinities could be due to the cyclic loading. This loading pattern perhaps does not allow the tendon to swell bey ond a threshold point due to the constant repetitive straightening of the collagen fibrils. Static and dynamic tests exhibit different Youngs modulus values when observed on a molecular basis43, and tensile loading has been shown to cause extrusion of water from the inside of the tendon to a bulk phase along the outside surface of the tendon44. For future tests however, a consistent, physiological salinity should be utilized. Increased water content in a tendon causes increased relaxation, allowing more freedom for molecular rearrangement; whereas, a dehydrated tendon will have less freedom22,42. Thornton et al. found that dehydrated tendons displayed decreased creep,

PAGE 62

62 as opposed to counterpart hydrated tendons, but the effects of variations in solutions were completely reversible4 2. Haut et al. also found that tendons immersed in hypotonic water relaxed faster than those immersed in hypertonic solution, meaning that more hydrated tendons would display less stiffness than less hydrated tendons22. This t rend was found by the author in the two methods of tendon testing, with higher Youngs modulus values found in the loadto -failure test method, which did not have active hydration, and lower Youngs modulus values found for tendons fully hydrated in water immersion. It is assumed that the greatest influence in disparity between the two test methods on the same AT tendon resulted from hydration of the tendon. Also important to note, while Haut et al. did find differences in stiffness due to hydration, she also found that ultimate loads and elongation were not affected by the bath environment, which is important in validating the ultimate tensile test results of the CryoGrip loadto -failure test conducted by the author22. In addition to the effects of h ydration to Youngs modulus values, temperature plays an important role in viscoelastic behavior21. Many of the reported values in literature were derived from tests which were conducted at room temperature10,15,36. In the case of freeze clamps, the tend on is completely frozen within the grips and unless a source of heat is added, the mid-substance tissue of the tendon will subsequently be exposed to the cold temperatures as well, thus increasing Youngs modulus values. The CryoGrip method used in the l oad -to -failure test included a warm -water jacket to keep the mid-substance portion of the tendon at a physiological range of about 37 C. Freeze clamps are an accepted and popular mode of gripping because they are able to hold the tendon without slipping at high forces, and they are not destructive to the tissue

PAGE 63

63 at the gripping interface10,15,20. The utilization of warm water tubing is an addition to the design that helps maintain more accurate results of tendon performance in a physiological range. Whil e the tendon exterior was maintained at about 37 C, this temperature may not have been uniformly distributed from the exterior fibers to the center of the tendon. This feature would be more of a concern for thicker tendons that have more interior tissue with further contact to the tubing envelope. The marathon test fluid was maintained at a temperature of 37 C by using a temperature probe and an internal heater; therefore, the AT tendon of this test was maintained at a steady temperature for the duration of the ~2.5 h test. Comparison of temperature effects cannot be quantitatively calculated between the two tests because of factors such as hydration between the two test methods, but temperature is reported to have a critical effect on tendon stiffness Wang et al determined in creep testing that there was a reduction by the order of a magnitude in tendon lifetime when temperatures were raised from room temperatures to 37C29. To study potential knee laxity as a result of temperature effects on allogr afts during reconstructive surgery, Ciccone et al. simulated operating room conditions and exposed tensioned hamstring tendon grafts to room temperature and then increased the tendon temperature to 34C21. He found that increasing tendon temperature decreased tension and stiffness by approximately 40% and 70%, respectively, thus reinstating the effects of temperature on viscoelastic properties21. Understanding the response of tendons to temperature and hydration through testing methods such as those administered by the author can provide critical information to surgeons and patients undergoing ACL

PAGE 64

64 reconstructive surgeries. If a tendon becomes less stiff once it is exposed to warm body temperatures, laxity could be experienced by the patient. Future Ap plications The present study of EDL, FDL, FHL, and PB tendons show s promise for allocation of future allograft material. Given that most knee ligament injuries are sports related, tendon properties under such physiological loads as running are useful to d etermine the most applicable tendon for a given task. In the future, allograft tendons may be selected for their mechanical properties in relation to the patients activity of choice. A long -distance runner would select a tendon exhibiting more complianc e, therefore being able to store more elastic energy, and have less hysteresis, therefore being capable of re-using more elastic energy. A football player or baseball player who has more explosive actions and requires more cutting activity would select an allograft with higher ultimate tensile s trength A model relating the biomechanics of EDL, FDL, FHL, and PB tendons of the load-to -failure test to the biomechanics of the AT tendon in the marathon simulation would provide an extensive host of biomechanical properties in which to properly assess the efficacy of a specific allograft for implantation. An extension of the current pilot study to that end would require validation testing of the four tendons in the marat hon simulation test. This would requir e double strands to be tested in order to sustain the frequency of the applied loads, and proper tensioning would be required to accomplish representative responses from these tendons. Given the nature of the two test methods and the different environment s to which the tendons are exposed, a relationship should be established in which the factors are associated

PAGE 65

65 APPENDIX A AVERAGE MARATHON TEST RESULTS Table A -1. Average load and unload c ycle for all marathon t endons Duration Extension (mm) Force (N) St ress (MPa) Strain (mm/mm) Cycle 1000 6.587 423.446 14.042 0.268 Cycle 5856 6.775 422.952 14.026 0.275 Cycle 11712 6.884 422.735 13.918 0.279 Table A -2. Average load c ycle for all marathon t endons Duration Extension (mm) Force (N) Stress (MPa) Strain (mm/mm) Cycle 1000 6.535 416.313 13.810 0.266 Cycle 5856 6.723 413.473 13.714 0.273 Cycle 11712 6.834 412.748 13.690 0.277 Table A -3. Average u nl oad c ycle for all marathon t endons Duration Extension (mm) Force (N) Stress (MPa) Strain (mm/mm) Cycle 1000 6.638 430.280 14.265 0.270 Cycle 5856 6.820 432.020 14.324 0.277 Cycle 11712 6.931 432.267 14.333 0.281

PAGE 66

66 APPENDIX B MARATHON SIMULATION REGRESSION LINE EQUATIONS Table B -1. Regression lines for rate of hysteresis Tendon ID Rate of Hysteresis Equ ation s R value 1 Hysteresis Rate= 2.00E 05 *ln(t)+ 0.0002 0.5746 2 Hysteresis Rate= 2.00E 05 *ln(t)+ 0.0002 0.54 00 3 Hysteresis Rate= 2.00E 05 *ln(t)+ 0.0002 0.755 6 4 Hysteresis Rate= 3 .00E 05 *ln(t)+ 0.0002 0.6279 5 Hysteresis Rate= 2 .00E 05 *ln(t)+ 0.0002 0.6144 6 Hysteresis Rate= 2.00E 05 *ln(t)+ 0.0002 0.6079 7 Hysteresis Rate= 3 .00E 05 *ln(t)+ 0.0003 0.4474 8 Hysteresis Rate= 6 .00E 05 *ln(t)+ 0.0005 0.6865 9 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0004 0.6433 10 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0005 0.6432 11 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0004 0.6292 12 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0005 0.6189 13 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0005 0.6233 14 Hysteresis Rate= 4 .00E 05 *ln(t)+ 0.0004 0.593 0 15 Hysteresis Rat e= 4 .00E 05 *ln(t)+ 0.0004 0.6101 16 Hysteresis Rate= 4 .00E 05 *ln(t)+ 0.0004 0.6219 17 Hysteresis Rate= 4 .00E 05 *ln(t)+ 0.0004 0.6793 18 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0005 0.648 19 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0005 0.6838 20 Hysteresis Rate= 5 .00E 05 *ln(t)+ 0.0005 0.6332

PAGE 67

67 Table B -2 Regression lines for Youngs modulus Tendon ID Youngs Modulus Equation s R value 1 E= 0.7901 *ln(t) +532.94 0.0212 2 E= 1.388 *ln(t)+ 638.35 0.1432 3 E= 15.29*ln(t) + 631.82 0.9473 4 E= 15.06*ln(t) +461.02 0.9901 5 E= 88.00*ln(t) +1190.0 0.9342 6 E= 12.59*ln(t) +726.00 0.8845 7 E= 5.382*ln(t) +321.06 0.3764 8 E=21.116*ln(t) +398.16 0.6716 9 E=13.215*ln(t)+266.59 0.684 0 10 E=13.308*ln(t) +256.49 0.6178 11 E=7.6898*ln(t)+252.88 0.7284 12 E=4.6416*ln(t)+314.45 0.1607 13 E=6.04*ln(t)+0.0181 0. 394 0 14 E=6.3559*ln(t)+368.71 0.382 0 15 E=6.5841*ln(t)+219.27 0.6311 16 E=3.024*ln(t)+143.18 0.6997 17 E= 1.769*ln(t)+285.9 0.0402 18 E=7.5402*ln(t)+187.06 0.6357 19 E=14.626*ln(t)+255. 3 0.7253 20 E=4.4479*ln(t)267.45 0.1272

PAGE 68

68 Table B -3 Regression lines for strain Tendon ID Strain Equation s R 1 0.9864 2 0.9612 3 0.9858 4 0.9965 5 0. 0167*ln(c) 0.0023 0.7937 6 0.9983 7 0.9852 8 0.8098 9 0.8639 10 0.848 0 11 0.8116 12 0.9264 13 0.8631 14 0.9459 15 0.9253 16 0.9213 17 0.8471 18 0.9654 19 0.9469 20 0.3712 0.8356

PAGE 69

69 Table B -4 Regression lines for extension Tendon ID Extension Equation R 1 Ext=0.0506*ln(c)+4.6514 0.9864 2 Ext=0.0365*ln(c)+4.2491 0.9612 3 Ext=0.0689*ln(c)+3.8596 0.9859 4 Ext=0.1491*ln(c)+5.5904 0.9965 5 Ext=0.6398*ln(c) 0.09 0.7937 6 Ext=0.0836*ln(c)+2.7473 0.9982 7 Ext=0.2512*ln(c)+6.8105 0.9852 8 Ext=0.0885*ln(c)+8.9571 0.8098 9 Ext=0.0749*ln(c)+8.2001 0.8639 10 Ext=0.0847*ln(c)+10.67 0.848 0 11 Ext=0.1059*ln(c)+5.334 0.8116 12 Ext=0.1018*ln(c)+5.987 0.9264 13 Ex t=0.0711*ln(c)+6.8814 0.8276 14 Ext=0.138*ln(c)+6.4162 0.9459 15 Ext=0.0822*ln(c)+5.6208 0.9253 16 Ext=0.0755*ln(c)+6.9018 0.8969 17 Ext=0.3261*ln(c)+4.3837 0.8471 18 Ext=0.1052*ln(c)+4.8413 0.9553 19 Ext=0.0892*ln(c)+5.6122 0.9469 20 Ext=0.1908*ln( c)+6.5487 0.8356

PAGE 70

70 APPENDIX C MARATHON SIMULATION : STRAIN VS. MODULUS Figure C -1. Average strain vs. average Youngs modulus with damaged tendons indicated by the ( X ) symbol. No correlation was found between strain values and Youngs modulus v alues in relation to tendon damage.

PAGE 71

71 APPENDIX D TENTH CYCLE PERFORMANCE COMPARISON Table D -1. Tendon comparison for each test method Tendon Force Profile (N) Frequency (Hz) Modulus Cycle 9.5 (Mpa) Modulus Cycle 10 (Mpa) Marathon specimen 3 50 25 0 N 0.67 338.56 343.66 Marathon specimen 3 250 600 N 1 528.92 534.1 1 CryoGrip FDL 50 250 N 1 1829.03 1863.6 2 CryoGrip AT 50 250 N 1 2615.62 2651.0 7

PAGE 72

72 LIST OF REFERENCES 1 Cohen SB, Sekiya JK Allograft safety in anterior cruciate ligament reconstruction. Clin Sports Med. 2007;26:597-605. 2 Millenium Research Group US markets for orthopedic soft tissue solutions 2009. Toronto: Millenium Research Group, Inc, 2008. 3 Centers for Disease Control and Prevention. (2006, September 1). About Tissue Transplants. . 4 Clark JC, Rueff DE, Indelicato PA, Moser M Primary ACL reconstruction using allograft tissue. Clin Sports Med. 2009;28:22344. 5 Nagano Y, Ida H, Akai M, Fukubayashi T Biomechanical characteristics of the knee joint in female athletes during tasks associated with anterior cruciate ligament injury. Knee. 2009;16:1538. 6 Wren TA, Lindsey DP, Beaupr GS, Carter DR. Effects of creep and cyclic loading on the mechanical properties and failure of human Achill es tendons. Ann Biomed Eng. 2003;31:710 -7. 7 Woo SL, Wu C, Dede O, Vercillo F, Noorani S Biomechanics and anterior cruciate ligament reconstruction. J Orthop Surg Res. 2006;1:2. 8 Henson J, Nyland J, Chang HC, Caborn DN Effect of cryoprotectant incubation time on handling properties of allogeneic tendons prepared for knee ligament reconstruction. J Biomater Appl. 2008;0:110. 9 Royalty RN, Junkin DM Jr, Johnson DL. Anatomic double-bundle revision anterior cruciate ligament surgery using fresh-frozen allograft tissue. Clin Sports Med. 2009;28:311-26. 10. Hamner DL, Brown CH Jr, Steiner ME, Hecker AT, Hayes WC. Hamstring tendon grafts for reconstruction of the anterior cruciate ligament: biomechanical evaluation of the use of multiple strands and tensioning techniq ues. J Bone Joint Surg Am. 1999;81:549 -57. 11. Blythe A, Tasker T, Zioupos P ACL graft constructs: In vitro fatigue testing highlights the occurrence of irrecoverable lengthening and the need for adequate (pre)conditioning to avert the recurrence of knee i nstability. Technol Health Care. 2006;14:33547. 12. Noyes FR, Barber -Westin SD Revision anterior cruciate surgery with use of bone patellar tendon-bone autogenous grafts. J Bone Joint Surg Am. 2001;83:11311143.

PAGE 73

73 13. Steckel H, Starman JS, Baums MH, Klinger HM, S chultz W, Fu FH The double bundle technique for anterior cruciate ligament reconstruction: a systematic overview. Scand J Med Sci Sports. 2007;17:99108. 14. Grood ES, Noyes FR Cruciate ligament prosthesis: strength, creep, and fatigue properties. J Bone Joi nt Surg Am. 1976;58:1083 -8. 15. Haut Donahue TL, Howell SM, Hull ML, Gregersen C A biomechanical evaluation of anterior and posterior tibialis tendons as suitable singleloop anterior cruciate ligament grafts. Arthroscopy. 2002;18:58997. 16. Harner CD, Xerogean es JW, Livesay GA, Carlin GJ, Smith BA, Kusayama T, Kashiwaguchi S, Woo SL. The human posterior cruciate ligament complex: an interdisciplinary study. Ligament morphology and biomechanical evaluation. Am J Sports Med. 1995;3:736-45. 17. Noyes FR, Grood ES Th e strength of the anterior cruciate ligament in humans and Rhesus monkeys. J Bone Joint Surg Am. 1976;58:1074-82. 18. Noyes FR, Butler DL, Grood ES, Zernicke RF, Hefzy MS. Biomechanical analysis of human ligament grafts used in knee-ligament repairs and recons tructions. J Bone Joint Surg Am. 1984;66:34452. 19. Hollis JM, Pearsall AW, Stutz JD, Russell GV The biomechanical properties of the tibialis anterior, tibialis posterior, and peroneus longus tendons as potential graft sources. Poster presented at: 47th An nual Meeting, Orthopaedic Research Society; 2001 February 2528, San Francisco, CA. 20. Pearsall AW 4th, Hollis JM, Russell GV Jr, Scheer Z. A biomechanical comparison of three lower extremity tendons for ligamentous reconstruction about the knee. Arthroscopy. 2003;19:1091-6. 21. Ciccone WJ 2nd, Bratton DR, Weinstein DM, Elias JJ Viscoelasticity and temperature variations decrease tension and stiffness of hamstring tendon grafts following anterior cruciate ligament reconstruction. J Bone Joint Surg Am. 2006;88:107 1 8. 22. Haut TL, Haut RC The state of tissue hydration determines the strain-rate sensitive stiffness of human patellar tendon. J Biomech. 1997;30:7981. 23. Butler DL, Guan Y, Kay MD, Cummings JF, Feder SM, Levy MS Locationdependent variations in the material properties of the anterior cruciate ligament. J Biomech. 1992;25:511 -8. 24. Woo SL, Abramowitch SD, Kilger R, Liang R Biomechanics of knee ligaments: injury, healing, and repair. J Biomech. 2006;39:110.

PAGE 74

74 25. Maganaris CN, Paul JP In vivo human tendon mechanical properties. J Physiol. 1999;1:307-13. 26. Schimizzi A, Wedemeyer M, Odell T, Thomas W, Mahar AT, Pedowitz R. Effects of a novel sterilization process on soft tissue mechanical properties for anterior cruciate ligament allograf ts. Am J Sports Med. 2007;35:6126. 27. Provenzano PP, Heisey D, Hayashi K, Lakes R, Vanderby R Jr Subfailure damage in ligament: a structural and cellular evaluation. J Appl Physiol. 2002;92:36271. 28. Wang XT, Ker RF, Alexander RM Fatigue rupture of wallaby tail tendons. J Exp Biol.1995;198:847-52. 29. Wang XT, Ker RF Creep rupture of wallaby tail tendons. J Exp Biol. 1995;198:83145. 30. Duthon VB, Barea C, Abrassart S, Fasel JH, Fritschy D, Mntrey J Anatomy of the anterior cruciate ligament. Knee Surg Sports Tr aumatol Arthrosc. 2006;14:20413. 31. Billat VL, Demarle A, Slawinski J, Paiva M, Koralsztein JP Physical and training characteristics of top -class marathon runners. Med Sci Sports Exerc. 2001;33:208997. 32. De Zee M, Bojsen -Moller F, Voigt M Dynamic viscoelast ic behavior of lower extremity tendons during simulated running. J Appl Physiol. 2000;89:13529. 33. Boston Athletic Association (2009). 113th Boston Marathon Top Finishers. . 34. Derrick TR, Dereu D, McLean SP Impacts and kinematic adjustments during an exhaustive run. Med Sci Sports Exer. 2002;34:9981002. 35. Guyton AC, and Hall JE. Textbook of Medical Physiology 9th edition. PA, W.B. Saunders Company. 1996; 7. 36. Schechtman H, Bader DL Fatigue damage of human tendons. J Biomech. 2002;35:34753. 37. Pike AV, Ker RF, Alexander RM The development of fatigue quality in highand low -stressed tendons of sheep (Ovis aries). J Exp Biol. 2000;203:2187-93. 38. Fukunaga T, Kawakami Y, Kubo K, Kanehisa H. Muscle and tendon interaction during human movements. Exerc Sport Sci Rev. 2002;30:10610.

PAGE 75

75 39. Kubo K, Kanehisa H, Fukunaga T Effects of viscoelastic properties of tendon structures on stretch shortening cycle exercise in vivo. J Sports Sci. 2005;23:85160. 40. Schechtman H Bader DL. In vitro fatigue of human tendons. J Biomech. 1997;30:82935. 41. Reeves ND, Narici MV, Maganaris CN Strength training alters the viscoelastic properties of tendons in elderly humans. Muscle Nerve. 2003;28:7481. 42. Thornton GM, Shrive NG, Frank CB Altering ligament water content affects ligament pre-stress and creep behaviour. J Orthop Res. 2001;19:845 -51. 43. Sasaki N, Odajima S Stress -strain curve and Young's modulus of a collagen molecule as determined by the X ray diffraction technique. J Biomech. 1996;29:6558. 44. Han S, Gemmell SJ, Helmer KG, Grigg P, Wellen JW, Hoffman AH, Sotak C Changes in ADC caused by tensile loading of rabbit achilles tendon: evidence for water transport. J Magn Reson. 2000;144:21727. 45. Dubinskaya VA, Eng LS, Rebrow LB, Bykov VA Comparative study of the state of water in various human tissues. Bull Exp Biol Med. 2007;144:2947.

PAGE 76

76 BIOGRAPHICAL SKETCH Meridith Myrick was born in Atlanta, Georgia where s he was raised with the foundation of two loving parents and a best friend for a sister. In 2006, Meridith earned her B.S. in biological engineering with an emphasis in biomedical engineering from the University of Georgia, and remains a loyal Bulldog fan. Upon graduation, Meridith worked in the field of civil engineering in the great city of Savannah, Georgia. She quickly realized, however, that her interests were only in the biomedical field, so in 2008, Meridith enrolled in the University of Floridas J. Crayton Pruitt Family Department of Biomedical Engineering as a masters s tudent. Within the first month of enrollment in graduate school, Meridith received an internship position with RTI Biologics, Inc. to work with the Sports Medicine Group. She worked for them on many enriching projects, including her thesis project, which was jointly coordinated through UF and RTI Biologics, Inc. Upon graduation from the University of Florida with a masters degree in biomedical engineering, Meridith moved back to Atlanta to start a career in the biomedical field. She will be a very joyful bride to Jon Ussery on March 27, 2010, and looks forward to a life of love, learning, and laughter.