font a. o o 'u "#31. ... u . a ct... . . .p... .. ... ... ... ..fo ...4 a. g . .... 4.. . . it... .92“. . . . . X. . .... u.7T-....«. ... . "......” .a ...-“”91. ””0 0.1.0... .50“..hh. . .....I o; :“w ‘Vuarwo “.nw’jymn m "Imfi-hcb) s ‘.8 ..- . I“. “a .. $.81 . ... . . 6 .4... ... 4...}... ... .... t .4 . . .....t u .. .. 3... ...-r. 2.3 - . .- ...! o ... I - 54v. 3 . u . .. .. .o....... .4 . . . ..al... 70 1 cl. . .. . o... .4. .. ...... 335A: ... Pint". _. 9...... our-:- . ...... .... . . ... "‘ .v t . .33..)oo31u...) .. I44: 3 n... r- ”a"... 8’. n. 3. . n... .... a. . .... a. . 4 5. ‘5’: b :- .~...o . .. . v... . .. . .1 ....9-. v...- 3. .23.... 45.1;- . . . .... .... u ... {No -...»-. L .o a '.,..:... . .03..“ ..!~ I. .50 . 701‘. I o ¢¢t ...-o. :ot. .. .0... 1 . a g \ . . .A ....u .4. ..... . . o... ... ..ovSIO... .0 r .2 . . . . . .33.- ... on“ I... .0 o- :0..- . .Io. . I... I . 0 o .o .3. u .. .. o....l¢l¢h«| o .0-5 bio-0:. no .5 0 O Cc. . .0. 4. .. ... . . .. Lo . ...? .... $418...‘ 0 t3. ....‘fi...’ 39 . .. .3 A. v.0... . co. 0 0. tv- - Q. 4 . 3 . . I no- ‘9‘ Ava unatn';hn ‘4 ... . u . D in ..- .).w.” c - ~ .I u R; 0.: ’1 ...-2.3... ca:- 1.‘. .a" c 3.. . :2! 1.412.. ..Ioflh...-.¢..V\Q. can .. \..7 E... 32.1.. ...-.30.. I! III: ' .5. . 3—“. L ......L .173... . .935. '0." ... :- ~ ...... {Jo-.5. .lo.‘ 0 a o . vialwu auto“. 44"”. “‘1’ v~.¢ ‘0‘}... .O'..l‘-rflt" ...." I v . ...: .— . 4 0. Go up. . o .. . III!- v trot“: It...” . ‘2’:- cu k3)....uihaoavo‘fls{:~¢.«0”.na.ub o liIH ” 3.... #14:"... “u : E 3.4.. I... ... . . lo....»o..\’.‘0|.c.cl v . .0. 4 . I- . . 3...? .29: . ... ... I . 4 . . . . , . . 1.001....V9.ooififiohmhsfla .6413 J .0 fa 1 .¢ 3.25....ia ....v. o. .I..‘o . 4...... ‘1‘ , . I: v” 333-. .. $004709 mm “fin-.44? . u 5S :0. "9...... So... .6, .‘n. :3. . 3 no. ...-a. on: ... ...fian V. A. It| 3 ..Jhp..«9k31.‘flh¢flllrcu .313 ...... that-‘11. .3... so). fl £91333“. v lat-n cud-60....345K1: ...-2a.. . 3.... can .00. SI.“ _ ... ...-00:: . .I . -.v| on... n.3ov...‘ . n I v. I. .... .. ... A ....‘oé... . to 0....- ..o 1' I: I . . A. . .4 ... n .. ..6 ...-..o . c :7 3. Q. ,. ... ... o 1. l5.-!-5 4. o..oi$v‘ 1.13.3.5... .. 4.3:. 0.. o o . ' to ... . u 0.... .s ..roi 3 I . I. .. ..-I.I’9.O.I".'§. '- . a . ... .. 33.... .o a... . ...! ... .a ,. .§.-..i-..o i. .. n .0 . . no .... s . ...! ..v-.:o .....- . 03. .u, v ...-0!.uooau‘ ... ....v.. . 233.3:50. 1.... a .95... . .. A. I. . o- v. .. . I . ... . u .u o u .....o.£3n"o¢ ....aplLSv: u. . . 0.. o. o... :0 .0 I... Q .4 "0t .0 ,~ .90... 5.5 .9... ‘7 .. 4 . o. .... o. . . .00. . a u . ... . nu. .4 7} b [III 4 5!... .I - 0.. u a .1. I nil-1f... n titito .l- 00“...... . ...-...; o .1. O. . 0‘. u. '10- ...»..‘h . .l. :- . .. .3. .. .‘vn: . . . . , .. . ... o‘- . u v... :0. .3... ...-.19.}: ... .3 .r . ~ _u t' . 1. . p .7. a 0A... '0 .....iuo". . ... . o $04.20;! . .. .. . 3 .o.n...oa...a07‘al. .... ... .....l G. . ...}... D. . . h. :-O.O.r O. . . . . :0. ...... 3.1.. . -4. ._ .... . ... oz . .- .. .- ..» . . . x 2 93:35. ...-0.!" . . o . ... p.‘ -. p... 3‘ It's. 3 .... . .a‘. 3 ..pa. .4 4 ‘0 o . y no v. .... .,. Qt! Slun'on It ‘0..- . u. u .o u .04., 4 ’..c‘-JI.:. . .0. 0.. .n .... I..- :11. 0.0.1.1 .. ..., . o a 0.. ...A.- .~ )...U . ...-o. , 0.. .00. .- . In. ‘- . .. L .4 . 3H,. .»\.o.£ . ...‘$.s‘).uo€2. £17.. Vthuuhvflhouaua-quoflio.8ahnu.noun.4. ...v . ... .0.- 3 .0: o‘,|’..to.~: S‘;!l,¢¢‘-. ‘53. . O.‘V~‘.oo .d‘aflot ’9‘ “09‘J\.’.}l"‘ 0‘0..- .0 V .‘.‘......L. .... o 43‘: to .o nu‘uOur .- a I .!b . . .. .... . 0.. .. A . o. .u 3...... ..- . . . on}. . Too... : .‘o z. ....Ii ...... ....3A.os..... .... vii-Etc!!! “filiifaazfi ...v . 31 ...-3.2...) .. 9... ‘g..$..o 1.1!. it: sthmntv. h; 0.". ...s.oo§..o.¢.r . . 0d n‘.‘.0 a .. u u . ...: 01:... .0 v . o .. ...,f c. t. a... .-'..u, in u. ..I I ‘9 L . o a ...... n o z .... .35. .5...‘I'\ ...utu '1'." .l..~o.-t.cu . .D.. .1. .0..- 901 o ... . O 01.3.. 4o. :- .. . ... gt: 0... .o 0.11. ... .. . .. ...O‘v‘..-n- ......o..5. - .. ... 4 0"..- 'cl‘ .. . .41 I I... - C . out no... 1(3......‘ ‘1 . . . . . .o Q. 4...... to I . .- , 312......c‘: ‘i :"D\I.u’$'! . .3." I., t . . 0 ‘.v . ...-.... .3 4. 3.0- v'; w » ..90. n . z’v‘. 30' c 106..- I’chlcznuuuo.31"§..r ..." a...§.3u-.5.’. .ooglé‘o-Iasz: o 1“"...013, lc‘ooo’u \15 o. A 4.... c.odb..lw.s...u: 3.3.“ .c. hex-’53“:- ov‘o .51... . .‘lvva... . .0. .4....... cl 3 I. .x ugh. ‘4.-‘¢..,- - :‘ .0. cult. 7. s.‘.lt )oOa 9:40: 4.. .uo'a 4O..In.elo..f.53si r. 3:) ... c . ... 0.? J .0. £0.13... ¢ o. ...-6‘ IL!“- .. $ 3; . . . .9: ...! .2 u. 0.. ..l- .3.’.: fi“ . o I. . .-,... ..II.-..O‘o:. - .It.... s 6.3. .c ...-I ... . . . a 1. “3:. . - 4 Cu’...‘ 39-..... ...... 1 ... 37.0%,.9. 2. ...-.... $...1!’5.3..:J . 2 . ...! . h:..9,0u.’a$uol¢:onlv$..0 . ... a o. . . . .OivOO . '0. T on Io t. O"uo-»¢4§0 on: .40. .0. .. o .0 o . .7 . ... .o .... . . 0' .- .... .1 on... coolnos... . . o . ‘IJ . . I o. ‘ . a .0 .. . o A u a . .v O!" r. ..In..ovu.:. : 9-,. ‘. 3......‘4. ...-C. .4! 3’... .u ‘ .......l.3.3 3.0.0:; 3.5.?! sq.) .oa’ltu..$!1§3.2tofu“: ,to.!8;?-Slu .1 0:... {rob-0...... - $321.1”). . 34:11. .70 it... 92?}: ..t. .1 .3, .l. 3‘.5...uu"!d:3‘o\! . ....p 0.0 o. . 0-.f-.§Dtoo...a... - o . Q“2pa.. C...) .v :7.- ‘L. .v. : a a?) 0.. DA. .. IOIVI‘,C . o _ ...... Y... .13. vac-1.39.5.al. . ....tv O.‘OI'.‘.L"." z 0.... 0 no. . ...-.0 on .- ! 0". o. ‘KQ . I ..‘v IQI‘ . . . on. 5.. ul!”!v!;’93.~. b. ‘ . . n . . 4 n c i A“. o. I '0'! 02. viii-‘36:...” 1’ .a 12... .I- .1 1' . . \ ....Il:w....$ 3...: 3 II a :0:- j .4.3.“..lu’9?.‘.ua¢l.....o 16;... 1.2.. a): 533%....1 u. .1. . .. p. .0 . 1.3. ...: :- ”20...... s...- .3 ..f. 4:. o .0 . o o..l..o..¢o¢.- .- :- :- 4 ...... log-7 ..\ S . . . $0. 0 2 41941. . . . .. ‘¢...fl..:oz 11.0.53-..- I I I u c n . - - 0".t‘. ... . .v . . 4 . . . . . 4 .o . o . . ..Oo . . . . . . - 4 . .. . u .o I .. . . a. . . . . . : . . . ‘0. £30 . o-o. A019. .4 VI. . ...c n o - . . . c. on .....O o to. .v v :0 b. .. . .10. cut C... 2v: a. . . ... n ...c.‘- .....nértu.‘ 0.. .03... .10: p .4 A o . u . o ‘2 .0- . . n. I . . in. ...... c on. .I . ... Xe . ‘ v . a: . Z 4‘ . ...vo‘.vo.s_v I. a a‘ :0 . w." 0.. ..3‘.’ - .. a... K. I ..‘c...... .00... . . u . I. f 1. . r. . a. I. . ...-.- - ....3 I .01.... ‘ 40. 4... ..IO 0 .4 ..IICV..01.- I. _ .4 .. . a . }.§-.2. ..d“ . o . 1a. . . . . 4 9.. ..‘U‘ .4. v. I . ‘- 4 . . . . . .. - . ..oodi.-a.\a. . - . . 4 . . . . .. . .4. . 4 - w. - . . .. . .4 . . 4. . . , .. . . . . . . w. n... . . . _: . . _ . . . . _ -..: . . . . .9... - D . 4 . . . ... u§ f.~.o‘..o O. . . 4 ..‘a no .5 }.u '0:- .. . v 4 c. - ... n. 0.0 v.5... . II 1.: 7 . .4 I: a .00. (4 o. . Y . v . . . o. 0.. p . .0: l....v¢ r . . on... ..t . 3.1!oOoh‘. . . v. .. .10 o'l‘... .0. .34 I.a.. . QAI¢U'-. . a .. .... ‘9... o .I. . .. Poi!..... . . .l.nl.,~ot. .. ,n 190:. o _. ... . - .... .- .. .. .. ...‘\7(...n.. . v . . .. . .. . . . 3:1. .. V. . '0‘. ..4 1.01 no... ‘1 . . . 4. .. I I . .1. x y . .ou \. ... ~ ... ? - 39.119.- . u . ,1. . - . .. . . . , n‘ w. . . .. ,|’ .v v Q .i.,..o...o.-: v. - 1"2. . :4 .o .u.. ....o. Y .o.|uq.i. o .. . . .0..- 1 . o-o‘4n.)a ‘ _. ... .63.. v. cl. .1!- l|.u.t . . . u ....‘0 c - 4.1vangg‘1 ‘~ 7 s . .... I. . ct ..o .0... ‘ 9.. - ... - .... on. out C u I .o . o n I 5 ‘ . v.13 . . a. . .0. ..‘I .... -..... 4 D .. .oa ...... . 7.. ...-0‘ . I: - . 1‘... s‘\.&.¢n‘l-o ... .. Out 3...... o . O... O... 'U: 4 40A. . .. . .39. u... 0 710.50. . . I .. 4 . . . e .0 v! o .. ,u o t c . . . . c. z. r . . . K . .... . l. 0.; ... 0.0... 4 .0. n } .... a: . u u v . n .4 v... . .n ‘ .. 1.. u ....lanlvo‘tno30t5a4 ... . . ‘ a . v. - .0 1-. ~ o. . . .. Ooll}..!.". . c Q I‘ on; 50 at o i n .. 1.. . . ‘4... o.«. W... . . . l O a '4 4‘4. .. ... o . 0' III. p v: 0.... u I. . .I. O ... ...O .o t. 7 . o . ..O‘ 6‘ I 2 u a .. . . n . ... c‘ . ... o c.- ..J o'...1.l 1.... . 0 t t. u . i . u n . . o. I . l ... ii. I .8. n . 00.: .- ‘ n i .- . . n . . . u. .9 .woC... . 4 . .. , . «5...!» .334. ..ME . u ._ - _ - .....-7: u... 5%.... . .. a......f....m.....i.p. \ t oo o.... I... It ... .4 1.. .u ' ‘10.me Yaw .VJJ'I:§.0”:£HO-4.Vfla LWVeru‘u.m?fio AWN-3 fi‘L‘ "J "— u‘.0 .n “- fia- \‘ O“ . "‘. x. ._ a u - v .. . .I '0. 3.7.. . 3- P. to. .. . ...... vi. ...: ...l 0.0... u ......v'lfi. .0. ... o .. o I... . ‘caoslgz.’ . a. . rv’E! . .. u I ... oJ . Q .i . .5 . |. u: ... . . . . . 2;": v «a! . 0 ...I o 0.. .3 . o ‘0‘ 4 Iéu‘ coinrif 5).: o, .i ...: . . u u .- otl'l...ol.u1u.?c:¢‘d c . tvul...ob.libva‘..o.o?. a? ‘paczkmm . p ...”...o .ho. m. I... ~92.on .‘. . Pl... . _ ¢ 0". p . . 1A.. . u .0 . O I. 1' ..v..".au.v. I. "t... V‘ 0.. ... . . D o 5%.. . _ . Q n ‘..l . ‘19... ~02. a .p 0|. . . .- I...Q.-.u... . E, c.» O, I 0.5- .' .‘oa.-n.o‘v’..§¢- Noam." ’0 3.59;.0 C ..‘é‘u‘ 0" u..“hqlo-‘nl".-.o 5... . . . . g. . o. ‘o. . . .1i , 4 ..o - a... Iv... ...-..- cw...» ... OI... ... .x...» .. .. ..Ia. 1.2.. A... ...4 ch: . o O . a». In. D. I: ..u 0.9.40.0..33. 20...}. E .Ai.’o.:uO ml?...!r.;l no“. o .. .. u. . o - . 3 9 _. .4. v .32.... . . . v .....J .3.. of... m . c . 1 . .14; ... 2.1: o .35; ...: I... ... :HMSIS . “¢.}.I81 3 1:7 .7 .....l. ...-.4... K». .. ......ooo . v.38.- ..I . . “I. o 0.. .o l on 9 5.. . ... . .... 4!. .Q .... ... «.0: .. ‘0. ...: .23...“ 2.. 1313.1. .. ‘ .n¢.:‘¢n. 0.0 . . . ... . I... . . o c .. ....Q-. '00... ....o . .40.”. ‘u . o o o . ...“. . 0.0.. o .... O . n. o. ...r. . . a ... n .- 9 . . . I 'Il “TESTS 1003 LIBRARY Michigan. State UnWemmy This is to certify that the dissertation entitled BIOMECHANICAL RESPONSE OF THE KNEE TO INJURY LEVEL FORCES IN SPORTS LOADING SCENARIOS presented by ERIC G. MEYER has been accepted towards fulfillment of the requirements for the PhD. degree in Mechanical Engineering flew/aka” VMajor Professor’ 3 Signature flag //1 07009’ / / Date MSU is an Allirmative Action/Equal ()ppm'runih' Employer PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 KzlProj/AccsPresICIRCIDaIeDue.indd BIOMECHANICAL RESPONSE OF THE KNEE TO INJURY LEVEL FORCES IN SPORTS LOADING SCENARIOS By Eric G. Meyer A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Mechanical Engineering 2009 ABSTRACT BIOMECHANICAL RESPONSE OF THE KNEE TO INJURY LEVEL FORCES IN SPORTS LOADING SCENARIOS By Eric G. Meyer Injuries to the knee are among the most common injuries in sports. A frequent and serious sports injury is that of an anterior cruciate ligament (ACL) rupture in the knee. Injury mechanisms have been documented from sports medicine patients suffering non- contact ACL tears. A hypothesis of the study was that the external tibial and valgus femoral rotations frequently identified after ACL injury are not representative of the relative displacements that cause isolated ACL failure to occur. The ACL is the primary restraint for anterior tibial subluxation, a co-primary restraint for internal tibial rotation and hyperextension, and a secondary restraint for valgus bending. A second hypothesis of this study was that tibiofemoral compression will produce anterior tibial subluxation and isolated ACL injuries, while other loading mechanisms will produce combination ligament injuries. Lower extremity joint injury is often accompanied by undiagnosed cartilage or bone damage in the form of fissures or “bone bruises”, respectively. Post- traumatic osteoarthritis has been demonstrated to occur at a high incidence rate following ACL injury. This disease may be initiated by the acute compressive trauma that occurs at the moment of ligamentous injury. A final hypothesis of the study was that the mechanism-based clinical classification of knee injuries and bone bruise patterns would correspond to characteristic distributions of high levels of contact pressure and osteochondral microdamage across the tibial plateau for each loading mechanism. The four specific loading mechanisms investigated here were; tibiofemoral compression, internal tibial torsion, hyperextention and valgus bending. This dissertation combines human cadaver studies with a computational model to validate the bone bruise “footprint” patterns associated with each injury mechanism. The peak forces/moments, relative knee joint displacements/rotations and type of failure were documented. Isolated ACL injuries occur from tibiofemoral compression, but internal tibial torsion and valgus bending caused combined medial collateral ligament injury. Hyperextension caused combined posterior and anterior cruciate ligament injuries. Tibiofemoral compression produced anterior tibial subluxation leading up to ACL injury. After failure, there were significant increases in external tibial rotation and valgus knee bending. Therefore, the vertical ground reaction force and muscle contraction that produces tibiofemoral compression should be considered as an important loading mechanism for studies of sports ACL injury scenarios. In addition, each loading mechanism produced distinct contact pressure distributions which correlated well with the location of osteochondral microdamage. The tibiofemoral compression, hyperextension and valgus bending loading mechanisms produced regions of contact pressure exceeding 30 MPa. In the computational model, this contact pressure produced maximum shear stresses in the articular cartilage, subchondral bone and trabecular bone which exceeded the threshold for predicted tissue damage. Therefore, even if knee joint motion is constrained and the ligamentous injuries are prevented, there is a long term risk of developing post-traumatic osteoarthritis from these levels of knee loading. The data presented in this dissertation may be applicable to injury prediction/prevention, and for clinicians to help diagnose injuries associated with ACL trauma. DEDICATION Thank you Suzi, for your unwavering love and support. With you anything is possible. iv ACKNOWLEDGEMENTS I would like to thank my mentor Dr. Roger Haut for his expertise, leadership, support and dedication throughout my research at the Orthopaedic Biomechanics Laboratories (OBL). I would also like to acknowledge Drs. Eann Patterson, Dashin Liu, G. Thomas Mase, and John Powell for their insightfiil advice and for serving on my committee. I also wish to gratefully acknowledge Drs. Michael Shingles, Walter Smith, Segin Baek, Robert Wiseman and Jillian Slade for insightful clinical discussion, as well as computation modeling and radiological assistance. I would like to thank Clifford Becket, Jane Walsh and Jean Atkinson for their help and advice. I also thank Dean Mueller, Director of the Anatomical Donations Program, University of Michigan for help to procure these test specimens. Finally, I would like to acknowledge everyone who worked along side me at the OBL; Michael Lavagnino, Dan Isaac, Tim Baumer, Mark Villwock, Feng Wei, Jerrod Braman, Nuirit Golenburg, Steve Rundell, Lynn Martin, Mike Sinnott and Dan Phillips for their help and friendship. I am grateful for the financial support I received from the Orthopaedic Biomechanics Laboratories, the College of Osteopathic Medicine, the Department of Mechanical Engineering and the College of Engineering. This study was supported by a grant from the Centers for Disease Control and Prevention, National Center for Injury Prevention and Control (CEOOO623). TABLE OF CONTENTS LIST OF TABLES ............................................................................... viii LIST OF FIGURES ................................................................................. x CHAPTER 1: Introduction Injury in Sports ............................................................................. l Knee Biomechanics ........................................................................ 5 Acute Knee Injuries ........................................................................ 20 Post-traumatic Osteoarthritis ............................................................ 33 Summary and Objectives ................................................................. 39 References ................................................................................. 41 CHAPTER 2: Anterior cruciate ligament injury induced by tibiofemoral compression producing anterior subluxation of the tibia in unconstrained human knee joints. Abstract .................................................................................... 57 Introduction ................................................................................. 58 Methods .................................................................................... 59 Results ...................................................................................... 62 Discussion ................................................................................. 68 References ................................................................................. 74 CHAPTER 3: Osteochondral microtrauma in regions of high contact pressure associated with tibiofemoral compression during ACL rupture. Abstract ..................................................................................... 76 Introduction ................................................................................. 77 Methods ..................................................................................... 78 Results ...................................................................................... 83 Discussion .................................................................................. 88 References .................................................................................. 94 CHAPTER 4: Anterior cruciate ligament injury induced by internal tibial torsion and the associated osteochondral microtrauma in regions of high contact pressure. Abstract ..................................................................................... 97 Introduction ................................................................................ 98 Methods .................................................................................... 100 Results .................................................................................... 103 Discussion ................................................................................ 1 10 References ................................................................................ l 16 CHAPTER 5: Tibiofemoral contact pressures and osteochondral microtrauma during ACL rupture due to hyperextension of the human knee. Abstract ................................................................................... 1 18 Introduction ............................................................................... 119 Methods ................................................................................... 120 vi Results .................................................................................... 123 Discussion ................................................................................ 128 References ................................................................................ 132 CHAPTER 6: Tibiofemoral contact pressures that induce cartilage damage and subchondral bone microcracks during valgus bending of the knee. Abstract ................................................................................... 135 Introduction ............................................................................... 136 Methods ................................................................................... 137 Results .................................................................................... 141 Discussion ................................................................................ 146 References ................................................................................ 15 1 CHAPTER 7: Tibiofemoral contact pressures generate cartilage and subchondral bone stress during ACL rupture: A finite element analysis. Abstract ................................................................................... 1 54 Introduction ............................................................................... 1 56 Methods ................................................................................... 158 Results .................................................................................... 164 Discussion ................................................................................ 170 References ................................................................................ 1 75 CHAPTER 8: Discussion Cause of Isolated ACL Injury ........................................................... 178 Tibiofemoral Compression ........................................................ 181 Post-traumatic Osteoarthritis ........................................................... 185 Conclusions ................................................................................ 189 References ................................................................................ 191 RESEARCH PUBLICATIONS ................................................................. 193 APPENDIX ........................................................................................ 196 vii 1.1: 1.2: 2.1: 2.2: 2.3: 2.4: 2.5: 2.6: 2.7: 2.8: 3.1: 3.2: 3.3: 3.4: 3.5: 4.1 4.2: 4.3: 4.4: 4.5: 4.6: 5.1: 5.2: 5.3: 5.4: 6.1: LIST OF TABLES Structural properties of human cadaver ACL specimens in tension ..................... 9 Structural properties of the ACL and grafts in tension ..................................... 33 Series 1 specimen information and test parameters ....................................... 61 Series 2 specimen information and test parameters ....................................... 61 Data from series 1, pre-failure experiments .................................................. 63 Data from series 2, pre-failure experiments .................................................. 63 Data from series 1, failure experiments ....................................................... 66 Data from series 2, failure experiments ....................................................... 66 Dissection documentation of injured structures following series 1 failure tests ...... 67 Dissection documentation of injured structures following series 2 failure tests 68 Histological scoring to quantify cartilage and subchondral bone damage ............ 82 Pressure film data from series 1 subfailure experiments ................................. 84 Pressure film data from series 2 subfailure experiments ................................. 84 Pressure film data from series 1 failure experiments ...................................... 85 Pressure film data from series 2 failure experiments ...................................... 85 : Torsion specimen information and test parameters ...................................... 101 Data from internal tibial torsion pre-failure experiments ................................. 104 Data from internal torsion failure experiments on each specimen .................... 106 Dissection documentation of soft tissue injury following failure tests ................ 106 Pressure film data from subfailure experiments ........................................... 108 Pressure film data from failure experiments ............................................... 108 Hyperextension specimen information and test parameters ........................... 121 Peak force during pre-failure and failure tests and knee joint injury .................. 124 Pressure film data, average (SD), from sub-failure and failure tests ................. 126 Pressure film data, average (SD), from sub-failure and failure tests ................. 126 Valgus bending specimen information and test parameters ........................... 139 viii 6.2: 6.3: 6.4: 7.1: 7.2: 8.1: 8.2: 8.3: Maximum valgus bending moment and knee joint injuries .............................. 142 Data from maximum contact pressure distribution ........................................ 143 Maximum pressure and cartilage and subchondral bone damage ................... 144 Maximum Tresca stresses in the articular cartilage ....................................... 167 Maximum Tresca stresses in the subchondral bone ...................................... 168 Joint motions prior to failure for each loading mechanism .............................. 179 Failure forces and injuries sustained by the ligaments of the knee ................... 183 Classification of loading mechanism, bone bruise patterns and injuries ............ 188 ix 1.1: 1.2: 1.3: 1.4: 1.5: 1.6: 1.7: 1.8: 1.9: 1.10 1.11 1.12: 1.13: 1.14: 1.15: 1.16: 1.17: 1.18: 1.19: 1.20 1.21 1.22 1.23 1.24 1.25 1.26 LIST OF FIGURES Anatomical structures of the knee ................................................................ 6 Anatomical features of diarthrodial joints such as the knee ................................ 6 Knee muscle groups and their motion effect ................................................... 7 The effect of age and orientation on the ultimate load of the ACL ....................... 9 ACL load-elongation at two orientations ........................................................ 9 Tensile response of the ACL divided into four regions based on load ranges ....... 11 Layered structure of cartilage and subchondral bone .................................... 12 Osteoarthritis disease progression from healthy to end stage ........................... 13 Flexibility method for the normal and ACL-deficient knee joints ........................ 15 : Knee ligaments and their function during tibial rotation ................................. 16 : ACL force generated from internal or external tibial torque ............................. 17 ACL force generated from varus or valgus bending moments ......................... 17 Hall-effect strain transducer attached to the ACL .......................................... 18 Tilt of the tibial plateau and resultant anterior displacement of the tibia ............. 19 Frequency distribution of knee laxity between normal and ACL-deficient knees .. 20 Video sequence of a contact ACL injury to an NFL player .............................. 24 Landing from a jump ACL injury mechanism ............................................... 24 Video frame and time sequence of the ACL injury to a NBA player .................. 25 Hyperextension loading method for producing anterior knee dislocations ......... 27 : Valgus bending method for producing knee injuries ...................................... 28 : ACL rupture experiments from excessive force in the quadriceps tendon .......... 28 : Stop-jump landing and the resultant proximal tibial anterior shear force 31 : Forces acting on the lower leg during the simulated drop-landing .................... 31 : Radiographic diagnosis of CA by loss of joint space. .................................... 34 :Valgus bending injury mechanism causing bone bruise ................................. 37 : Hyperextension injury mechanism causing bone bruise ................................. 38 1.27: Surface fissure of the articular cartilage and microcrack of subchondral bone 38 2.1: Knee specimens attached to the compressive testing fixtures ............................... 60 2.2: Representative load/moment versus time plots for pre-failure and failure tests 64 2.3: A representative load and motion versus time plot during a failure test ............. 65 2.4: A representative load and motion versus time plot during a failure test ............... 65 2.5: Anterior cruciate ligament rupture and avulsion specimens .............................. 67 2.6: Relative TF joint motion during compression experiments ............................... 68 2.7: Sagittal view of the relative TF joint motion during compression experiments ...... 70 2.8: Coupled internal rotation of the tibia and valgus rotation of the femur ................ 70 3.1: Knee specimens attached to the TF compression testing fixtures ..................... 79 3.2: Representative pre-soak and post-soaked MRI sagittal slices .......................... 81 3.3: Photographs of representative cartilage and subchondral bone microtrauma 83 3.4: Representative shape and magnitude of the pressure distribution ..................... 85 3.5: The relative increase in signal intensity between pre-soak and post-soak scans ...86 3.6: Average 1: SD histological scores for regions within the medial and lateral facets. 87 3.7: Gross surface fissures stained with India ink 87 4.1: Diagram of the torsion testing fixture ......................................................... 101 4.2: Knee specimen attached to the torsion testing fixture ................................... 102 4.3: Representative torque versus time plots for pre-failure and failure tests ............. 103 4.4: A representative torque and motion versus time plot during a failure test .......... 103 4.5: Representative shape and magnitude of the pressure distribution ................... 107 4.6: The relative increase in signal intensity of the tibial plateau cartilage ............... 109 4.7: Average 1: SD histological scores for regions within the facets ........................ 110 4.8: Gross surface fissures stained with India ink for representative specimens ....... 110 4.9: Sagittal view of the relative TF joint motion during torsion experiments ............ 111 4.10: Sagittal view of the relative TF joint motion during torsion experiments ........... 112 4.11: Coupled internal rotation of the tibia and valgus rotation of the femur ............. 112 xi 5.1: Diagram of the hyperextension testing fixture ............................................... 121 5.2: Knee specimen attached to the hyperextension testing fixture ........................ 122 5.3: Representative bending moment vs time plots for pre-failure and failure tests 124 5.4: Rotation of the tibia and femur and total knee hyperextension ........................ 125 5.5: Joint displacements for pre-failure and failure tests ...................................... 125 5.6: Representative contact pressure distributions for hyperextension failure ........... 126 5.7: Average 1 SD histological scores for regions in the medial and lateral facets 127 5.8: Gross surface fissures stained with India ink for representative specimens 128 6.1: Diagram of the valgus bending test fixture ................................................... 139 6.2: Knee specimen attached to the valgus bending testing fixture ......................... 140 6.3: Representative contact pressure patterns for valgus bending ......................... 143 6.4: Gross surface fissures stained with India ink for representative specimens 145 6.5: Gross cartilage damage and corresponding subchondral bone damage ........... 145 7.1: Representative contact pressure patterns .................................................. 156 7.2: Knee joint geometry ............................................................................... 159 7.3: Layer of articular cartilage created on tibia plateau ....................................... 160 7.4: Tissue zones and material properties for bone and cartilage .......................... 161 7.5: Contact model with input displacement along tibial axis ................................. 162 7.6: Generalized pressure distribution patterns ................................................. 163 7.7: Representative pressure distribution ......................................................... 164 7.8: Contact pressure distribution for the preliminary knee model .......................... 164 7.9: Contact results for an applied displacement along the tibial axis ..................... 165 7.10: Cartilage surface Tresca stress distributions ............................................. 166 7.11: Percentage of specimens with cartilage microdamage in each region ............ 167 7.12: Coronal slices through the cartilage at the point of maximum Tresca stress 169 7.13: Percentage of specimens with subchondral bone microdamage .................... 170 8.1: Percentage of specimens with cartilage and subchondral bone microdamage 186 xii CHAPTER ONE INTRODUCTION INJURY 1N SPORTS Sports Participation Participation in sports, recreation and exercise is increasingly p0pular and widespread in American culture. Approximately 50% of youths (aged 5-17 yrs.), accounting for more than 25 million people, regularly participate in vigorous physical activity (Cosgarea & Schalzke, 1997). This includes approximately 22 million children participating in agency sponsored sports (Little League, Pop Warner, etc.) and almost 6 million adolescents in interseholastic sports (Seefeldt and Ewing, 1996). The 1997 Centers for Disease Control and Prevention “Guidelines for schools and communities for promoting lifelong physical activity” states that the benefits of regular physical activity in childhood and adolescence are; improves strength and endurance, helps build healthy bones and muscles, helps control weight, reduces anxiety and stress, increases self esteem and may improve blood pressure and cholesterol levels. There are also estimates of 25 million middle-aged people, “weekend warriors,” participating in recreational sports. This number, currently between 15-20% of the US. population, depending on age and gender characteristics (US Dept Health, 1998), continues to rise with emphasis on physical fitness in all age groups. The benefits of moderate and vigorous physical activities are also clear in adulthood. In a study of 50-70 year old Harvard alumni, greater energy expenditure was associated with increased longevity and was proportional to the level of vigorousity of the activities (Lee and Paffenbarger, 2000). These benefits also included decreased blood pressure, decreased risk of coronary heart disease, hypertension, colon cancer and diabetes and decreased obesity (Pate et al., 1995). In elite athletes, the benefits of exercise are also a longer life expectancy and lower risk of ischemic heart disease and diabetes (Raty et al., 1997). Along with a general emphasis on physical fitness, female participation in sports has seen a dramatic increase since the passage of Title IX in 1972. In 1974, the percent of female athletes was only 32% of males’ participation (Seefeldt and Ewing, 1996), but that figure climbed to 70% in 2007 (NFHS, 2008). In high school basketball there are nearly 450,000 female participants in the US. (NFHS, 2008), however in popular sports like basketball and soccer, girls have shown a much higher risk of injury than boys (Powell and Barber-Foss, 2000). Injury incidence Each year between 3 to 4.5 million children and adolescents are injured during sports participation (Conn et al., 2003; Hergenroeder, 1998) with more than 775,000 of the injuries to young athletes requiring physician visits. An estimated 800,000 lower extremity injuries are sustained nationally by American high school athletes (Fernandez et al., 2007). Adult sports participants report musculoskeletal injuries at nearly twice the frequency of sedentary individuals (27-31% versus 15-17%, respectively) (Hootman et al., 2002). In Scandinavia, sports injuries represent 10-19% of all acute injuries treated in emergency rooms (Bahr and Krosshaug, 2005). Llong-term participation in vigorous physical activities increases the risk of acute and chronic injuries, such as ligament sprains or osteoarthritis (OA), respectively (Hootman et al., 2002). Injuries to the lower extremity are among the most frequent injuries in all levels of sports and often account for more than 50% of reported injuries (Fernandez et al., 2007). The knee is one of the most common sites of sports injury requiring an emergency room visit (Bahr and Krosshaug, 2005). In particular, one in ten female athletes at the intercollegiate level suffers a season-ending knee injury annually (NCAA, 2001). Additionally, female athletes participating in jumping and cutting sports have a three to six-fold higher incidence of serious knee injury than males (Ferretti et al., 1992; Hewett et al., 1999; Powell and Barber-Foss, 2000). Risk factors for injury A comprehensive model to describe the etiology of sports injuries would be very complex and include both intrinsic (age, sex and body composition) and extrinsic (shoe/surface traction, bracing and environmental factors) risk factors for a particular athlete as well as an understanding of the inciting event (injury mechanism) that is associated with the onset of injury (Krosshaug et al., 2005). Each of these variables will have an effect on the load level in the lower extremity as well as the athlete’s tolerance to that load before an injury occurs. Intrinsic risk factors are generally difficult to control in order to reduce a particular athlete’s likelihood of a lower extremity injury. Each athlete has a different set of these variables which can predispose them to more frequent injuries than an athlete with a different level of maturity, joint anatomy, or history of injury. Other intrinsic factors can be modified slightly with training. Physical fitness can be improved, skills in performing certain tasks, such as landing can be taught and psychological motivations, like competitiveness can be monitored. Exposures to extrinsic risk factors are easier to control, provided the variables linked with an increased susceptibility for injury can be identified in the first place. Therefore, knowledge about the types and magnitudes of force that produce injury is needed in order to implement policies that protect athletes from unnecessary risk. Examples of such policies to increase safety for the participants are rules that ban certain activities, mandated sports equipment, and the design of competition sites. In some cases, however, tradeoffs and compromises make it difficult to control the extrinsic factors. While linear friction at the shoe-surface interface is necessary for athletic performance (Shorten et al., 2003), it is generally accepted that excessive rotational friction between the shoe and surface induces dangerous forces in vulnerable anatomic structures and may be a factor in lower extremity injuries. A study of ACL injuries in high school football players documented a significant relationship between cleat design and the amount of torsional friction and the risk of ACL injury (Lambson et al., 1996). In fact, the cleat design with the highest torsional moment was associated with an ACL injury rate two and a half times higher than of all other cleat designs combined. Although all injuries were sustained on natural grass, the authors also measured the fictional moments of each cleat on artificial turf and found these values to be even higher. Differences in injury frequency have also been documented for natural grass versus Astroturf surface types. These differences in injury frequency may be due to changes in the structure and materials (Hammer, 1981), the running speed of the players (Stanitski et al., 1974) or the coefficient of fiiction between the surface and shoe (Andreasson et al., 1986). To complicate matters further, injury risk also depends on the player’s position and the type of play at the time of injury, both of which influence the loading mechanisms on the lower extremity. KNEE BIOMECHANICS Knee anatomy The knee is a synovial joint in the leg where three bones, the femur, tibia and patella meet (Figure 1.1). The femur and tibia are two long bones in constant contact that rotate relative to each other (similar to a hinge) in order to produce knee flexion. The femur has two articulating surfaces, the medial and lateral condyles that are in contact with the medial and lateral tibial plateaus. The patella, or kneecap, is a sesamoid bone that rests on the anterior articulating surface of the femur to protect the knee and act as a lever for transferring muscle forces between the upper and lower leg. Diarthrodial joints, such as the knee, allow movement by transferring forces between muscles and bones with very little friction, while also providing cushioning and distributing the forces over large areas (Figure 1.2). The knee has two fibrocartilaginous pads, the medial and lateral menisci between the femur and tibia which act as shock pads to cushion forces transmitted between the tibia and femur. In addition, articular cartilage on each bone’s articulating surface provides nearly frictionless motion and additional contact force distribution between bones. There is a fibrous capsule of synovium, ligaments and tendons that hold the joint together, as well as maintain the synovial fluid in a “closed system”. The exterior of the knee joint is made up of the medial collateral ligament (MCL), lateral collateral ligament (LCL), patellar ligament, and quadriceps tendon. The anterior cruciate ligament (ACL) and posterior cruciate ligament (PCL) are located in the interior of the knee joint. Femur (thighbone) Patella (underside) Anterior Trochlea C-rUC'ate (patellofemoral Ligament groove) Lateral Posterior Femoral 1 ,. ~ Cruciate Co ndyle I. '. . ; nga ment Medial Lateral . :l Collateral Meniscus ' Ligament Medial Lateral Meniscus Collateral ' Ligament Tibial Plateau Tibia (shinbone) Fibula Tibial Tuberosity Figure 1.1 Anatomical structures of the anterior knee (Self Medical Graphics). Figure 1.2 Anatomical features of diarthrodial joints such as the knee (Mow and Mak, 1987) *Irnages in this dissertation are presented in color.* The knee joint primarily functions as a hinge to produce flexion between the articular surfaces femoral condyles and tibial plateaus (Kapandji, 1987). The major muscle groups that produce knee flexion and extension are the hamstrings and quadriceps, respectively (Figure 1.3). The hamstrings are really four individual muscles that originate along the edges of the pelvis and insert at the proximal tibia and fibula. The quadriceps also encompasses four individual muscles that originate on the pelvic girdle and insert into the patella and reach the tibial tuberosity via the patellar ligament. Muscle force plays an important role in the stability of the knee by assisting the knee ligaments in constraining joint motion. Figure 1.3 Knee muscle groups and their motion effect (Seif Medical Graphics). Tissue properties and failure Ligament: Ligaments are sofi tissues that connect two bones together, and generally have high strength and high stiffness (Amoczky, 1983; Butler et al., 1989; Girgis et al., 1975; Kennedy et al., 1974). The primary cells in ligamentous tissue, fibroblasts are surrounded by an extracellular matrix that is composed of highly oriented macromolecules and water. The macromolecules include a majority of collagen (75% of dry weight), along with lower concentrations of elastin (<5%), proteoglycans (<1%) and glycoprotiens (2-3%). The major strength-bearing elements are the collagen fibers, which are composed of type I (90%) and type III (10%) collagen throughout the midsubstance and type II and type X collagen at their insertions into bone. Water (60% of wet weight) and ground substance (glycosaminoglycans, proteoglycans, and glycoproteins) provide lubrication and spacing between fibers and with other tissues. The fluid phase is also responsible for viscoelastic properties of ligaments and for diffusion of nutrients and metabolites to the cells. Some ligaments, such as the ACL do not have a good blood supply, and therefore do not heal well. Others, such as the MCL, have a robust vascular innervation and therefore heal more rapidly when damaged. Nerve fibers and receptors are primarily located at the bony insertions, therefore tearing the midsubstance of a ligament such as the ACL can occur with relatively little pain. The composition is variable between ligaments, depends on the location within a particular ligament, and changes with age and activity as the tissue responds to the mechanical environment (Woo et al., 1991). These changes in composition, especially of the collagen fibers, affect the ligament’s tensile properties. The ACL, in particular, loses 25-40% of its stiffness and 50-70% of its strength between younger (16-28 yrs) and older (34-86 yrs) groups of individuals (Table 1.1, Figure 1.4). The ultimate failure strength of the ACL also depends on the orientation and rate of loading during experimental testing. When the ACL is aligned with the axis of the tibia its failure strength is 25-33% lower than in its normal anatomical axis at 30° of knee flexion (Figure 1.5). Table 1.1 Structural properties of human cadaver ACL specimens in tension. Stiffness (Nlmm) Strength (N) Younger Older Younger Older Noyes and Grood, 182 (33) 129 (39) 1725 (269) 734 (266) Raueh et al., 1987 203 (34) 124 (39) 1716 (538) 814 (356) 30007 2000 1 1000‘ Ultimate Load (N) . Anatomical Orientation O Tibial Orientation 0 v 20 V i V ' V V 40 50 60 70 80 90 100 Donor Age (Years) Figure 1.4 The effect of age on the ultimate load of the ACL (Woo et al., 1991 ). 250° ' —— ANATOMICAL AXIS --- TIBIAL AXIS 2000 - 1500 - I 1000 - I I I I I 500 - . I I I ’ i I I I i I I I I u 0 2 4 6 81012141618 ELONGATION (mm) Figure 1.5 ACL load-elongation at two orientations (Woo et al., 1991). Defining the material properties of ligaments has been somewhat difficult, due to the complex anatomy and compositional organization of these tissues. The average elastic modulus of the ACL is 278 MPa, while the ultimate tensile strength is 40 MPa and occurs at approximately 12% strain (Butler et al., 1980). A more appropriate way of describing these material properties would be a viscoelastic, non-linear stiffening response. For relatively low strain rates (.66%/sec to 66%/sec) there is little viscoelastic influence on the tissue’s stiffness, but the failure strain and energy absorbed before failure are significantly decreased compared with the higher strain rate (Noyes et al., 1974). In addition, at a higher rate of loading, the failure mechanism is more commonly ligamentous rupture, versus avulsion at a lower rate. Non-linear stiffening occurs due to progressive recruitment of crimped collagen fibrils and can be described in terms of four regions; clinical response, physiologic loading, microfailure and complete failure (Figure 1.6) (Butler, 1989). The physiological region is nearly linear and can range fiom approximately 169 N of force seen in the ACL during normal walking (Morrison, 1970) to 630 N of force during jogging (Chen and Black, 1980). During the microfailure region and leading up to complete failure there are a series of unloadings that occur due to failure of individual collagen fibril bundles (Noyes and Grood, 1976). 10 COMPLETE FAILURE 3 PHYSIOLOGIC 0 LOADING 5:25 5 50° :33: -‘ CLINICAL 35:52: O O ......A.. J TEST 0 :::::::Eiiiii I /,ngf JOlNT DlSPLACEMENT (mm) Figure 1.6 Tensile response of the ACL divided into four regions based on load ranges v.v.v . O O O O O 0 0.0... O O. O 0.. O. . '0 (Butler, 1989). Cartilage: The frictionless motion and cushioning properties of diarthrodial joints are provided by the articular cartilage (AC) layer on the contacting surfaces of the bones. This material is a connective tissue that is composed of cells (chondrocytes), a fibrous matrix (collagen), a ground substance (proteoglycans) and interstitial fluid (mostly water). The solid phase (chondrocytes, collagen and proteoglycans) accounts for 15-30% of the wet weight of AC. The remaining 70-85% of the weight is water that pressurizes the cartilage. Proteoglycans have a negative charge that attracts electrolyte ions. This creates an osmotic gradient between the interstitial and the extracellular fluid. Collagen fibers provide the structural support for the surface tension that is developed by the pressurized cartilage. As the cartilage is loaded and compressed during normal activities, fluid is squeezed out, similar to squeezing a sponge. Articular cartilage is avascular (no blood supply), so the fluid flow is important for transporting nutrients and waste products into and out of the cartilage. 11 Articular cartilage is characterized by a high degree of structural anisotropy (Figure 1.7), however at birth and during development the tissue is more isotropic in organization. The equilibrium tensile modulus of adult human knee cartilage is 5-20 times higher in the superficial zone than in the middle and deep zones (Akizuki et al., 1986). By contrast, the confined compression modulus of bovine knee cartilage is significantly higher in the deep versus the superficial zone (Schinagl et al., 1997). These differences are attributed to the orientation of collagen fibrils parallel with the surface in the superficial zone (Williamson Et al., 2003) and to the higher content of collagen and glycosaminoglycan (GAG) in the deep zone (Klein et al., 2007). Articular surface Zones Superficial tangential -[ 37/— m\§ (10-20%) “”966:- _ 3L /“*-' "'5 Middle (40-60 A.) Aségj/ J/fi’xx' % o 1 “ W. l Deep (30/o)__x- t (H- . ‘ dirt. ... T'demark .- _~,.Xc.gh J . i i ’ , -—Subchondral bone ‘ Cancellous bone Calcified cartilage /& -- Figure 1.7 Layered structure of cartilage collagen network and interface region with subchondral bone (Mow and Mak, 1987). Age related changes in the mechanical properties of articular cartilage differ between locations within the body (Kempson, 1991) and within the superficial, middle and deep zones (Kempson, I982). The Young’s modulus of cartilage increases until approximately 30 years of age at a value of 123 MPa (Ding, 2000). Between 30—50 years of age the material properties remain constant and then show a significant decline in the 12 following years. The thickness of cartilage also decreases with age and especially with OA (Ding, 2000). Cartilage fissures are fiequently documented in cadaver and animal research studies of blunt impact. The number of fissures and their depth are related to the applied impact load and the contact pressure on the surface. They have also been shown to increase over time post irnpact in a rabbit model (Haut, 1989). There are two theories for how OA initiation and progression occur (Figure 1.8). The first is that fissures and damage of the AC occur due to a mechanical insult thus changing ability to adequately absorb and transmit loads. This initiates bone remodeling with increased stiffness that in- tum further damages the overlying cartilage and the cycle continues. The other theory involves a similar cycle, except the mechanical insult causes trauma initially to the SB in the form of occult microcracks or “bone bruises”. These cracks initiate remodeling of the SB, which in turn damages the AC by increasing the stresses seen in the overlying tissue, and the chronic degradation cycle begins (V ellet et al. 1991). Healthy articular Thickening 0f the cartilage surface ——> subchondral bone Surface fibrilation Complete erosion .3 ' oftheAC .., 1 . Figure 1.8. Osteoarthritis disease progression from healthy to end stage. 13 Bone: Bone is composed of a collagen network, a mineral phase and three types of cells; osteophytes, osteoblasts and osteoclasts. The joints of the lower extremity are composed of a hard shell of cortical bone creating the subchondral plate and trabecular bone filling in the diaphaseal ends. The thickness of the subchondral plate varies across the articular surfaces, in particular between the medial and lateral tibial plateaus. The underlying trabecular bone also varies in material properties, especially centrally beneath the cruciate insertion points. The medial tibial plateau has a central region of increased strength, and while the lateral tibial plateau is generally weaker, the posterior region does have slightly elevated trabecular bone properties (Hvid, 1988). Loss of bone mineral density is the main factor in the decreased Young’s modulus and ultimate stress of trabecular bone with age (Ding, 2000). Occult injuries such as subchondral bone microcracks are documented in a number of impact studies with and without gross fracture of bone (Ewers et al., 2000; Newberry and Haut, 1996; Haut and Atkinson, 1995). High rate impacts cause more microcracks than low rate impacts (Ewers et al., 2000). This microcracking is hypothesized to cause chronic subchondral bone thickening and remodeling post impact. Biomechanical function of the knee joint The knee joint functions primarily in flexion and extension with a normal range of motion within 130°. During passive flexion the femoral condyles rotate and the contact point moves posteriorly on the tibial plateaus (Pinskerova et al., 2004). Initially during flexion there is also a small amount of internal tibia rotation. Articulation between the medial and lateral compartments of the knee, as well as the cruciate ligaments, guides this motion. 14 Biomechanics research has used two sectioning techniques to determine the primary and secondary contributions of the knee ligaments to preventing relative joint motion. One method is to apply known displacements prior to and after removal of a specific ligament and measure the amount of force required to produce the movement. This is known as the stiffness method. The other method is to apply known forces to the knee and to measure the resulting displacements before and after ligament sectioning. This is known as the flexibility method (Figure 1.9). When a primary restraint to a specific joint motion is sectioned, the force will decrease in the stiffness method or the displacement will increase in the flexibility method. If sectioning a specific ligament does not have one of these effects until after the primary restraint is removed, then it is a secondary restraint to that particular motion. FORCE (N) I 7 ’1‘ / Normal / i— / I so N / " g , / ACL cut l I ’ ’ l l- I ’ / I l- ,1 l c ’// 1 mm 1 A 1”! 4 g r 1_ A. I I I A I I 1 Post .....— DIS. (MM) Tm: Figure 1.9 Flexibility method force—displacement for the normal and ACL-deficient knee joints (Markolf et al., 1984). Ligament sectioning has been used to evaluate the ACL’s role in preventing various types of joint displacements (Gabriel et al., 2004; Hallen and Lindahl, 1965; Hull, 15 I997; Markolf et al., 1984; Mills and Hull, 1991). The ACL is a primary restraint for anterior tibial subluxation (Butler, 1980; Fukubayeshi et al., 1982; Shoemaker and Markolf, 1995; Zantop et al., 2007). For an extended knee the ACL is also considered a primary restraint for internal tibia rotation (Figure 1.10) along with the MCL (Flemming et al., 2001; Hame et al., 2002; Markolf et al., 1995; Senter and Hame, 2006). On the other hand, the ACL is a secondary restraint for varus and valgus rotations (Figure 1.10) of the femur (Grood et al., 1981; Markolf et al., 1976; Nielson et al., 1984) as well as external tibia rotation (Figure 1.11) (Nielson and Helmig, 1985; Shoemaker and Daniel, 1990) and hyperextemsion (Bizot et al., 1995; Fomalski et al., 2008; Kennedy, 1963; Schenck et al., 1999). Other primary restraints in the knee are as follows; the PCL resists posterior tibia subluxation, the MCL resists valgus rotation of the femur, the LCL resists varus rotation of the femur and the posterior capsule resists knee hyperextension. Internal Tibial Rotation Figure 1.10 Knee ligaments and their function during tibial rotation (Childs, 2002). 16 ACL Resultant Force (N) Varus Moment (Nm) Valgus Moment (Nm) ' Figure 1.11 ACL force generated from varus/valgus bending (Markolf et al., 1990). ACL Resultant Force (N) lntemal Torque (Nm) External Torque (Nm) Figure 1.12 ACL force generated from tibial torsion (Markolf et al., 1990). Other methods for determining a ligament’s role in resisting specific joint motions include direct measurement of strain (Beynnon et al., 1992) or force (Markolf et al., 1990) by implanting strain gages (Figure 1.13), buckle transducers, or load cells. In subjects undergoing arthroscopic surgery, but with normal gait and an intact ACL, a strain transducer was implanted on the ACL and a variety of external loads were applied (Flemming et al., 2001). The results of this study show that an internal tibia torque of 10 17 Nm increases ACL strain, while a 10 Nm external torque does not. At 20° of knee flexion, weightbearing also significantly increases ACL strain. Combined loading of a 10 Nm internal torque and 40% of body weight, however, does not produce ACL strains higher than the isolated loading cases. This study also supports experimental investigations on cadaver knee specimens that document the ligament forces of the knee under isolated and combined loading conditions. For knee flexion angles less than 30° combined loading of a 10 Nm internal tibial torque and 100 N anterior tibial shear force produces ACL forces that exceed the externally applied shear force (Markolf et al., 1995). In another study, two knees were flexed 30° and the tibia internally rotated 20°, resulting in the ACL providing 30-40% of the ligamentous restraining force (Seering et al., 1980). In knees at 0° of flexion, an isolated 10 Nm internal torque produces an average ACL tensile force of 230 N (Hame et al., 2002). Posterior cruciate ligament Anterior cruciate ligament Figure 1.13 Hall effect strain transducer attached to the ACL (Beynnon et al., 1992). Normal in vivo weight bearing on the knee induces tensile strain in the ACL from anterior subluxation of the tibia (Fleming et al., 2001; Li et al., 1998; Markolf et al., 1981). The primary function (providing 85% of support) of the ACL is, in fact, to limit anterior tibial subluxation (Butler et al., 1980; Fukubayeshi et al., 1982; Torzilli et al., 18 1994). This motion occurs from TF compression for all joint flexion angles greater than 15° due to an inherent 10-15° posterior tilt of the tibial plateau (Figure 1.14) (DeJour et al., 1994; Genin et al., 1993; Giffin et al., 2004; Li et al., 1998; Torzilli et al., 1994). QUADRICEPS TENDON PATELLA TF COMPRESSIVE LOADING PATELLA LIGAMENT 15 DEGREES ANTERIOR SHIFT Figure 1.14 Tilt of the tibial plateau and anterior displacement of the tibia (Meyer, 2005). At 20° of knee flexion the average anterior/posterior (AP) laxity for a 200 N force applied to intact knees is 10 mm (Markolf et al., 1984). Using a commercially available clinical laxity measurement instrument, the KT-IOOO (Medmetric Corp, San Diego, CA) the maximum anterior laxity of in vivo knees at 30° of knee flexion was 8.5 mm (Daniel et al., 1985) or 5 mm for a 100 N anterior force (Sullivan et al., 1984). After ACL sectioning this anterior subluxation increases to 15 mm (Sullivan et al., 1984). There is a high amount of scatter in comparing right-left AP laxity differences in both normal and ACL-deficient populations, but generally a difference in laxity >3 mm is indicative of an ACL disruption (Figure 1.15) (Daniel et al., 1985). The AP laxity has been shown to be 19 the greatest at 30° of knee flexion and increases when axial rotation of the tibia is allowed (Fukubayashi et al., 1982). The sensitivity to diagnose a complete ACL disruption using this method is 96% (Daniel et al., 1985). \NonMAL \ I ACL DEFICIENT FREQUENCY O 3 6 9 12 15 ANTERIOR LAXITY (mm) Figure 1.15 Knee laxity in normal and ACL-deficient knees (Daniel et al., 1985). ACUTE KNEE INJURIES Ligamentous injury An estimated annual cost (direct and indirect) for scholastic sports injuries is $1.3 billion (Hergenroeder, 1998). The knee is one of the most frequently injured joints in the human body, accounting for 19-23% of all injuries (Arendt and Dick, 1995; Deitch et al., 2006; Hootman et al., 2002). Athletes in particular, suffer knee injuries at an even higher rate (nearly 40%) compared to other parts of the body (Majewski et al., 2006). The most common diagnosis for knee injuries in athletes is internal knee trauma at approximately 45%, followed by 34% for minor knee distortions, 11% for cartilage lesions, 5.5% for contusions, 3.3% for dislocations, and 1% for fractures (Majewski et al., 2006). Internal knee trauma is a classification that includes damage or tearing to the cruciate or collateral ligaments and menisci. The ACL, medial collateral ligament (MCL), and medial meniscus are the most frequently injured structures in the knee (Majewski et al., 2006). 20 Epidemiological studies have shown there are 80,000—250,000 ACL tears in the USA each year (Griffin et al., 2000), with a total cost of two billion. dollars (Hewett et al., 2006) In professional football, knee sprains account for approximately 10% of injuries and while ACL rupture occurs in only 11% of cases, it represents a severe injury with a time loss of six months of more (Powell and Schootrnan, 1992). MCL injuries are more frequent (55-73% of knee injuries), and 45% of players with this injury miss more than three games. Other sports with high rates of ACL injuries include; squash, handball, volleyball, basketball, soccer, and skiing (Majewski et al., 2006) The average age for ACL injured patients is 26 years old, with males representing a majority of the cases due to higher levels of participation in sports and recreational activities (Daniel and Fithian, 1994; Griffin et al., 2000). One of the most important aspects of these injuries is the impact the injury has on a player’s ability to continue participation. The games and practices that are missed due to injury affect the player’s skill development and personal satisfaction of the game (Hergenroeder, 1998). Injuries to high school athletes usually result in three to six days off, but approximately 45% of injuries require a week or more (Fernandez et al., 2007). There is a significant difference in the percentage of season-ending injuries suffered by girls (12.5%) versus boys (8%). This is most likely due to internal knee trauma injuries, as girls have to undergo surgery for this injury twice as frequently as boys. At a professional level, missed practices and games account for millions of dollars of lost revenue for the team (Powell and Schootrnan, 1992). 21 Following an ACL tear, most patients who undergo surgical reconstruction (65- 88%) can return to their sport within the first year and even nonoperatively treated patients return (19-82%) if they can regain knee stability in rehabilitation (Myklebust and Bahr, 2005). However, the rate of athletes competing at an elite level declines quicker than uninjured athletes, with only 30% of soccer players still competing three years after an ACL injury, versus 80% in the uninjured control population (Roos et al., 1995). There is also a much higher rate of reinjury (2-13%) or injury to a related structure such as the menisci, cartilage, or other ligaments (9-22%) if the athlete continues after an ACL tear (Myklebust and Bahr, 2005). The long-term cost of an acute injury may in fact be higher still, due to a significantly increased risk of developing chronic joint degradation, also known as post- traumatic osteoarthritis (OA). After ACL rupture approximately 50% of patients display signs of OA within ten years and nearly all patients have OA after 15-20 years (Myklebust and Bahr, 2005). Additionally, ACL reconstruction has been shown to have no effect on the high rate of post-traumatic OA (Fink et al., 2001; Myklebust et al., 2003; Von Porat et al., 2004). Sports participation, in general, has also been shown to increase the rate of OA, and certain sports such as weightlifting and team sports are associated with a higher risk of hip and knee OA, respectively. In contrast, moderate levels of exercise and activity appear safe and do not lead to OA (Wolf and Amendola, 2005). Analysis of knee injury mechanisms Most of the proposed ACL injury mechanisms are based on patient questionnaires and video analysis of the injury events. Approximately 91% of ACL injuries occur during sports activities (Paul et al., 2003) and of these, non-contact ACL injuries occur more 22 frequently than injuries involving player-to-player contact (Agel et al., 2005; Boden et al., 2000; Griffin et al., 2000; Hewett et al., 1999; Renstrom et al., 2008; Yu and Garrett, 2007). ACL injury results in immediate functional instability of the knee joint. A classic sign of an ACL tear is hearing a “pop” that occurs while cutting or pivoting and is followed by pain (Cameron et al., 2000; Fegin, 1979). Following the injury, ACL- deficient patients may experience giving-way episodes and are more likely to develop additional injuries in the knee joint, including meniscal tears (Allen et al., 2000; Bellabara et al., 1997; Cipolla et al., 1995) and osteoarthritis of the knee (Kannus and Jarvinen, 1987). Four mechanisms of noncontact isolated ACL rupture have been proposed in the clinical literature: The loading mechanisms are internal rotation of the tibia (Arnold et al., 1979; McNair et al., 1990), anterior shear of the tibia (Ettlinger et al., 1995), knee valgus bending (Boden et al., 2000; Krosshaug et al., 2007a; Olsen et al., 2004); and. hyperextension of the knee (Heinrichs, 2004; Meyers and Harvey, 1971; Snearly et al., 1996). Direct anterior shear of the tibia without contact from another player is quite uncommon, except in skiing when the entire bodyweight is on tail of the ski, which can induce the “phantom-foot ACL injury mechanism”. Hyperextension also may be more common in player-to-player contact scenarios (Figure 1.16), but has been documented to occur in noncontact situations with a large axial tibial load, such as jumping on a trampoline (Kwolek et al., 1998). Others propose that most ACL injuries occur in extended knees due to a combination of tibial rotation and valgus bending as well as an axial tibial load (Fauno and Wulff Jakobsen, 2004; Matsumoto et al., 2001; Shoemaker et al., 1988; Speer et al., 23 1995). Many studies have emphasized the frequency that ACL tears occur during landing from a jump on one or both legs (Figure 1.17) (Bahr and Krosshaug, 2005; Boden et al., 2000; F agenbaum and Darling, 2003, Hewett et al., 2006; Krosshaug et al., 2007a). Figure 1.17 Landing from a jump ACL injury mechanism (Boden et al., 2000). In a recent ACL injury mechanism survey the most common activities were basketball, football and soccer (skiing injuries were excluded) and 38 out of 71 noncontact ACL injuries occurred while decelerating during or just before a change in direction (Boden et al., 2000). An additional 26 ACL injuries occurred during landing from a jump (Figures 1.18). In a study of only soccer game-related ACL injuries, 56 out of 105 players were changing direction towards the side of their injured knee, while 10 24 120 A i Flexion oi . .3 , -- Varus v = — Internal - 2 I ; ".... ..., mm»- "d 3’ s s w < §:.‘;?‘ra' O E {0... g -i O I: l P .-‘ i o ...»- " w . WV § \lioquua-O". l l 40, l l I I i ' l I l A B C A 4 a 0 , AP g i , -- m- I. : —Vertical .2 2' s 2 . = i i .3 ..., ,....... .. a i ...-.1 s o a. . 2 ...-~.. g ’ inn“ karma? ‘3 I I I? w/ v D I I I f :3: i so e .2 f 9. .3." i ‘9 f : -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 Time (s) Figure 1.18 Video frame sequence of the ACL injury to a NBA player. (A) initial ground contact, (B) after 33 ms and (C) after 100 ms and the corresponding time sequence of knee joint angle and ground reaction force (Krosshaug et al., 2007b). 25 were turning towards the uninjured side (Fauno and Wulff Jackobsen, 2004). An additional 26 players sustained their injuries when landing after jumping to head the ball, and the authors concluded that “tackling and kicking do not contribute significantly to ACL ruptures in soccer. Skiing, also has one of the highest rates of ACL injuries, accounting for 25-30% of knee injuries (Pujol et al., 2007; Speer, 1995). These injuries are mainly associated with internal twisting or combined loading during a hard landing (Ettlinger et al., 1995; Shoemaker et al., 1988). Li et al. suggest “that excessive compressive loads caused by impact load along the tibial shaft (e.g., landing from a jump) may contribute to injury of the ACL, especially when the knee is flexed.” While injury of the ACL has been documented in landings with the knee relatively straight (Speer et al., 1992), the knee may also be flexed as much as 60-80° during a landing (Hewett et al., 1996). Hypothesis 1: “The external tibial and valgus femoral rotations frequently identified after ACL injury are not representative of the relative displacements that cause isolated ACL failure to occur.” Experimentally produced knee injuries Kennedy et a1. (1974) tested five cadaver knees in internal rotation (at 20° of flexion) and described the ACL becoming very taut, but premature fractures occurred at the bone clamps before ACL injuries. Additional studies have been conducted in external tibia rotation to characterize MCL injury mechanisms (Kennedy and Fowler, 1971; Shoemaker and Markolf, 1982). Some of those specimens also had ACL injuries occur as the joints were externally rotated beyond the MCL failure point and impingement occurred against the medial edge of the lateral femoral condyle (Csintalan et al., 2006; 26 Kennedy and Fowler, 1971). Kennedy et al. (1963) also tested knees in hyperextension by applying an anterior directed force to the proximal femur with the tibia rigidly constrained (Figure 1.19). They described tearing in the posterior capsule at approximately 30° of hyperextension, followed by complete anterior dislocation of the knee joint (Kennedy, 1963). Other hyperextension studies did not produce ACL injury (Bizot et al., 1995), or used three-point bending with fixed pivot points which caused distraction between the tibia and femur (Fomalski et al., 2008; Schenck et al., 1999). The knee joint has also been tested with a lateral-medial (valgus) bending moment (Figure 1.20) and those tests resulted in MCL rupture with one knee also suffering an ACL injury (Kerrigan et al., 2003). Isolated muscle force has also been shown to produce ACL injury for quadriceps forces of 4 kN (Figure 1.21) (DeMorat et al., 2004). 27 Preload (400 N) L ad cell 41% lmpactor -——-*- ..___ Mobile plate Figure 1.20 Valgus bending method for producing knee injuries (Kajzer et al., 1999). Figure 1.21 ACL rupture experiments from excessive force in the quadriceps tendon (DeMorat et al., 2004). In cadaver experiments, excessive axial compression loads in the tibia will produce injury to the ACL when the knee joint is free to translate in the anterior-posterior direction (Jayaraman et al., 2001). In 4/5 joints at 60° and 4/5 joints at 90° of flexion the ACL was ruptured at axial tibial loads of 4.4:tl.1 kN and 4.6:t1.2 kN, respectively. Constrained experiments on the other hand, which do not allow anterior-posterior (AP) or 28 medial-lateral (ML) motion of the tibia with respect to the femur result in fracture to the medial and lateral tibial plateau, medial femoral condyle and femoral notch at 8.0 $1.8 kN (Banglmaier et a1. 1999). Additionally in constrained joints the load to prevent motion of the femur relative to the tibia is 1.2 :l:0.5 kN (Jayaraman et al., 2001). This may be the tensile load in the ACL prior to rupture in unconstrained experiments. Recently, a porcine study confirmed that ACL injuries occur for unconstrained knees at 70° of flexion with TF compressive loads of approximately 3 kN (Yeow et al., 2007). Hypothesis 2: “Tibiofemoral compression will produce anterior tibial subluxation and isolated ACL injuries, while other loading mechanisms will produce combination ligament injuries.” Knee kinetics In rimning, the ground reaction force is approximately 2.5 times body weight (Cavanagh and Lafortune, 1980). During controlled jump landings ground reaction forces can be 4 or 6 times body weight for untrained females and males, respectively (Hewett et al., 1996). The internal loads in the knee joint are much higher than that, due to the combined force of the contracting muscles in order to control knee flexion. During running the TF contact force is estimated to be 15 times body weight (Glitsch and Baumann, 1997). Thus for the untrained, 50th percentile male in an uncontrolled jump landing on a relatively straight leg, joint compressive forces could well exceed 5 kN. This level of GRF may injure or even grossly rupture the ACL, based on the above data from the cadaver experiments. That is, where the ACL injury-mitigating influences of neuromuscular joint control via the gastrocnemius and hamstrings muscles are not considered (Hewett et al., 1996). On the other hand, muscle forces also substantially 29 increase the compressive force in the TF joint, due to the balance of forces effect and may contribute to ACL injuries by this mechanism. Recent studies on the incidence of ACL rupture in female athletes have concentrated on potential differences in neuromuscular controls between the sexes, and the training of hamstrings and quadriceps muscles in helping to aid in the limitation of anterior directed motion of the tibia, which appears to occur during a jump loading (Fagenbaum & Darling, 2003; Hewett et al., 1996). Quadriceps muscle force produces strain in the ACL (Withrow et al., 2006). Stop-jump landings include a posterior directed ground reaction force (Figures 1.22 and 1.23), but the necessary extension moment in the quadriceps needed to balance the knee produces an anterior shear force in the proximal tibia instead (Sell et al.,2007). Lower extremity muscle fatigue may increase an athlete’s risk for noncontact ACL injury. Increased injury rates occur during the later portion of games in a variety of sports which indicates that fatigue is a risk factor for injuries (Gabbett and Dumrow, 2005). The anterior shear force on the proximal tibia is significantly increased for both male and female subjects in the fatigued state during a stop-jump landing (Chappell et al., 2005). Previous studies have suggested that proximal tibial anterior shear force may be an indication of increased strain in the ACL (Hutchinson and Ireland, 1995). 30 Proximal tibial axial force Knee extension moment Proximal tibia shear force Posterior ground reaction force . . Vertical ground reaction force Figure 1.22 Stop-jump landing and the resultant proximal tibial anterior shear force (Chappell et al., 2005). TD. 2 Q 0 IL Ii 2 (I) -600 . 800 i —Total -o-Patellar Tendon , -1000 .1». —TF Contact -°-Hamstrings l U —GRF +Gastrocnemius l -1200 »— —~ — —i---~-—~ 4 —~--. - . “J 0 50 100 150 200 Time (ms) Figure 1.23 Shear forces (+ anterior tibia shear) acting on the lower leg during the simulated drop-landing. (Pflum et al., 2004). 31 ACL injury treatment The treatment goal of the ACL-injured patient is to prevent recurrent knee injury while allowing the patient to return to their desired work and level of sports participation. If left untreated, a torn ACL can cause an increased incidence of meniscal tears, in addition to the anterior and rotary instabilities (Noyes et al., 1983). In “high-risk lifestyle” patients, those who participate in frequent jumping or cutting sports, ACL reconstruction is usually recommended. Two types of autogenous graft tissues are commonly used: bone-patellar tendon-bone (BPTB) graft and doubled semitendinosus and gracilis tendons (DLSTG) graft (Howell, 2001). These graft choices are made based on case of harvesting, donor site morbidity, initial strength and stiffness and fixation technique. Both grafts have initial strength and stiffness values that exceed the normal ACL’s mechanical properties (Table 1.2). However, numerous animal studies have shown that between 1 month and 1 year post reconstruction the grafts lose 50-80% of their material properties through remodeling. An additional consideration for the ACL reconstruction is its ability to restore normal knee kinematics, especially to control anterior translation of the tibia. Recently, kinematic studies have also identified tibial rotation as a key motion to restore following ACL reconstruction. Anterior tibial translation has been shown to be similar between ACL reconstruction and normal knees; however, ACL reconstruction fails to restore rotational or valgus kinematics (Tashman et al., 2004; Woo et al., 2002). One of the most important variables for the issue of kinematic restoration is the positioning and orientation of the ACL reconstruction graft. 32 Table 1.2 Structural properties of the ACL and grafts in tension. _ Area Stiffness Strength Tissue 2 (mm ) (Nlmm) (N) ACL (Noyes et al., 1984) 44 (4.0) 182 (33) 1725 (269) BPTB (Noyes et al., 1984) 51 (2.8) 685 (86) 2900 (260) DLSTG (Hamner et al., 1999) 53 (5.3) 776 (204) 4090 (295) POST-TRAUMATIC OSTEOARTHRITIS Risk factors and epidemiology Nearly 50% of Americans over the age of 65 have some form of arthritis (CDC Fact Sheets, 1997) with total costs of the disease approximately $80 billion per year. Osteoarthritis, or degenerative joint disease, is the most common musculoskeletal disease and the most common form of arthritis affecting 20.7 million people in the USA alone (1999). 0A affects the cartilage and subchondral bone of diarthrodial joints (Lane, 1996). It is characterized by irregular loss of cartilage in areas of high load, sclerosis of subchondral bone (SB), subchondral cysts and osteophytes. ' Clinical diagnosis of osteoarthritis comes only when a significant reduction of the joint space is seen radiographically (Figure 1.24) (Hamerman, 1989), although MRI is proving to be a useful tool in diagnosing OA at an earlier point (Hodgson et al., 1992). From the patient’s perspective this disease is characterized by diarthrodial joint pain and tenderness, loss of range of motion and localized inflammation around the affected joint. Since the main function of diarthrodial joints is to allow body movement and locomotion, this disease has grave consequences for a patient’s quality of life. 33 ‘6‘ Loss of .loint Space Osteophytes Figure 1.24 Radiographic diagnosis of CA by loss of joint space. Loss of joint space is also associated with osteopyhtes of the bone (Felson et al., 1997). Association between end stage OA from one particular cause have been difficult due to the long incubation time before chronic changes occur and radiographic evidence appears, typically at least 10-20 years (Wright, 1990). In many patients the disease is due to a lifetime of high stresses in a particular joint from an occupation or recreational activity (Dieppe and Dieppe, 1992). A significantly higher percentage of patients with ligament tears or sprains go on to develop OA as a result of the change in the way forces are transferred through the joint after an injury. There is also the possibility of a single mechanical insult initiating the disease process, especially if there is bone fracture or soft tissue damage near the articulating surfaces (Chapchal et al., 1978; Davis et al., 1989; Nag Sports participation in general has also been shown to increase the rate of OA (Cosgarea and Schatzke, 1997; Lane, 1996), and certain sports such as weightlifting and team sports (Kujala, 1995) are associated with higher risk of hip and knee OA, respectively (Chapacha, 1978; Davies-Tuck et al., 2007; Davis et al., 1989; Gelber et al., 2000;Tepper and Hochberg, 1993). In contrast, moderate levels of exercise and activity appear safe and do not lead to OA (Wolf and Amendola, 2005). But, while moderate 34 contact forces in the knee are okay and even beneficial for healthy joints (US Dept Health, 1998), excessive contact forces during an acute injury or applied repetitively during high impact sports induce long term joint degradation by damaging the articular cartilage and/or subchondral bone. History of a joint injury, particularly to the knee or hip, increases the risk of CA in cross-sectional and case-control studies (Davis et al., 1989; Tepper and Hochberg, 1993). Furthermore, people who injure a knee before the age of 22 years have a greater than 3-fold increased risk of diagnosed 0A in that joint by the mid 50$ (Gelber et al., 2000). Two specific types of injury are strongly associated with subsequent knee OA: cruciate ligament damage and meniscal tears (Felson, 2004). In fact, 50-70% of patients with complete ACL rupture and associated injuries have radiographic changes consistent with chronic disease after 15 to 20 years (Gillquist and Messner, 1999). And, surgical reconstruction of the torn ACL is not effective in mitigating the incidence of joint 0A (Myklebust and Bahr, 2005; Von Porat et al., 2004), as a significant proportion of these patients develop clinical symptoms of OA 5-10 years post-injury (Asano et al., 2004; Daniel et al., 1994; Frederich & O’Brien, 1993). Bone bruises Subtle damage to cartilage and subchondral bone can occur without radiographic evidence of bone fracture (Pritsch et a1. 1984). Recent studies have focused on identifying occult bone trauma and relating it to clinical findings. These radiographically occult injuries to the bone, otherwise referred to as occult fractures or “bone bruises”, may account for patient pain (Kapelov et al., 1993). In over 80% of clinical ACL injury cases, a characteristic osteochondral lesion occurs in the tibial plateau and/or the femoral condyles (Mink and Deutsch, 1984; Speer et al., 1992). These occult injuries have been 35 shown to be accompanied by chondrocyte necrosis and surface lesions (Fang et al., 2001; Johnson et al., 1998) and result in an overt loss of cartilage overlying geographic bone bruises in 48% of patients within six months of injury (Faber et al., 1999; Vellet et al., 1991).Bone bruises are associated with microcracks of the subchondral and/or trabecular bone (Rangger et al., 1998; Speer et al., 1990) and may be caused by compressive trauma during the associated ligamentous injury. Radiographic images have also been used to relate bone marrow edema patterns with injury mechanisms through their characteristic “footprint” (Kaplan et al., 1992; Kaplan et al., 1999; Sanders et al., 2000; Viskontas et al., 2008) (Figures 1.25 and 1.26). Figure 1.25 Valgus bending injury mechanism causing a pattern of bone bruise in the lateral tibial plateau and femoral condyle (Sanders et al., 2000). 36 Figure 1.26 Hyperextension injury mechanism causing a pattern of bone bruise in the anterior tibial plateau and femoral condyle (Sanders et al., 2000; Hayes et al., 2000). Osteochondral microdamage Lower extremity trauma causes acute pain followed by a chronic disease process that can lead to an end-stage disease such as OA (States, 1986). Biomechanically, the cartilage material properties, such as the tensile, compressive and shear moduli change. The hydraulic permeability of the cartilage also changes due to degradation of the collagen, causing increased water content and excessive swelling. Additionally, the SB thickness and stiffness change as it undergoes remodeling due to changing stress levels. The progression of this disease involves chronic fragmentation of the cartilage surfaces and remodeling of SB. Histological methods are a common way of documenting subfracture injuries such as cartilage fissuring and occult microcracking at the calcified cartilage/subchondral bone interface (Figure 1.27). These studies use semi-quantitative scoring to analyze the condition of tissue as a result of impact and chronic degradation. Impact loading has been shown to initiate damage in articular cartilage between 11-36 MPa of contact pressure, depending on the thickness, rate of loading, species and location 37 (Atkinson et al., 1998; Haut, 1989; Repo and Finlay, 1977; Torzilli et al., 1999). Haut (1989) documents that at the point of fracture up to 60% of the human patellofemoral contact area exceeds 25 MPa. This level of pressure was also shown to cause cartilage fissures in an in vitro cartilage explant model (Repo and Finlay 1977). Figure 1.27 Surface fissure of the articular cartilage (A) and an occult microcrack at the subchondral bone interface (B) (Meyer et al., 2004). Even when the ACL is not injured, presumably due to appropriate muscular contraction preventing its rupture in the trained athlete, high TF contact forces would occur and depending on the ACL injury mechanisms may be sufficient to generate osteochondral microdamage. Histological microcracks of the subchondral plate occur in cadaver knees under high compressive loads, producing 18-21 MPa of contact pressure in the TF joint (Banglemaier et al., 2001) and in a rabbit model at similar pressures (Isaac et al., 2008). Articular cartilage fibrillation and cell death have also been documented in explant studies when pressures exceed a critical threshold stress (25 MPa) (Buckland- Wright et al., 2000; D’Lirna et al., 2001; Repo and Finlay, 1977; Spindler et al., 1993; Torzilli et al., 1999). Contact pressure distributions in the knee joint have been measured during physiological levels of TF compression (Ahmed and Burke, 1983; Brown and Shaw, 38 1984; Fukubayashi and Kurosawa, I980; Riegger-Krugh et al., 1998; Thambyah et al., 2005), varus-valgus bending (McKellop et al., 1991), and intemal-external tibial rotation (Yildirim et al., 2007). The TF contact pressures have also been computed by finite element analysis for a number of loading mechanisms, but not for failure level forces/moments (Bendjaballah et al., 1995; Bendjaballah et al., 1997; Jilani et al., 1997). Additional knee models have predicted the stress distributions that occur in the articular cartilage during TF contact, but they used rigid models for bone, and thus are incapable of predicting damaged areas in the subchondral bone (Haut-Donahue et al., 2002; Li et al., 2001). Experimental studies have documented osteochondral damage in the tibial plateau from TF compression in porcine knees (Yeow et al., 2008) and rabbit knees (Isaac et al., 2008). So far no study has created a finite element model to relate TF contact pressures with stress distributions in the cartilage and subchondral bone in order to validate the characteristic footprints of bone bruises and overlying cartilage degeneration that occur with specific injury mechanisms. Hypothesis 3: “The mechanism-based clinical classification of knee injuries and bone bruise patterns will correspond to characteristic distributions of high levels of contact pressure and osteochondral microdamage across the tibial plateau for each loading mechanism.” SUMMARY AND OBJECTIVES The current study presents three hypotheses that can be classified under the broader topics of sports medicine injury mechanisms, the biomechanics of injury, and post-traumatic osteoarthritis. Recently, numerous surveys have been conducted based on patient questionnaires and videos of ACL injuries to determine the loading mechanism 39 that cause this frequent and serious sports injury. The two most common scenarios were axial rotation of the tibia and valgus bending of the knee. But, as one of the studies pointed out; “whether the consistent valgus collapse observed in the videos was actually the cause of injury or simply a result of the ACL being torn is open for discussion.” Therefore the first objective of this dissertation was to measure the pre-failure and failure characteristics of each of these loading mechanisms in order to determine the cause-and- effect relationship between knee joint motion and potential ACL failure. A second objective was based on our previous experiments showing that isolated ACL failure occurs due to anterior subluxation of the tibia caused by TF compressive forces in the flexed knee. An objective was to provide additional data recognizing the importance of this loading mechanism in the extended knee during non-contact, isolated ACL injuries. Finally, there is a large body of evidence for the association of ACL injury with long- terrn development of post-traumatic osteoarthritis. A final hypothesis of the current study was that chronic joint degeneration may be due, in part, to osteochondral microdamage that occurs at the time of acute ligamentous injury. Another objective was to record the contact pressure distribution in the TF joint that occurs during each injury mechanism. Based on these contact pressures a 3-D computational model was to be developed that would be able to predict the experimentally observed osteochondral damage patterns for each proposed injury mechanism of loading. The microdamages predicted and recorded in the current study are applicable to the clinical literature for comparison with the current mechanism-based classification of knee injuries and bone bruise patterns. Strategies for dealing with post-traumatic osteoarthritis may use information based on the injury mechanism and locations of 40 osteochondral microdamage when developing a treatment regimen. Additionally, this study will provide emphasis for the importance of TF compression in producing ACL injury. Injury prevention strategies may then be focused on the types of motion shown to be most important for producing non-contact ACL injury. 41 References Agel J, Arendt E, Bershadsky B. Anterior cruciate ligament injury in National Collegiate Athletic Association basketball and soccer. Am J Sports Med. 2005;33(4):524-531. Ahmed A, Burke D. In-vitro measurement of static pressure distribution in synovial joints- Part 1: Tibial surface of the knee. J Biomech Eng. 1983;105:216-225. Akizuki S, Mow V, Muller R, et al. Tensile properties of human knee joint cartilage: 1. Influence of ionic conditions, weight bearing, and fibrillation on the tensile modulus. J Orthop Res. 4:379-392, 1986. Albright J, Powell J, Smith W. Medial collateral ligament knee sprains in college football: Brace wear preferences and injury risk. Am J Sports Med. 1994;22(1):2-11. Allen C, Wong E, Livesay G, Sakane M, Fu F, Woo S. Importance of the medial meniscus in the anterior cruciate ligament-deficient knee. J Orthop Res. 2000;18:109- 115. Andreasson G, Lindenberger U, Renstrom P, et a1. Torque developed at simulated sliding between sport shoes and an artificial turf. Am J Sports Med. 1986;14(3):225-230. Arendt E, Dick R. Knee injury patterns among men and women in collegiate basketball and soccer: NCAA data and review of literature. Am J Sports Med. 1995;23(6):694- 701. Amoczky S. Anatomy of the anterior cruciate ligament. Clin Orthop Rel Res. 1983;172:19-25. Arnold J, Coker T, Heaton L, Park J, Harris W. Natural History of Anterior Cruciate Tears. Am J Sports Med. 1979;72305-313. Asano H, Muneta T, Ikeda H, Yagishita K, Kurihara Y, Sekiya 1. Arthroscopic evaluation of the articular cartilage after anterior cruciate ligament reconstruction: a short-term prospective study of 105 patients. Arthroscopy. 2004;20:474-481. Atkinson P, Haut R. Subfracture insult to the human cadaver patellofemoral joint produces occult injury. J Orthop Res. 1995;13:936-944. Atkinson T, Haut R, Alterio N. Impact-induced fissuring of articular cartilage: An investigation of failure criteria. J Biomech Eng. 1998;120:181-187. Atkinson PJ, Haut RC. Injuries produced by blunt trauma to the human patellofemoral joint vary with flexion angle of the knee. J Orthop Res. 2001;19:827-833. Atkinson P, Cooper T, Anseth S, Walter N, Kargus R, Haut R. Association of knee bone bruise frequency with time postinjury and type of soft tissue injury. Orthop. 2008;31:440. 42 Bahr R, Krosshaug T. Understanding injury mechanisms: a key component of preventing injuries in sport. Br J Sports Med 39;2005:324—329. Banglmaier R, Dvoracek-Driksna D, Oniang’o T, Haut R. Axial compressive load response of the 90° flexed human tibiofemoral joint. Stapp Conf Proc. 2001;42:127- 140. Bellabarba C, Bush-Joseph C, Bach B. Patterns of meniscal injury in the anterior cruciate-deficient knee: a review of the literature. Am J Orthop. 1997;26:18-23. Bendjaballah M, Shirazi-Adl A, Zukor D. Biomechanics of the human knee joint in compression: reconstruction, mesh generation and finite element analysis. Knee. 1995;2r69-79. Bendjaballah M, Shirazi-Adl A, Zukor D. Finite element analysis of human knee joint in varus-valgus. Clin Biomech. 1997;12:139-148. Beynnon B, Howe J, Pope M, Johnson R, Fleming B. The measurement of anterior cruciate strain in vivo. Int Orthop. 1992;16:1-12. Bizot P, Meunier A, Christel P, Witvoet J. Experimental passive hyperextension injuries of the knee: Biomechanical aspects of their consequences. Revue de Chirurgie Orthopedique. 1995;81 :21 1-220. Boden B, Dean G, Feagin J, Garrett W. Mechanisms of Anterior Cruciate Ligament Injury. Orthop. 2000;23:573-578. Brautigan B, Johnson D. The epidemiology of knee dislocations. Clin Sports Med. 2000;19:387-397. Brown T, Shaw D. In vitro contact stress distribution on the femoral condyles. J Orthop Res. 1984;2:190—199. Buckland-Wright J, Lynch J, Dave B. Early radiographic features in patients with anterior cruciate ligament rupture. Ann Rheum Dis. 2000;59:641-646. Butler D, Noyes F, Grood E. Ligamentous restraints to anterior-posterior drawer in the human knee. A biomechanical study. J Bone Jt Surg. 1980;62Az259-270. Butler D. Anterior Cruciate Ligament; Its Normal Response and Replacement. J Orthop Res. 1989;7z910-921. ‘ Cameron M, Mizuno Y, Cosgarea AJ. Diagnosing and managing anterior cruciate ligament injuries. J Musculoskeletal Med. 2000;17:47-53. Cavanagh P, Lafortune M. Ground reaction forces in distance running. J Biomech. 1980;13:397-406. 43 Centers for Disease Control. Guidelines for schools and communities for promoting lifelong physical activity. 1997 Chapchal G. Posttraumatic Osteoartritis After Injury of the Knee and Hip Joint. Reconstr Surg Trauma. 1978;16: 87-94. Chappell J, Herman D, Knight B, Kirkendall D, Garrett W, Yu Bing. Effect of fatigue on knee kinetics and kinematics in stop-jump tasks. Am J Sports Med. 2005;33:1022- 1029. Chaudhari A, Briant PL, Bevill S, Koo S, Andriacchi T. Knee kinematics, cartilage morphology and osteoarthritis after ACL injury. Med Sci Sports Exer. 2008;40:215- 222. Chen E, Black J. Materials design analysis of the prosthetic anterior cruciate ligament. J Biomat Res. 1980;14:567-586. Childs S. Pathogenesis of anterior cruciate ligament injury. Orthop Nursing. 2002;21:35- 40. Cipolla M, Scala A, Gianni E, Puddu G. Different patterns of meniscal tears in acute anterior cruciate ligament ruptures and in chronic ACL-deficient knees: classification, staging and timing of treatment. Knee Surg Sports Traumatol Arthrosc. 1995;3z130- 134. Conn ll, Annest J, Gilchrist J. Sports and recreation related injury episodes in the US population, 1997-99. Inj Prev. 2003;9zl 17-123. Cosgarea A, Schatzke M. Knee problems in the young athlete: A clinical overview. J Musculoskeletal Med. 1 997; 14(8):96-109. Csintalan R, Ehsan A, McGarry M, Fithian D, Lee T. Biomechanical and Anatomical Effects of an External Rotation Torque Applied to the Knee; A Cadaveric Study. Am J Sports Med. 2006;34:1623-1629. Daniel D, Stone M, Sachs R, Malcom L. lnstrumented measurement of anterior knee laxity in patients with acute anterior cruciate ligament disruption. Am J Sports Med. 1985;13:401-407. Daniel D, Fithian DC. Indications for ACL surgery. Arthroscopy. 1994;10:434-441. Daniel D, Stone M, Dobson B, Fithian D, Rossman D, Kaufman K. Fate of the ACL injured patient: A prospective outcome study. Am J Sports Med. 1994;22:632-644. Davis M, Ettinger W, Neuhaus J, Cho S, Hauck W. The association of knee injury and obesity with unilateral and bilateral osteoarthritis of the knee. Am J Epidem. 1989;130(2):278-288. 44 Davies-Tuck M, Wluka A, Wang Y, Teichtahl A, Jones G, Ding C, Cicuttini F. The natural history of cartilage defects in people with knee osteoarthritis. Osteoarthritis Cart. 2008;16:337-342. Deitch J, Starker C, Walters S, Moseley J. Injury risk in professional basketball players: A comparison of Women’s National Basketball Association and National Basketball Association Athletes. Am J Sports Med. 2006;34(&):1077-1083. DeJour H, Bonnin M. Tibial translation after anterior cruciate ligament rupture. Two radiological tests compared. J Bone Jt Surg. 1994;768:745-749. DeMorat G, Weinhold P, Blackburn T, Chudik S, Garrett W. Aggressive quadriceps loading can induce noncontact anterior cruciate ligament injury. Am J Sports Med. 2004;32:477-483. Derscheid G, Garrick J: Medial collateral ligament injuries in football. Nonoperative management of grade I and grade II sprains. Am J Sports Med. 1981;9(6):365-368. Dieppe P, Dieppe P. Osteoarthritis. Rheum. 1992;31:132-133. Ding M. Age variations in the properties of human tibial trabecular bone and cartilage. Acta Orthop Scand Suppl. 2000;292:1-45. D’Lima D, Hashimoto S, Chen P, Colwell C, Lotz M. Human chondrocyte apoposis in response to mechanical injury. Osteoarthritis Cart. 2001 ;9:712-719. Ettlinger C, Johnson R, Shealy J. A method to help reduce the risk of serious knee sprains incurred in alpine skiing. Am J Sports Med. 1995;23:531-537. Ewers B, Jayaraman ‘V, Banglmaier R, Haut R. The effect of loading rate on the degree of acute injury and chronic conditions in the knee after blunt impact. Stapp Conf Proc. 2000;44:299-314. Faber K, Dill J, Amendola A, Thain L, Spouge A, Fowler P. Occult osteochondral lesions after anterior cruciate ligament rupture six-year magnetic resonance follow-up study. Am J Sports Med. 1999;27(4):489-494. Fagenbaum R, Darling W. Jump landing strategies in male and female college athletes and the implications of such strategies for anterior cruciate ligament injury. Am J Sports Med. 2003;31:233-240. Fang C, Johnson D, Leslie M, Carlson C, Robbins M, Cesare P. Tissue distribution and measurement of cartilage oligomeric matrix protein in patients with magnetic resonance imaging-detected bone bruises after acute anterior cruciate ligament tears. J Orthop Res. 2001;19:634-641. Fauno P, Wulff Jakobsen B. Mechanism of Anterior Cruciate Ligament Injuries in Soccer. Int J Sports Med. 2004;27:75-79. 45 Feagin J. The syndrome of the torn, anterior cruciate ligament. Orthop Clin North Am. 1979;10:81-90. Felson D, McAlindon T, Anderson J, Naimark A. Defining radiographic osteoarthritis for the whole knee. Osteoarth Cart. 1997;52241-250. Felson D. An update on the pathogenesis and epidemiology of osteoarthritis. Radiologic Clinics N Am. 2004;42:1-9. Fernandez W, Yard E, Comstock R. Epidemiology of lower extremity injuries amoung US. high school athletes. Acad Emergency Med 14;2007:641-645. F erretti A, Papandrea P, Conteduca F, Mariani PP. Knee ligament injuries in volleyball players. Am J Sports Med. 1992;20:203-207. Fink C, Hoser C, Hackl W, Navarro R, Benedetto K. Long-term outcome of operative or nonoperative treatment of anterior cruciate ligament rupture — Is sports activity a determining variable? Int J Sports Med. 2001;22:304-309. Fleming B, Renstom P, Beynnon B, Engstrom B, Peura GD, Badger GJ, Johnson RJ. The effect of weight-bearing and external loading on anterior cruciate ligament strain. J Biomech. 2001;34:163-170. F omalski S, McGarry M, Csintalan R, Fithian D, Lee T. Biomechanical and anatomical assessment after knee hyperextension injury. Am J Sports Med. 2008;36:80-84. Friederich N, O'Brien W. Gonarthrosis after injury of the anterior cruciate ligament. Unfallchir Versicherungsmed. 1993;86:81-89. Fukubayashi T, Kurosawa H. The contact area and pressure distribution pattern of the knee: A study of normal and osteoarthrotic knee joints. Acta Orthop Scand. 1980;51:871-879. Fukubayashi T, Torzilli PA, Sherman MF, Warren RF. An in vitro biomechanical evaluation of AP motion of the knee. J Bone Jt Surg. 1982;64A:258-264. Gabbett T, Dumrow N. Risk factors for injury in subelite rugby league players. Am J Sports Med. 2005;33:428-434. Gabriel M, Wong E, Woo S, Yagi M, Debski R. Distribution of in situ forces in the anterior cruciate ligament in response to rotatory loads. J Orthop Res. 2004;22:85-89. Gelber A, Hochberg M, Mead L, Wang N, Wigley F, Klag M. Joint injury in young adults and risk for subsequent knee and hip osteoarthritis. Annals Internal Med. 2000;133:321-328. Genin P, Weill G, Julliard R. The tibial slope. Proposal for a measurement method. J Radiol. 1993;74:27-33. 46 Giffen J, Vogrin T, Zantop T, Woo S, Harner C. Effects of increasing tibial slope on the biomechanics of the knee. Am J Sports Med. 2004;32:376-382. Gillquist J, Messner K. Anterior cruciate ligament reconstruction and the long-term incidence of gonarthritis. Sports Med. 1999;27:143-156. Girgis F, Marshall J, Al Monajen A. The cruciate ligaments of the knee joint. Anatomical, functional and experimental analysis. Clin Orthop Rel Res. 1975;106:216-231. Glitsch U, Baumann W. The three-dimensional determination of internal loads in the lower extremity. J Biomech. 1997;30:1123-1131. Griffin L, Agel J, Albohm M: Non-contact anterior cruciate ligament injuries: risk factors and prevention strategies. J Am Acad Orthop Surg. 2000;8: 141-150. Grood E, Noyes F, Butler D, Suntay W. Ligamentous and capsular restraints preventing strainght medial and lateral laxity in intact human cadaver knees. J Bone Jt Surg. 1981;63A21257-1269. Hallen L, Lindahl O. Rotation in the knee-joint in experimental injury to the ligaments. Acta Orthop Scand. 1965;36:400-407. Hame S, Oakes D, Markolf K. Injury to the Anterior cruciate ligament during alpine skiing; A biomechanical analysis of tibial torque and knee flexion angle. Am J Sports Med. 2002;30:537-540. Hamerman D. The biology of osteoarthritis. New Eng J Med. 1989;320:1322-1330 Hamner D, Brown C, Steiner M, Hecker A, Hayes W. Hamstring tendon grafts for reconstruction of the anterior cruciate ligament: Biomechanical evaluation of the use of multiple strands and tensioning techniques. J Bone Jt Surg. 1999;81Az549-557. Hammer D. Artificial playing surfaces. Athletic Training. 1981;16:127-129. Haut Donahue T, Hull M, Rashid M, Jacobs C. A finite element model of the human knee joint for the study of tibio-femoral contact. J Biomech Eng. 2002;124:273-280. Haut R. Contact Pressures in the Patello-Femoral Joint During Impact Loading on the Human Flexed Knee. J Orthop Res. 1989;7z272-280. Haut R, Atkinson P. Insult to the Human Cadaver Patello-Femoral Joint: Effects of Age on Fracture Tolerance and Occult Injury. Stapp Car Crash J. 1995;39:281-294. Hayes C, Brigido M, Jamadar D, Propeck T. Mechanism-based pattern approach to classification of complex injuries of the knee depicted at MR imaging. RadioGraphics. 2000;20:S 121-S 134. 47 Heinrichs A. A review of knee dislocations. J Ath Train. 2004;39:365-369. Hergenroeder A. Prevention of sports injuries. Pediatrics. 1998;101(6):1057-1063. Hewett T, Stroupe A, Nance T, Noyes F. Pylometric training in female athletes: Decreased impact forces and increased hamstrings torques. Am J Sports Med. 1996;24:765-773. Hewett T, Lindenfeld T, Riccobene J, Noyes F. The effect of neuromuscular training on the incidence of knee injury in female athletes: A prospective study. Am J Sports Med. 1999;27:699-706. Hewett T, Myer G, Ford K. Anterior cruciate ligament injuries in female athletes: Part 1, Mechanisms and risk factors. Am J Sports Med. 2006;34:299-311. Hodgson R, Carpenter T, Hall L. Magnetic Resonance Imaging of Osteoarthritis. Articular Cart and Osteoarthritis Raven Press Ltd, NY. 1992:629-667. Hootman J, Macera C, Ainsworth B, Addy C, Martin M, Blair S. Epidemiology of musculoskeletal injuries among sedentary and physically active adults. Med & Sci Sports Exercise. 2002;34:838-844. Howel S. Autogenous graft choices in ACL reconstruction. Cur Opin Orthop. 2001;12:149-155. Hull M. Analysis of skiing accidents involving combined injuries to the medial collateral and anterior cruciate ligaments. Am J Sports Med. 1997;25:35-40. Hutchinson M, Ireland M. Knee injuries in female athletes. Sports Med. 1995;19:288- 302. Hvid I. Trabecular bone strength at the knee. Clin Orthop. 1988;227:210-221. Isaac D, Meyer E, Haut R. Chondrocyte damage and contact pressures following impact on the rabbit tibiofemoral joint. J Biomech Eng. 2008;130:0410180. Jayaraman V, Sevensma E, Kitagawa M, Haut R. Effects of Anterior-Posterior Constraint on Injury Patterns in the Human Knee During Tibial-Femoral Joint Loading from Axial Forces Through the Tibia. Stapp Car Crash J. 2001;45:449-468. Jilani A, Shirazi-Adl A, Bendjaballah M. Biomechanics of human tibio-femoral joint in axial rotation. Knee. 1997;4:203-213. Johnson D, Urban W, Caborn D, Canarthos W, Carlson C. Articular cartilage changes seen with magnetic resonance imaging-detected bone bruises associated with acute anterior cruciate ligament rupture. Am J Sports Med. 1998;26:409-414. Kajzer J, Matsui Y, Ishikawa H, Schroeder G. Shearing and bending effects at the knee joint at low-speed lateral loading. SAE Paper. 1999-01-0712. 48 Kannus P, Javinen M. Conservatively treated tears of the anterior cruciate ligament: long- term results. J Bone Jt Surg. 1987;69A:1007-1012. Kapandji I. The physiology of the joints. Vol. 2 Lower Limb. Churchill-Livingstone, Edinburg. 1987. Kapelov R, Teresi L, Bradley WG, Bucciarelli NR, Murakami DM, Mullin WJ, Jordan JE. Bone Contusions of the Knee; Increased Lesion Detection with Fast Spin-echo MR Imaging with Spectroscope Fat Saturation. Radiology. 1993;189:901-904. Kaplan P, Walker C, Kilcoyne R, Brown D, Tusek D, Dussault R. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: Assessment with MR imaging. Musculoskeletal Rad. 1992;183:835-838. Kaplan P, Gehl R, Dussault R, Anderson M, Diduch D. Bone contusions of the posterior lip of the medial tibial plateau (Countercoup injury) and associated internal derangements of the knee at MR imaging. Radiology. 1999;211:747-753. Kempson G. Relationship between the tensile properties of human articular cartilage from the human knee and age. Ann Rheum Dis. 1982;41:508-511. Kempson G. Age-related changes in the tensile properties of human articular cartilage: a comparitive study between the femoral head of the hip joint and the talus of the ankle joint. Biochem Biophys Acta. 1991;10752223-230. Kennedy J. Complete dislocation of the knee joint. J Bone Jt Surg. 1963;45:889-904. Kennedy J, Fowler P. Medial and anterior instability of the knee: An anatomical and clinical study using stress machines. .1 Bone Jt Surg. 1971;53-A21257-1270. Kennedy J, Weinberg H, Wilson A. The anatomy and function of the anterior cruciate ligament: As determined by clinical and morphological studies. J Bone Jt Surg. 1974;56-Az223-235. Kerrigan J, Bhalla K, Madeley J, Funk, J, Bose D, Crandall J. Experiments for establishing pedestrian-impact lower limb injury criteria. SAE Paper. 2003-01-0895. Klein T, Chaudhry M, Bae W, Sah R. Depth-dependent biomechanical and biochemical properties of fetal, newborn, and tissue-engineered articular cartilage. J Biomech. 40:182-190, 2007 Krosshaug T, Andersen T, Olsen 0, Myklebust G, Bahr R. Research approaches to describe the mechanisms of injuries in sport: limitations and possibilities. Br J Sports Med. 2005;39:330-339. Krosshaug T, Nakamae A, Boden B, Engebretsen L, Smith G, Slauterbeck J, Hewett T, Bahr R. Mechanisms of anterior cruciate ligament injury in basketball: Video analysis of 39 cases. Am J Sports Med. 2007a;35:359-367. 49 Krosshaug T, Slauterbeck JR, Engebretsen L, Bahr R. Biomechanical analysis of anterior cruciate ligament injury mechanisms: three-dimensional motion reconstruction from video sequences. Scand J Med Sci Sports. 2007b;17:508-519. Kujala U, Kettunen J, Paananen H. Knee osteoarthritis in former runners, soccer players, weight litters and shooters. Arthritis Rheum. 1995;38(4):539-546. Kwolek C, Sundaram S, Schwarcz T, Hyde G, Endean E. Popliteal artery thrombosis associated with trampoline injuries and anterior knee dislocations in children.Am Surg. 1998;64:1183-1187. Lambson R, Barnhill B, Higgins R. Football Cleat Design and its Effect on Anterior Cruciate Ligament Injuries. A Three Year Prospective Study. Am J Sports Med. 1996;24(2):155-159. Lane N. Physical activity at leisure and risk of osteoarthritis. Annals Rheumatic Diseases. 1996;55:682-684. Lee I, Paffenbarger R. Associations of light, moderate and vigorous intensity physical activity with longevity: The Harvard alumni health study. Am J Epidem 151;2000:293-299. Li G, Rudy T, Allen C. Effect of combined axial compressive and anterior tibial loads on in situ forces in the anterior cruciate ligament: a porcine study. J Orthop Res. 1998;16:122-127. Li G, Lopez 0, Rubash H. Variability of a three-dimensional finite element model constructed using magnetic resonance images of a knee for joint contact stress analysis. J Biomech Eng. 2001;123:341-346. Lohmander L, Stenberg A, Englund M, Roos H. High prevalence of knee osteoarthritis, pain and functional limitations in female soccer players twelve years after anterior cruciate ligament injury. Arth Rheum. 2004;50:3145-3152. Majewski M, Susanne H, Klaus S. Epidemiology of athletic knee injuries: A lO-year study. Knee. 2006;13:184-188. Markolf K, Mensch J, Amstutz H. Stiffiiess and laxity of the knee- the contributions of the supporting structures. A quantitative in vitro study. J Bone Jt Surg. 1976;63A2570-585. Markolf K, Bargar W, Shoemaker S, Amstutz H. The role of joint load in knee stability. J Bone Jt Surg. 1981;63:570-585. Markolf K, Kochan A, Amstutz H. Measurement of knee stiffness and laxity in patients with documented absence of the anterior cruciate ligament. J Bone Jt Surg. 1984;66- A:242-252. 50 Markolf K, Gorek J, Kabo J, Shapiro M. Direct measurement of resultant forces in the anterior cruciate ligament: An in vitro study performed with a new experimental technique. Am J Bone Jt Surg. 1999;72:557-567. Markolf K, Burchfield D, Shapiro M, Shepard M, F inerman G, Slauterbeck J. Combined knee loading states that generate high anterior cruciate ligament forces. J Orthop Res. 1995;13:930-935. Matsumoto H, Suda Y, Otani T, Niki Y, Seedhom B, Fujikawa K. Roles of the anterior cruciate ligament in preventing valgus instability. J Orthop Sci. 2001;6z28-32. McKellop H, Sigholm G, Redfern F, Doyle B, Sarmiento A, Luck Sr J. The effect of simulated fracture-angulations of the tibia on cartilage pressures in the knee joint. .1 Bone Joint Surg Am. 1991;73:1382—1391. McNair P, Marshall R, Matheson J. Important features associated with acute anterior cruciate ligament injury. N Zealand Med J. 1990;103:537-539. Meyer E. Biomechanical response of tissues to various blunt impact orientations on the knee. MSU Thesis. 2004. Meyer E, Haut R. Excessive compression of the human tibio-femoral joint causes ACL rupture. J Biomech. 2005;38:2311-2316. Meyers M, Harvey P. Traumatic Dislocation of the knee joint. J Bone Jt Surg. 1971;53- A:16—29. Miller C, Shelton W, Barrett G, Savoie F, Dukes A. Deltoid and Syndesmosis Ligament Injury of the Ankle Without Fracture. Am J Sports Med. 1995;23(6):746-750. Mills 0, Hull M. Rotational flexibility of the human knee due to varus/valgus and axial moments in vivo. J Biomech. 1991;24(8):673-690. Mink J, Deutsch A. Occult cartilage and bone injuries of the knee: Detection, classification, and assessment with MR imaging. Radiology. 1989;170:823-829. Morrison J. The mechanics of the knee joint in relation to normal walking. J Biomech. 1970;3251-61. Mow V, Mak A. Lubrication of diarthrodial joints. In: Handbook of Bioengineering, edited by Skalak R and Chien S. McGraw-Hill, New York. 1987 :5.l-5.34. Myklebust G, Holm I, Maehlum S, Engelbretsen L, Bahr R. Clinical, functional and radiologic outcome in team handball players 6 to 11 years after anterior cruciate ligament injury. A follow-up study. Am J Sports Med. 2003;31:981-989. Myklebust G, Bahr R. Return to play guidelines after anterior cruciate ligament surgery. Br J Sports Med 39;2005: 127-131. 51 Nagel D, States J. Dashboard and bumper knee — Will arthritis develop? Am Auto Accident Med. 1977;21:272-278. National Federation of State High School Associations. 2007—2008 Athletics Participation Summary. Indianapolis, IN. 2008. National Collegiate Athletic Association. NCAA injury surveillance system summary. Indianapolis, IN. 2001:80-89. Newberry W, Haut R. The Effects of Subfracture Impact Loading on the Patello-Femoral Joint in a Rabbit Model. Stapp Car Crash J. 1996;40:149-159. Nielsen S, Ovesen J, Rasmussen O. The anterior cruciate ligament on the knee: An experimental study of its importance in rotatory knee instability. Arch Orthop Trauma Surg. 1984;103:170-174. Nielsen S, Helmig P. Instability of knees with ligament lesions: Cadaver studies of the anterior cruciate ligament. Acta Orthop Scand. 1985;56:426-429. Noyes F, Delucas J, Torvik P. Biomechanics of anterior cruciate ligament failure: An analysis of strain-rate sensitivity and mechanisms of failure in primates. J Bone Jt Surg. 1974;56:236-253. Noyes F, Grood E. The strength of the anterior cruciate ligament in humans and Rhesus monkeys. J Bone Jt Surg. 1976;58:1074—1082. Noyes F, Matthews D, Mooar P, Grood E. The symptomatic anterior cruciate-deficient knee. Part II: the results of rehabilitation, activity modification, and counseling on functional disability 1983 meniscal tears after ACL. J Bone Jt Surg. 1983;65:163-174. Noyes F, Butler D, Grood E, Semicke R, Hefzy M. Biomechanical analysis of human ligament grafts used in knee-ligament repairs and reconstructions. Am J Bone Jt Surg. 1984;66:344-352. Olsen 0, Myklebust G, Engebretsen L, Bahr R. Injury mechanism for anterior cruciate ligament injuries in team handball: A systematic video analysis. Am J Sports Med. 2004;32:1002-1012. Pate R, Pratt M, Blair S, Haskell W, Macera C, Bouchard C, Buchner D, Ettinger W, Heath G, King A. Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. J Am Med Assoc. 1995;275:402-407. Paul J, Spindler K, Andrish J, Parker R, Secic M, Bergfeld J. Jumping versus nonjumping anterior cruciate ligament injuries: A comparison of pathology. Clin J Sport Med. 2003;13:1-5. 52 Pflum M, Shelburne K, Torry M, Decker M, Pandy M. Model prediction of anterior cruciate ligament force during drop-landings. Med Sci Sports Exerc. 2004;36:1949- 1958. Pinskerova V, Johal P, Nakagoawa S, Sosna A, Williams A, Gedroyc W, Freeman M. Does the femur roll-back with flexion? BrJ Bone Jt Surg. 2004;86:925-931. Powell J, Schootrnan M. A Multivariate Risk Analysis of Selected Playing Surfaces in the National Football League: 1980-1989. An Epidemiological Study of Knee Injuries. AmJ Sports Med. 1992;20(6):686-694. Powell J, Barber-Foss, K. Sex-related injury patterns among selected high school sports. Am J Sports Med. 2000;28:385-391. Pritsch M, Comba D, Frank G, Horoszowski H. Articular Cartilage Fractures of the Knee. J Sports Med. 1984;24:299-302. Pujol N, Blanchi M, Chambat P. The incidence of anterior cruciate ligament injuries among competitive alpine skiiers. A 25-year investigation. Am J Sports Med. 2007;35:1070-1074. Rangger C, Kathrein A, Freund M, Klestil T, Kreczy A. Bone bruise of the knee: Histology and cryosections in 5 cases. Acta Orthop Scand. 1998;69:291-294. Rauch G, Allzeit B, Gotzen L. Tensile strength of the anterior cruciate ligament in dependence on age. Proc Biomech Human Knee Ligaments. Ulm, Germany 1987:24. Raty H, Kujala U, Videman T, Impivaara O, Battie M, Sama S. Lifetime musculoskeletal symptoms and injuries among former elite male athletes. Int J Sports Med. 1997;18:625-632. Renstrom P, Ljungqvist A, Arendt E, Beynnon B, et a1. Non-contact ACL injuries in female athletes: an International Olympic Committee current concepts statement. Br J Sports Med. 2008;42:394-412. Repo R, Finlay J, Survival of Articular Cartilage After Controlled Impact. J Bone Jt Surg. 1977;59A: 1068-1076. Rettig A. Medial and lateral ligament injuries. In: Scott WN, ed. Ligament and extensor mechanism injuries of the knee: Diagnosis and treatment. St. Louis: Mosby Year Book; 1991. Riegger—Krugh C, Gerhart T, Powers W, Hayes W. Tibiofemoral contact pressures in degenerative joint disease. Clin Orthop. 1998;348:233-245. Roos H, Adalberth T, Dahlberg L, Lohmander LS. Osteoarthritis of the knee after injury to the anterior cruciate ligament or meniscus: the influence of time and age. Osteoarth Cart. 1995;3z261-267. 53 Sanders T, Medynski M, Feller J, Lawhom K. Bone contusion patterns of the knee at MR imaging: Footprint of the mechanism of injury. RadioGraphics. 2000;20:S135-S151. Schenck R, Kovach I, Agarwal A, Brummett R, Ward R, Lanctot D, Athanasiou K. Cruciate injury patterns in knee hyperextension: A cadaveric model. J Arthro Rel Surg. 1999;15:489-495. Schinagl R, Gurskis D, Chen A, Sah R. Depth-dependent confined compression modulus of full-thickness bovine articular cartilage. J Orthop Res. 15:499-506, 1997. Seefeldt V, Ewing M. Youth sports in America: An overview. PCPFS Res Digest. 1996;2(11):l-18. Seering W, Piziali R, Nagel D, Schurman D. The Function of the Primary Ligaments of the Knee in Varus-Valgus and Axial Rotation. J Biomech. 1980;13:785-794. Sell T, Ferris C, Abt J, Tsai Y, Myers J, Fu F, Lephart S. Predictors of proximal tibia anterior shear force during a vertical stop jump. J Orthop Res. 2007;25:1589-1597. Senter C, Hame S. Biomechanical analysis of tibial torque and knee flexion angle: Implications for understanding knee injury. Sports Med. 2006;36(8):635-641. Shelboume K, Klootwyk T. Low-velocity knee dislocation with sports injuries: Treatment principles. Clin Sports Med. 2000;19:443-456. Shoemaker S, Markolf K. In Vivo Rotary Knee Stability; Ligamentous and Muscular Contributions. J Bone Jt Surg. 1982;64-A:208-216. Shoemaker S, Markolf K. Effects of joint load on the stiffness and laxity of ligament- deficient knees: An in vitro study of the anterior cruciate and medial collateral ligaments. J Bone Jt Surg. 1985;67-Azl36-146. Shoemaker S, Markolf K, Dorey F, Zager S, Namba R. Tibial torque generation in a flexed weight-bearing stance. Clin Orthop Relat Res. 1988;228:164-70. Shoemaker S, Daniel D. The limits of knee motion: in vitro studies. Eds. Daniel DM, Akeson WH, O’Connor JJ. Knee ligaments: structure, function, injury and repair. NY: Raven Press. 1990:153-161. Shorten M, Hudson B, Himmelsbach J: Shoe-surface traction of conventional and in- filled synthetic turf football surfaces. Proc XIX International Congress of Biomechanics, University of Otago, Dunedin, New Zealand: 2003. Snearly W, Kaplan P, Dussault R. Lateral-compartment bone contusions in adolescents with intact anterior cruciate ligaments. 1996;198:205-208. Speer K, Spritzer C, Basset F, Feagin J, Garrett W. Osseous injury associated with acute tears of the anterior cruciate ligament. Am J Sports Med. 1992;20:382-389. 54 Speer K, Warren R, Wickiewicz T, Horowitz L, Henderson L. Observations on the injury mechanism of anterior cruciate ligament tears in skiers. Am J Sports Med. 1995;23:77-81. Spindler K, Schils J, Bergfeld J, Andrish J, Weiker G, Anderson T, Piraino D, Richmond B, Medendorp S. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. Am J Sports Med. 1993;21:551-557. Stanitski C, McMaster J, Ferguson R: Synthetic turf and grass: A comparative study. Am J Sports Med. 1974;2(1):22-26. States JD. Adult occupant injuries of the lower limb. Proc Symp Biomech. 1986:97-107. Sullivan D, Levy 1, Sheskier S, Torzilli P, Warren RF. Medial restraints to anterior- posterior motion of the knee. J Bone Jt Surg. 1984;66A2930-936. Tashman S, Collon D, Anderson K, Kolowich P, Anderst W. Abnormal rotational knee motion during running after anterior cruciate ligament reconstruction. Am J Sports Med. 2004;32:975-983. Tepper S, Hochberg M. Factor associated with hip osteoarthritis: data from the First National Health and Nutrition Examination Survey. Am J Epidemiol. 1993;137:1081- 1088. Thambyam A, Goh J, Das De S. Contact stress in the knee joint in deep flexion. Med Eng Phys. 2005;27:329-335. Torzilli P, Deng X, Warren R. The effect of joint-compressive load and quadriceps muscle force on knee motion in the intact and anterior cruciate ligament-sectioned knee. Am J Sports Med. 1994;22:105-112. Torzilli P, Grigiene R, Borrelli J, Helfet DL. Effect of impact load on articular cartilage: Cell metabolism and viability and matrix water content. J Biomech Eng. 1999;121:433-441. US. Department of Health and Human Services. Physical activity and health: A report of the Surgeon General. Sudbury, Ma: Jones & Bartlett Pub. 1998:173-208. Vellet A, Marks P, Fowler P, Munro T. Occult Post-Traumatic Osteochondral Lesions of the Knee: Prevalence, Classification and Short-term Sequelae Evaluated with MR Imaging. Radiology. 1991;178:271-276. Viskontas D, Giuffre B, Duggal N, Graham D, Parker E, Coolican M. Bone bruises associated with ACL rupture. Correlation with injury mechanism. Am J Sports Med. 2008;36:927-933. Volpin G, Dowd G, Stein H, Bently G. Degenerative Arthritis After Intra-articular Fractures of the Knee: Long Term Results. J Bone Jt Surg. 1990;72:634-638. 55 Von Porat A, Roos E, Roos H. High prevalence of osteoarthritis 14 years after an anterior cruciate ligament tear in male soccer players: A study of radiographic and patient relevant outcomes. 2004;63:269-273. Wang J, Rubin R, Marshall J. A mechanism of isolated anterior cruciate ligament rupture. Am J Bone Jt Surg. 1975;57:411-413. Williamson A, Chen A, Masuda K, et a1. Tensile mechanical properties of bovine articular cartilage: variations with growth & relationships to collagen network components. J Orthop Res. 21:872-880, 2003. Withrow T, Huston L, Wojtys E, Ashton-Miller J. The relationship between quadriceps muscle force, knee flexion, and anterior cruciate strain in an in vitro simulated jump landing. Am J Sports Med. 2006;34:269-274. Wolf B, Amendola A. Impact of Osteoarthritis on sports careers. Clin Sports Med 24;2005: 187-198. Woo S, Hollis J, Adams D, Lyon R, Takai S. Tensile properties of the human femur- anterior cruciate ligament-tibia complex: The effect of specimen age and orientation. Am J Sports Med. l991;19(3):217-225. Woo S, Kanamori A, Zeminski J, Yagi M, Papagerorgiou C, Fu F. The effectiveness of reconstruction of the anterior cruciate ligament with hamstrings and patellar tendon: a cadaveric study comparing anterior tibial and rotational loads. J Bone Jt Surg. 2002;84:907-914. Wright V. Post-traumatic osteoarthritis- A medico-legal minefield. J Rheum. 1990;29:474-478. Yeow C, Cheong C, Ng K, Lee P, Goh ll. Anterior cruciate ligament failure and cartilage damage during knee joint compression. Am J Sports Med. 2008;36:934-942. Yildirim G, Walker P, Susman-Fort J, Aggarwal G, White B, Klein G. The contact locations in the knee during high flexion. Knee. 2007;14:379-384. Yu B, Garrett W. Mechanisms of non-contact ACL injuries. Br J Sports Med. 2007;47:i47-i51. Zantop T, Herbort M, Raschke M, F u F, Peterson W. The role of the anteromedial and posterolateral bundles of the anterior cruciate ligament in anterior tibial translation and internal rotation. Am J Sports Med. 2007;35:223-227. 56 CHAPTER TWO ACL INJURY INDUCED BY TIBIOFEMORAL COMPRESSION PRODUCING ANTERIOR SUBLUXATION OF THE TIBIA IN UN CON STRAIN ED HUMAN KNEE JOINTS ABSTRACT The ACL is a primary stabilizer of the knee for TF joint compression. During a majority of sports related non-contact ACL injuries, the injured leg supports nearly all of the ground reaction force, implying that high TF compressive loads are common during ACL injuries. However, clinical studies analyzing video of sports injuries frequently measure large amounts of external tibial rotation and valgus bending. The hypothesis of this study was that the magnitudes and types of motion induced by TF compression would significantly change after ACL rupture from the relative joint displacements present just before ACL injury. Compression experiments were conducted on 13 knee joints with repetitive tests at increasing intensity until catastrophic failure. ACL injury was documented in all cases at 6.0:l:l.9 kN. The femur displaced posteriorly relative to the tibia in pre-failure tests and with a higher magnitude in failure tests. During compression there was either valgus femur or internal tibia rotation in pre-failure tests, and significantly increased amounts of valgus femur rotation or external tibia rotation after the ACL failed. These new data show that the joint motions can vary in magnitude and direction before and after failure of the ACL. Biomechanical studies do not support external tibia rotation or valgus bending as mechanisms of isolated ACL rupture, although the current study showed that these motions do likely occur after ACL rupture when a large TF compressive load is present. 57 INTRODUCTION Biomechanical studies confirm that the ACL is a primary stabilizer of the knee for tibiofemoral (TF) joint compression (Torzilli et al., 1994; Li et al., 1998). In subjects undergoing arthroscopic surgery, but with normal gait and an intact ACL, a strain transducer was implanted on the ACL and a variety of external loads were applied, including 40% weight bearing (Flemming et al., 2001). This study reports that at 20° of knee flexion, weight bearing significantly increases ACL strain. The tibial plateau surface has an inherent posterior slope of 10-15° relative to the long axis of the tibia (Li et al., 1998) that varies across the medial and lateral facets. This inclination creates a coupled anterior tibial translation and axial rotation which increases the distance between the tibial and femoral insertions of the ACL (Blankevoort et al., 1996). A similar effect, called “anterior neutral shift” of the tibia, has been described for externally applied TF compressive loads (Torzilli et al., 1994). Since the ACL provides 85% of the ligamentous restraining force during anterior displacement of the tibia (Butler, 1989), these motions are of particular importance in ACL injury mechanism studies. Few studies have documented ligamentous force or relative joint displacements at failure levels under controlled loading of the knee joint. The goal of this study was to induce ACL rupture in the 30° flexed knee joint by isolated TF compression. This force will result in relative motion between the tibia and femur that strains the ACL. The maximum relative motions of the knee joint before and after failure of the ACL are also important for injury mechanism identification from video analysis of clinical ACL tears. The hypothesis of the study was that the magnitudes and types of motion observed after 58 ACL rupture would significantly change from the relative joint displacements present just before ACL injury. Specifically, the ACL is not a primary restraint to external tibia rotation (Markolf et al., 1995), but this motion is frequently identified as a mechanism of injury in video analysis of ACL injuries. Valgus collapse of the knee is also commonly identified in videos of ACL rupture, but Olsen et al. (2004) comments “whether the consistent valgus collapse observed in the videos was actually the cause of injury or simply a result of the ACL being torn is open for discussion.” Data from the current study may help explain the results of video based studies and provide a framework for the relative joint displacements that are important for determining ACL injury risk. METHODS Compression experiments were conducted on one TF joint from 13 cadavers (53.5i7.9 yrs) that were procured through university sources (see Acknowledgements). One joint was randomly chosen from each cadaver specimen pair and sectioned approximately 15 cm proximal and distal to the center of the knee. The skin and muscle tissues were removed leaving the knee joint capsule and collateral ligaments intact. The femur and tibia shafts were cleaned with 70% alcohol and potted in cylindrical aluminum sleeves with room temperature curing epoxy (Fibre Strand, Martin Senior Corp., Cleveland, OH). In one series of experiments (Series 1, Table 2.1), a sleeve bearing and horizontal stabilizer bar were attached to a servo-hydraulic materials testing machine (Model 1331, lnstron Corp., Canton, OH) to prevent a bending moment in the load transducer (Model #10101a-2500, lnstron Corp.). An additional thrust bearing allowed axial, intemal/extemal (IE) tibia rotation which was recorded by a rotary encoder (Model 59 #RCH25D-6000, Renco Encoders Inc., Goleta CA) (Figure 2.1A). The potted femur was secured to a fixture with the knee joint flexion angle set at 30° and the varus/valgus angle adjusted and fixed in a position where the femoral condyles were perpendicular to the tibia. This fixture was attached to an X-Y translational plate to allow posterior/anterior (PA) and medial/lateral (ML) motions of the femur relative to the tibia during loading which were recorded with linear encoders (Model #XOOZOIA, Renishaw, Hofrnan Estates IL). In the second series of experiments (Series 2, Table 2.2), femur varus/valgus (V V) rotation was unconstrained and also recorded by a rotary encoder, but tibia axial rotation was prevented (Figure 2.1B). PA and NIL displacement of the femur were also unconstrained and measured relative to the tibia similar to the series 1 experiments. These experiments were carried out on a slightly different servo—hydraulic materials testing machine (Models 312.21 and 204.52, MTS Corp., Eden Prairie, MN) and load transducer (Model FF L(l8/i12)U-(3/i2)SP/0355, Strainsert Co., West Conshohocken, PA). ‘ ‘ Linear A Linear B Actuator Actuator LP 1 " Load Cell I t t l U Rotary Encoder X-Y Plate 1 r 308 Flexion. .VarusNalgus An gle: ”Rotary Encoder ' W Moment Femur ' Arm Femur . : I 330° Flexion : iAngle I l A oad Cell l I J 8 PI I 1 XWPlate 1 I , l , Figure 2.1 Knee specimens potted, flexed, and attached to the compressive testing fixtures. A) Series 1: Axial rotation of the tibia was allowed. B) Series 2: VarusNalgus rotation of the femur was allowed. 60 All joints were repeatedly loaded using similar protocols with slightly varying parameters for compressive preload, control method, and rate of loading. Dynamic compressive loads were applied along the tibial axis by the linear actuator with a single haversine waveform. In series 1 (Table 2.1), the compression was applied in load control and increased by increments of 1000 N in successive tests until catastrophic injury of the joint. In some specimens a 5 kg weight was also attached to the femur via a pulley to apply a slight posterior preload. In series 2 (Table 2.2), the compression was applied in displacement control and increased by increments of 2 mm in successive tests until catastrophic injury. The injury type and failure load were documented for each specimen. Table 2.1 Series 1 specimen information and test parameters. S ecimen‘ Age , Se Heightlw eightCompressivel Posterior ' Rate 9 . (yrs) ' i (m) I (kg) (Preload (N) 'Preload (N); (Hz) 32057 L 50 l M l1.67 l 64 100 50 10 32058 R 64 ' F 1.73 l 73 100 50 10 32081 R 52 F 1.70 1 75 50 50 10 32087 R 54 - M 1.93 114 50 50 10 32153 R 59 M 1.78 100 50 50 . 10 . . . l 3 32181 L 54 M 1.78 84 50 i 0 2.5 .i. Lifliflc . «w..- . 1 .. _ .. . - . _ . -..- _ . .. __—_J 32184 L 62 F 11.75 . 52 50 ; 0 4 1 r . Table 2.2 Series 2 specimen information and test parameters. 13:. = 0:19,“ “1:3,": 05.2.2381“: is: f 32416 L 34 M l 1.83 68 100 2.5 32284 R 53 M i 1.83 I 111 100 2.5 32273 L i 54 M 1.75 78 100 2.5 T 32302 L 59 M 1.78 86 100 2.5 32516 R1 47 M i 1.47 113 100 2.5 32388 L N/A M 1 N/A 1 N/Am 100 2:5 61 The peak load, time to peak load and actuator displacement corresponding to the peak load (proximal tibia motion) were recorded in each test. The relative joint motions, PA and ML displacement along with IE rotation or VV rotation were documented at their peak values, regardless of when they occurred relative to the loading. For simplicity, only the pre-failure data from the test immediately prior to gross joint failure was considered in this study. Paired t-tests were used to test for significant differences (p<0.05) between these variables for the pre-failure and failure tests, and unpaired t-tests were used to compare data from different types of failure and for different test parameters. RESULTS The average number of compression tests performed on each specimen prior to gross failure of the knee joint was 7 (range 4-12). The peak posterior displacement of the femur in pre-failure tests was approximately 12 mm in both series of experiments (Tables 2.3 and 2.4). In most series 1 specimens there was medial displacement of the femur. In series 2 experiments there was no clear medial or lateral displacement pattern documented. All series 1 experiments, except specimen 32081R, had internal rotation of the tibia develop during joint compression. Series 2 experiments had a slight trend towards val gus femur rotation. 62 Table 2.3 Data from series 1, pre-failure experiments. ML indicates medial (+) and lateral (-), PA indicates posterior (+) and anterior (-), IE indicates internal (+) and extemal(-). .40 x \o“ 8&0 6“ c) «0%? \ o&®&@ 06‘0" ($9 69‘ $06“ .63 0Q 09% (5‘ o 0 (’0 o69 S"; of}Q ®b\ \3 00‘ {(0 00‘ Q/«\ \- o x b b \o o gs <9 QV‘ 6‘ \ \OS C’09Q 006 «$6 V0 1'3 \90 Q'PQ 6’?» 015" \edb 06“ (6‘9 Q0“— OS" ’99 Q06 «Q QQ’ Q 0'39 0'69 Q‘ 32057L 0.05 8.1 g 6.1 3.4 12.2 8.1 32058R 0.05 2.6 3.5 4.5 6.2 t 2.6 32081R 0.05 4.8 5.4 8.8 9.1 g -2.8 32087R 0.06 6.3 7.6 2.1 12.8 ,, 1.4 32153R 0.05 5.1 ,_ 4.6 -5.6 12.7 3.7 , 32181L 0.21 4.9 6.4 ,, 3.9 22.7 ,, 8.4 32184L 0.14 2.9 4.5 -2.7 9.4 5.9 Avg (SD) 4.8(1.9) 5.4(1.4) 2.1(4.8) 12(52) 3.9(40) _, Table 2.4 Data from series 2, pre-failure experiments. ML indicates medial (+) and lateral (-), PA indicates posterior (+) and anterior (-), W indicates valgus (+) and varus << __32416L , 0.21 7.7 8 ., .3 32284R 0.2 5.6 9 2.5 11 .04 32273L 0.2 5.6 7 0 32302L 0.21 3.7 5 3 7 ’32516R 0.2 _ 4.7 f_ -1.6 7 70 ,4 32388L 0.21 7.7 7 -2.3 14 2.6 Avg (SD) 5.8 (1.6) 7.2 (1.3) -0.8 (1.9) 12 (4.3) 2.1 (2.6) Failure tests showed an average higher peak compressive force than in the pre- failure tests (Tables 2.5 and 2.6) and a slightly lower time to peak load (Figure 2.2). The peak posterior displacement of the femur was significantly increased in failure versus pre-failure tests with a combined mean value of 262813 mm between both series of experiments. This maximum value occurred after most of the compressive load had been 63 released due to the injury (Figure 2.3). In series 1 experiments, the peak ML displacement was not significantly changed after failure of the ACL, but followed the general pattern of displacement established in the pre-failure tests. In series 2 experiments there was a significant increase in the lateral displacement of the femur in the failure tests. The IE rotation of the tibia in series 1 was significantly different between failure and pre-failure tests. During early loading in failure tests the tibia rotated internally like in the pre-failure tests, but after the peak load and ACL failure the direction of rotation changed (Figure 2.3), and the peak rotation was in the external direction for all specimens with a mean value of 6.1:t4.3°. In series 2 experiments there was a significant increase in the valgus femur rotation after failure (Figure 2.4). 7 ’ ——Pre-failure A6 , g , —Fai|ure 'U 5 i M 3 o 4 I .2 :1: 3- 2 Q n E 2 o O 1 a 0 . 1 -..,¥_.__.,,,, , .-- 4"“— v, 0 0.1 0.2 03 0.4 0.5 Time (sec) ' Figure 2.2 Representative (32273L) load/moment versus time plots for pre-failure and failure tests. 35 - - - - Posterior Femur Displacement (mm) 30 - -- "Proximal Tibia Displacement (mm) ’ . ' L # . — - Medial Femur Displacement (mm) , ’ 25 " Compressive Load (kN) . ' 20 - — -— External Tibia Rotation (deg) . ’ Failure ' , ' ' 20 40" 60 80 100 -10 Time (msec) Figure 2.3 A representative load and motion versus time plot during a failure test (32057L). The peak compressive loads and the corresponding proximal tibia displacements occurred at the failure time point as marked. All other motions were measured at their peak values, indicated by #. ---‘Posterior Displacem ent ~~ Proximal Displacement 30 1 ——MedialDisplacement# —Load(kN) —Valgus Rotation . 25 *5 Failure 1 5 10* 00 0001000534.... : -’ r\ r/ r .‘ W—w ., 0 10 20 V. 80 40 M0 / '. ,' -10 ~' Time(msec) Figure 2.4 A representative load and motion versus time plot during a failure test (32284R). The peak compressive loads and the corresponding proximal tibia displacements occurred at the failure time point as marked. All other motions were measured at their peak values, indicated by #. 65 Table 2.5 Data from series 1, failure experiments. #Significant difference between pre- failure and failure tests. 0 c9 \. 80°98 82,6 8“ s e“ s ,s x (25" (.50 \oq \‘9 06‘ 8% 0° '66 \z @0 Q ~99 8 66 O ’b \‘b v0 ‘5 (5’2 (S c} «0 o S 0 4.9 \t- ‘i‘ 00 Q6 «\ \, \{:\ \/ \ 0% QQIb. Q\’b 32057L 0.05 8.6 6.8 5.9 33.8 -10.7 32058R 0.05 3.2 4.3 72 1_2.6 f, -2.7 32081R 0.05, 5.4 5.8 9.7 _ 21.4 -,11.9, 32087R 0.05 7.2 7.8 3.7 13.8 _ -1._7 32153R 0.05 4.5 7.2_ _ -7.2 _17.2_ -5.5 32181L 0.20_ _ 5-8. 7.4 __ 3.7_ 52.1 ;8.3_ 32184L 0.10 3.3 4.8 -12.2 40.7 -1.6 Avg (SD) 5.4(2.0)# 6.3 (1.4)# 1.5(8.fl 27 (15)# -6.1(4.3)# Table 2.6 Data from series 2, failure experiments. #Significant difference between pre-failure and failure tests. *Significant difference between series 1 and series 2. ‘5 o 0 \ \ \‘9 V Q t <9 s <9 s e,“ 07’6 39% ‘4} <9 86‘0 \© 0&0 66‘ eeoo°® .(e \/ 09 \g 15° \6 Q .59 Q Q9 Q 0° 99* Q‘ 8* 06‘ b \X‘V 6° QT 31° \F §0° (9Q \0Q 006‘ $00 ‘9‘» 16$ 6Q? 03*. 660 oéqp‘? ‘60 631- 0'69 Q <55 Q <36 Q Q Q «8 32416L 0.18 8.8 10 -8.7 33 4.9 32284R ‘ 0.16 5.8 11 -5.5 28 11.2 32273L 0.18 7 6.8 9 -10.4 31 8.0 32302L 0.17 5.2 7 -5.9 26 6.5- 32516R 0.15 5.1 9 -1.4 8 13.3 32388L 0.2 8.5 8 -3.5 16 6.5 Avg (SD) 6.7 (1.6)# 9 (1.4)# -5.9 (3.3)115‘H 24 (10)# 8.4 (3.2)# All failures involved the ACL, including 8/13 specimens with a complete or partial midsubstance ACL rupture, usually near the femoral insertion (Figure 2.5). In the cases of partial rupture, damaged fibers were largely observed in the posterior-lateral bundle of the ACL. The remaining joints suffered avulsion fractures at the insertion of the ACL into the tibia (Tables 2.7 and 2.8). The peak compressive load for midsubstance ruptures was 6.8 kN, while the peak load for avulsion fractures was significantly lower at 4.3 kN. None of the relative joint motions differed significantly between these types of 66 joint failure, although the statistical power in this comparison was low due to small group sizes. 32184L) specimens. Table 2.7 Dissection documentation of injured structures following series 1 failure tests. l Specimen ' ACL PCL MCL 1 LCL Fracture Posterior-lateral near I 32057 L . . Intact Intact Intact No femoral insertion 32058 R AvuISIon fracture at t'b'al Intact Intact Intact No insertion 32081 R AvuIsron fracture at “me" Intact Intact Intact No insertion 1 32087 R 1 Partial tear nearfemoral l Intact . Intact Intact ‘ No insertion | ,, _ f l . __ I 32153 R Posterior lateral-near Intact Intact ‘ Intact No femoral insertion . o o t I 32181 L Complete. tear near . Avulsion from Intact Avul5ion from . No femoral insertion tibial insertion femoral insertion f v 7 7 . v, - .. | 32184 L AvuIsron fracture at tibial | Avulsion from Intact Intact No 1 insertion tibial insertion 67 Table 2.8 Dissection documentation of injured structures following series 2 failure tests. Specimen ACL PCL MCL LCL Fracture 32416 L Anterior-mIediaInear Intact Intact Intact No femoral insertion 32284 R Completetear near Intact Intact Intact No femoral insertion 32273 L Complete.tear near Intact Intact Intact No femoral insertion 32302 L P°°t°”°"'.°t°'°'."°°' Intact Intact intact No femoralinsertion . 32516 R Avulsion. fracture at Intact Intact Intact Lateral tibial insertion Plateau 32388 L Posterior-lateral'near Intact Intact Intact Medial femoral insertion Plateau DISCUSSION ACL injury occurred under excessive TF compression as the femur displaced posterior relative to the tibia until ACL failure and continued afterwards (Figure 2.6). It was especially interesting that the direction of tibia rotation was changed from internal rotation in pre-failure tests to external rotation after failure of the ACL. Initial Posrtion Final Position '1 f I "r 'le 9"” ' " 'a'l . IM ' ." ‘1 Figure 2.6 Sagittal view, for a fixed tibia, of the relative TF joint motion during compression experiments, resulting in ACL failure. Most biomechanical evaluations of knee joint response under compressive external loading have been at low force levels, or to compare relative joint motions for 68 intact and ACL sectioned knees. Isolated ACL rupture occurred in a previous study that applied high TF compressive loads to knees flexed 60, 90 and 120° (Meyer and Haut, 2005). The maximum force in those ACL failure experiments was 5.1i2.1 kN compared with 6.0i1.9 kN in the current study. In the previous experiments, allowing posterior displacement of the femur induced ACL rupture, while allowing other displacements but constraining posterior femur displacement produced femoral condyle and tibial plateau fractures (Jayaraman et al., 2001). Posterior displacement of the femur occurs during TF compression (applied along the tibial axis) due to the posterior slope of the tibial plateau (Torzilli et al., 1994). The ACL is a primary restraint to this relative anterior tibial translation because of its orientation in the sagittal plane (Butler, 1989). In contrast, axial rotation of the tibia in the compression experiments likely did not induce additional strain in the ACL. Li et a1. (1998) has shown that the addition of a compressive load increased anterior tibial translation and force in the ACL, but had no effect on the degree of internal tibia rotation at 30° of flexion. These coupled motions were likely due to the concave geometry of the medial facet versus the slightly convex lateral facet (Figure 2.7) (Kapandji, 1987). Before ACL failure there was more relative displacement in the lateral compartment, producing internal tibial rotation in pre-failure tests (Matsumoto, 1990). After ACL rupture, relative displacement in the medial compartment was released and produced external rotation of the tibia (Figure 2.8). This “countercoup” effect after ACL failure was suggested by Kaplan et al. ( 1999) to explain the mechanism that produces medial compartment bone contusions in ACL injured patients. In this scenario, a lateral compartment contusion occurs first during anterior tibia subluxation on an internally 69 rotated tibia. Then, after failure of the ACL the anterior tibia subluxation continues, but the tibia rotates externally relative to the femur. High Slope M“ " Anteriorly

Avg , Max Ma iArea> Avg , Max 25M3a Presme ‘ Pressue i 25NPa Presare Preemie ("mil (MPa) (mar) e (MPa) ' 3.205”- _ -131 1.35. _ -21. 1233 - 9:891. 35 _._, 19- __39._ 32058R _ 0-- _o _ 7 _ 19“ _ 42 ‘o_ 12 _ 21_ ' 32081R - 209 __ 1“ 14 g“ ”76“ __7 _17 4o_ 132037R_ 15L -1?" _18_ _35 1 _£81_. ,0 -2 12 _ 3.11.. : 32153R .- 149 o _ 13~ . __22_ 9%. _o_ _ 11 f _19_ '32181L__36 ___o_ "11.- .19 161 -26- -..18 ; 46“ . 32184L 49 o 11 18 93 _ 2 13 29 Avg(SD)l1(B(77) 8(14) 14(5) , 24(10) 148(106), 10(14) 15(3) ‘ 31 (11) _ Table 3.3 Pressure film data from series 2 prefailure experiments. Medial Conpartment Lateral Compartment i i ... .2.::p.t.:..~..::.. .... 9.2.22... ..2. — 7 "7 (M__ ‘17 (Min? ___(M_-__. "tire-)7 '7 32416L 217 . 4 16 31 228 o ' 16 25 32284R 65 ‘ o 12 19 8 0 i 11 16 32273L o i o 7 1o 8 o , 11 17 32302L 0 j o 7 10 o o I 6 1o 32516R o o 7 10 189 2 '1 17 29 32388L 228 i 51 20 39 1 o , 11 14 £50) 85(109) ' 9(20) 12(6) ‘ 20 (13) 72(106) 0(1) 12(4) 19(7) 84 Table 3.4 Pressure film data from series 1 failure experiments. Medial Conpartment Lateral Con'partment Area > A Max Area > A Max Area 25 W3 Pregue Presire Area 25 MPa Prague Premre (mm’) (Wa) (Inn?) (Wa) 32057L 306 88 21 37 230 72 20 39 32058R o o 7 10 27 2 14 32 32081R 213 2 a 14 28 75 5 15 4o 32087R 278 61 19 50 260 0 12 23 32153R 159 3 15 33 23 0 11 23 32181L 281 11 16 36 246 50 20 50 32184L 73 8 14 47 130 o 12 22 Av SD 187(116) 25(35) 15(4) 34(13) 142(104) 18(29) 15(4) 33(11) Table 3.5 Pressure film data from series 2 failure experiments. Medial Cormartment Lateral Con'partment Area Area > Avg Max Area Area > Avg Max 25 Pa Plume Pressure 25 MPa Pressure Presire (rm?) (W8) (mm (MPa) 32416L 240 51 20 44 250 6 18 30 32284R 173 0 14 28 66 0 14 22 32273L 85 O 14 24 150 0 12 24 32302L 148 1 16 30 3 0 11 18 32516R 14 0 12 18 195 7 17 30 1 32388L 246 61 21 39 12 0 12 23 M30) 151 (90) 19 (29) 16 (3) 30 (10) 113 (101) 2 (3) 14 (3) 24(5) Anterior B 3 MPa 35 MPa Posterior Figure 3.4 Representative shape and magnitude of the pressure distribution on the tibial plateau of failure experiments with axial rotation (A: 32181) or valgus rotation (B: 32416) unconstrained. 85 The relative MRI signal intensity from the cartilage increased from the pre to post-soaked scans in all specimens, with average increases of 15% to 30% depending on the region (Figure 3.5). The largest increase was in the posterior-medial region. 60°/o - o: o o\° l 40% - 30% - 20% - Change in Relative Signal A ‘36 06 o\ o\ I I Anterior Central Posterior Anterior Central Posterior Medial Compartment Lateral Compartment Figure 3.5 The relative increase in signal intensity of the tibial plateau cartilage between pre-soak and post-soak MRI scans. Analysis of the histological samples revealed occult injuries for all specimens following gross failure, and were divided into three regions: anterior, central, and posterior (Figure 3.6). Microdamage was located central to posterior in both compartments of these compression specimens. In both facets of the compression specimens the damage in the central and posterior regions were not different, but both regions had significantly more damage than the anterior region. There were similar amounts of damage in the two series of TF compression experiments and between medial and lateral facets. Specific types of damage that were noted were vertical microcracks, which were located in the deep cartilage zone, and typically extended into the subchondral bone; and horizontal microcracks, which were located along the tidemark and ran parallel with the articular surface. There were also fissures of the articular 86 cartilage that were highlighted when the surface was wiped with india ink (Figure 3.7). The location of subchondral bone and cartilage damage was central or posterior on both facets of compression specimens. Lateral Medial Figure 3.6 Average 1: SD histological scores for regions within the medial and lateral facets of series 1 (A) and series 2 (B) specimens. 2i;i"-Significant differences from the central and posterior regions. Figure 3.7 The appearance of the surfaces after staining with India ink for representative specimens from series 1 (A & C) and series 2 (B & D) experiments. 87 DISCUSSION Compression experiments resulted in regions of high contact pressure on the posterior medial and lateral tibial plateaus. MRI was used to measure the relative fluid uptake of the cartilage after experimental testing and soaking for 24h in PBS. This provided an indirect method of evaluating the amount of firnctional matrix damage to cartilage (Grushko et al., 1989). While the cartilage from all specimens underwent some increase in signal, the center and posterior medial plateau cartilage had the most fluid uptake. This occurred in a region of the cartilage where high contact pressures were developed. While bone bruises in the medial tibial plateau are less cormnon than in the lateral compartment (Kaplan et al., 1992), medial compartment knee OA is a more frequent and conspicuous condition. Therefore, any damage in that region may be of particular importance (Ahlback, 1968; Davies-Tuck et al., 2008). A common injury scenario for ACL rupture includes a valgus bending moment at the knee produced by abduction of the femur that results in high stresses in the lateral compartment (Kaplan et al., 1992). However, the current contact pressures were distributed over both the medial and lateral tibial plateaus with the average contact force not exceeding a 60%-40% distribution between compartments. In subfailure experiments there were higher forces in the lateral compartment, but in failure experiments the higher contact pressures and areas occurred in the medial compartment. A “countercoup” effect after ACL failure was suggested by Kaplan et al. (1999) to explain the mechanism that produces medial compartment bone contusions during ACL injuries. In this scenario the medial compartment contusion occurs after a lateral compartment contusion during the compensatory varus bending that may occur as the knee forces are reducing. The authors’ 88 conclusion that, “Only injuries with large forces could result in medial bone contusions...” is also supported by the current data where differences in contact force distribution between subfailure and failure tests were documented in compression specimens. This theory, however, does not take into account baseline differences in the cartilage and subchondral bone material properties of the two plateaus. The bone mineral density (BMD) of the tibial plateau for a young, non-OA population is approximately 15% higher in the medial compartment than in the lateral compartment (Hurwitz et al., 1998). And, the thickness of medial cartilage is approximately 21% lower than that in the lateral compartment (Muhlbauer et al., 2000). The lower BMD of the lateral compartment would correspond to an approximately 50% lower ultimate failure stress (Goldstein et a1 1983). This may explain why bone bruises are more common in the lateral compartment, even when medial to lateral compartment loading is of similar magnitude during ACL rupture. Additionally with thinner cartilage medially and assuming comparable medial and lateral moduli, one may expect higher shear stresses to occur in overlying cartilage in the medial compartment than in the lateral compartment for similar contact pressures (Haut Donahue et al., 2002; Li et al., 1998). A previous study was conducted by this laboratory in which compressive TF joint loading was applied, until ACL failure, with the knees flexed either 60, 90, or 120° and varus-valgus constrained (Jayaraman et al., 2001). In those experiments the average pressures developed in the medial compartment also exceeded those generated in the lateral compartment. Additionally, the medial compartment contact pressures in each flexion case exceeded the average pressures generated in the current study, while the pressures in the lateral compartment compared well to those of the current study. These 89 previous data and the corresponding current data suggest that the incidence of cartilage damage and bone bruising may also depend on the knee joint flexion angle during excessive compression. The histological analysis revealed vertical microcracks at the interface between cartilage and subchondral bone on both the medial and lateral plateaus, without signs of gross tibial plateau fracture. These regions of microdamage generally coincided with areas that developed the highest contact pressures. To our knowledge, this is the only study to document controlled, experimentally induced occult microcracks at the cartilage- bone interface in ACL injured human cadaver knees. In a previous study, occult microcracks were also documented in the tibial plateau of knee specimens subjected to TF impacts at similar compressive loads (Banglmaier et al., 1999). Those experiments simulated automotive accidents. The knee joint was therefore flexed 900 and relative motion between the tibia and femur was constrained to prevent injury to the ACL (Jayaraman et al., 2001). The current study also documented additional damage and matrix compression lines in regions of the cartilage overlying microcracks at the cartilage-bone interface in compression specimens. This demonstrated how localized chondral or osteochondral fracture may occur in patients with an acutely ruptured knee ligament. Characteristic clinical lesions appear as chondral softening, fissuring, or overt chondral fracture and have been documented principally in the lateral compartment. These osteochondral lesions of the posterior-lateral aspect of the tibia and/or the anterior- lateral aspect of the lateral femoral condyle are seen in over 80% of clinical ACL tears (Speer et al., 1992). These data compare well with the regions of high contact pressure in 90 compression specimens of the current study. These lesions may be associated with an underlying bone bruise identified on MRI in patients with ruptures of the ACL, as biopsy specimens from these patients reveal degeneration and death of chondrocytes (cartilage cells) in this area (Fang et al., 2001; Johnson et al., 1998). The current study provided contact pressure data for one ACL injury mechanism. This information may be useful to better diagnose occult soft tissue, cartilage and subchondral bone damage when knowledge of the injury mechanism is provided. Conversely, if the injury mechanism is unknown, a specific pattern of secondary injury may be helpful in ruling out some joint loading scenarios. Patterns of bone contusion have been linked to certain mechanisms of ACL injury (Sanders et al., 2000). However, the current data support internal tibia and valgus femur rotation as the mechanism for pivot shift injuries (Matsumoto, 1990), not external tibia rotation as was documented previously (Sanders et al., 2000). The combined evidence of high contact pressures in the posterior tibial plateau of compressive specimens, and bone bruises seen in similar locations clinically may also be an indication of anterior drawer of the tibia during the injury. This effect has been previously noted by this laboratory during compression loading of unconstrained TF joints (J ayaraman et al., 2001). An “anterior neutral shift” of the tibia has also been seen during weightbearing (Friederich et al., 2001) that may be due to an inherent posterior slope of 10-15° in the tibial plateau (Li et al., 1998). There were a number of limitations in the current study that need to be addressed in the future. Insertion of pressure film may have altered the pressure profiles (Wu et al., 1998). This referenced study suggests that the recorded pressures from the pressure sensitive film may overestimate the actual joint contact pressures by as much as 14-28%, 91 depending on the level of contact force, the thickness of the film, the cartilage material properties, and the geometry of the joint. Hence, the potential errors associated with the pressures recorded with the pressure sensitive film used in the current study would be difficult to correct. However, a documentation of their locations and relative intensities during the various loading scenarios presented in this dissertation will be informative. Due to the inherent nature of the pressure film, only summations of the pressure over the entire experiment were documented. The large increase in contact area in the medial compartment between pre-failure and failure tests of compression experiments was an indication that some sliding artifact may have occurred after ACL failure. Another limitation of the study was that the average age of the specimens represents a small portion of the population most active in sports activities. The avulsion fractures of the ACL attachment to the tibia are not reflective of commonly treated injury in sports medicine clinics, but instead occur as a function of low BMD and are a consequence of using a more aged specimen population. There was a difference in the compressive load to cause ACL rupture versus avulsion, but no differences for the pressure or histology data. Finally, the repetitive nature of these experiments may also be a limitation. The consequence of microdamages accumulating in soft tissue structures, such as the ACL, and cartilage fissures or subchondral bone microcracks should be investigated by a single loading study at or near the failure loads established in the current study. In summary, the results of this study support the hypothesis that there is damage to articular cartilage overlying MRI detected bone bruises in patients with ACL tears. These osteochondral microdamages are due to high contact forces generated in the joint 92 during rupture of the ACL The location of these tissue injuries corresponds to the joint motion occurring at the time of injury. ACL injury patients may be at risk of osteochondral damage, especially if the mechanism of injury involves a high compressive loading component, such as during a jump landing. 93 REFERENCES Ahlback S. Osteoarthrosis of the knee. A radiographic investigation. Acta Radiologica Sup1277. 1968;277:1-61. Atkinson P, Newberry W, Atkinson T, Haut R. A method to increase the sensitive range of pressure sensitive film. J Biomech. 1998a;3 1 :855-859. Atkinson P, Walsh J, Haut R. The human patella: A comparison of three preparation methods. J Histotech 1998b;21:151-153. Banglmaier R, Dvoracek-Driksna D, Oniang’o T, Haut R. Axial compressive load response of the 90° flexed human tibiofemoral joint. 43rd Stapp Car Crash Conf Proc. 1999;43:127-139. Colombo C, Butler M, O’Byrne E, Hickman L, Swartzendruber D, Selwyn M, Steinetz B. A new model of osteoarthritis in rabbits. Development of knee joint pathology following lateral menisectomy and section of fibular collateral and sesamoid ligaments. Arth Rhuem. 1983;26:875-886. Davies-Tuck M, Wluka A, Wang Y, Teichtahl A, Jones G, Ding C, Cicuttini F. The natural history of cartilage defects in people with knee osteoarthritis. Osteoarthritis Cart. 2008;16:337-342. D’Lima DD, Hashimoto S, Chen PC, Colwell CW, Lotz MK. Human chondrocyte apoposis in response to mechanical injury. Osteoarthritis Cart. 2001;91712-719. Fang C, Johnson D, Leslie MP, Carlson CS, Robbins M, Cesare PE. Tissue distribution and measurement of cartilage oligomeric matrix protein in patients with magnetic resonance imaging-detected bone bruises afier acute anterior cruciate ligament tears. J Orthop Res. 2001;19:634-641. .Felson D. An update on the pathogenesis and epidemiology of osteoarthritis. Radiologic Clinics N Am. 2004;42:1-9. Fleming B, Renstom P, Beynnon B, Engstrom B, Peura GD, Badger GJ, Johnson RJ. The effect of weight-bearing and external loading on anterior cruciate ligament strain. J Biomech. 2001;34:163-170. Friederich N, O'Brien W. Gonarthrosis after injury of the anterior cruciate ligament. Unfallchir Versicherungsmed. 1993;86:81-89. Goldstein S, Wilson D, Sonstegard D, Matthews L. The mechanical properties of human tibial trabecular bone as a function of metaphyseal location. J Biomech. 1983;16:965-969. 94 Golenberg N, Kepich E, Haut R. Histomorphological and mechanical property correlations in rabbit tibial plateau cartilage based on a fibril-reinforced biphasic model. Int J Exp Comp Biomech. 2008; In Press. Grushko G, Schneiderrnan R, Maroudas A. Some biochemical and biophysical parameters for the study of the pathogenesis of osteoarthritis: A comparison between the processes of ageing and degeneration in human hip cartilage. Connective Tissue Res. 1989;19:149-176. Haut Donahue T, Hull M, Rashid M, Jacobs C. A finite element model of the human knee joint for the study of tibio-femoral contact. J Biomech Eng. 2002;124:273-280. Hurwitz D, Sumner D, Andriacchi T, Sugar D. Dynamic knee loads during gait predict proximal tibial bone distribution. J Biomech. 1998;31:423-430. Jayaraman V, Sevensma E, Kitagawa M, Haut R. Effects of anterior-posterior constraint on injury patterns in the human knee during tibiofemoral joint loading from axial forces through the tibia. 45th Stapp Car Crash Conf Proc. 2001;45:449-468. Johnson D, Urban W, Caborn D, Canarthos W, Carlson C. Articular cartilage changes seen with magnetic resonance imaging-detected bone bruises associated with acute anterior cruciate ligament rupture. Am J Sports Med. 1998;26:409-414. Kaplan P, Walker C, Kilcoyne R, Brown D, Tusek D, Dussault R. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: Assessment with MR imaging. Musculoskeletal Rad. 1992;183:835-838. Kaplan P, Gehl R, Dussault R, Anderson M, Diduch D. Bone contusions of the posterior lip of the medial tibial plateau (Countercoup injury) and associated internal derangements of the knee at MR imaging. Radiology. 1999;211:747-753. Li G, Rudy T, Allen C. Effect of combined axial compressive and anterior tibial loads on in situ forces in the anterior cruciate ligament: a porcine study. J Orthop Res. 1998;16:122-127. Li G, Lopez 0, Rubash H. Variability of a three-dimensional finite element model constructed using magnetic resonance images of a knee for joint contact stress analysis. J Biomech Eng. 2001;123:341-346. Matsumoto H. Mechanism of the pivot shift. J Bone Jt Surg. 1990;72-Bz816-821. Mazieres B, Blanckaert A, Thiechart M. Experimental post-contusice osteoarthritis of the knee: Quantitative microscope study of the patella and femoral condyles. J Rheum. 1987;14:119-121. Mink J, Deutsch A. Occult cartilage and bone injuries of the knee: Detection, classification, and assessment with MR imaging. Radiology. 1989;170:823-829. 95 Muhlbauer R, Lukasz S, Faber S, Stammberger T, Eckstein F. Comparison of knee joint cartilage in triathletes and physically inactive volunteers based on magnetic resonance imaging and three-dimensional analysis. Am J Sports Med. 2000;28:541- 546. Rangger C, Kathrein A, Freund M, Klestil T, Kreczy A. Bone bruise of the knee: Histology and cryosections in 5 cases. Acta Orthop Scand. 1998;69:291-294. Repo R, Finlay J. Survival of Articular Cartilage after Controlled Impact. J Bone Jt Surg. l977;59-A:1068-1076. Sanders T, Medynski M, Feller J, Lawhom K. Bone contusion patterns of the knee at MR imaging: Footprint of the mechanism of injury. RadioGraphics. 2000;20:8135-8151. Speer K, Spritzer C, Basset F, F eagin J, Garrett W. Osseous injury associated with acute tears of the anterior cruciate ligament. Am J Sports Med. 1992;20:382-389. Spindler K, Schils J, Bergfeld J, Andrish J, Weiker G, Anderson T, Piraino D, Richmond B, Medendorp S. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. Am J Sports Med. 1993;21:551-557. Tepper S, Hochberg M. Factor associated with hip osteoarthritis: data from the First National Health and Nutrition Examination Survey. Am J Epidemiol. 1993;137:1081-1088. Vellet A, Marks P, Fowler P, Munro T. Occult posttraumatic osteochondral lesions of the knee: Prevalence, classification, and short-term sequelae evaluated with MR imaging. Radiology. 1991;178:271-276. Wu J, Herzog W, Epstein M. Effects of inserting a Pressensor Film into articular joints on the actual contact mechanics. J Biomech Eng. 1998;120:655-659. 96 CHAPTER FOUR ANTERIOR CRUCIATE LIGAMENT INJURY FROM INTERNAL TIBIAL TORSION AND THE ASSOCIATED OSTEOCHONDRAL MICROTRAUMA IN REGIONS OF HIGH CONTACT PRESSURE ABSTRACT Clinical studies have proposed the mechanisms of ACL injury, based on patient questionnaires and video analysis of the events. The most commonly referenced loading mechanism is internal tibial rotation. The level of contact pressure developed in the human knee joint and the extent of articular cartilage and underlying subchondral bone injuries depends on the mechanism of applied loads/moments during rupture of the ACL. Seven knees, flexed to 30°, were loaded in internal tibial torsion until injury. Pressure sensitive film recorded the magnitude and location of contact. Histology and MRI were used to document microtrauma to the tibial plateau cartilage and subchondral bone. All specimens suffered ACL injury, either in the form of a midsubstance rupture or avulsion fracture at 3313 Nm of internal tibial torque. Failure occurred at 58il9° of internal rotation, and valgus rotation of the femur increased significantly after ACL injury. After loading, the articular cartilage in regions of the tibial plateau showed increased MRI signal intensity, corresponding to an increased susceptibility to absorb water. Histologically, there were fewer microcracks in the subchondral bone and less articular cartilage damage in torsion than compression experiments. Significant damage occurs to the articular cartilage and underlying subchondral bone during rupture of the ACL. The types and extent of these tissue injuries are a function of the mechanism of ACL rupture. 97 INTRODUCTION Clinical studies have proposed the mechanisms of ACL injury, based on patient questionnaires and video analysis of the events. The most commonly referenced loading mechanism is internal tibial rotation (Arnold et al., 1979; McNair et al., 1990; Boden et al., 2000; Olsen et al., 2004). In one study, athletes described internal tibial rotation as the injury mechanism in approximately 82% of cases (Arnold et al., 1979). In a more recent survey 38 out of 71 noncontact ACL injuries occurred while decelerating during or just before a change in direction (Boden et al., 2000). In a study of soccer ACL injuries, 56 out of 105 players were changing direction towards the side of their injured knee (Fauno and Wulff Jakobsen, 2004). Skiing also has one of the highest rates of ACL injuries, accounting for 25—30% of knee injuries (Speer et al., 1995). These injuries are mainly associated with internal twisting of the tibia relative to the femur or combined torque and compression during a hard landing (Ettlinger et al., 1995; Shoemaker etal., 1988). Biomechanical studies have also shown that internal tibial torque induces tensile forces in the ACL (Li et al., 1998; Markolf et al., 1995), especially when the knee is between full extension and 30° of flexion. More recently, knees loaded with an isolated 10 Nm internal torque produced an average ACL tensile force of 230 N (Hame et al., 2002). In another study, two knees at 30° of flexion were internally rotated 20°, resulting in the ACL providing 30-40% of the ligamentous restraining force (Seering et al., 1980). In subjects undergoing arthrOSCOpic surgery, but with normal gait and an intact ACL, a strain transducer has been implanted on the ACL and a variety of external loads were applied, including tibial torques of 10 Nm (Flemming et al., 2001). The study reports that 98 at 20° of knee flexion, internal torsion of the tibia increases ACL strain, while external torque does not. The rotational restraint capability of the AC L has been described using a computational model of the four main ligaments and articular contact of the knee (Blankevoort et al., 1996). Both the ACL and MCL were found to provide approximately 125 N of force when subjected to a 3 Nm internal tibial torque. Interestingly, the articular contact restraint force was twice the total ligamentous restraint force. Since the model had no applied axial compressive load, this articular contact must have been induced by tension from the knee ligaments due to the balance of forces. Few studies have documented ligamentous forces or relative joint displacements at failure levels under controlled loading of the knee joint. The goal of the current study was to induce ACL rupture in a 30° flexed knee joint by isolated internal tibial torsion. This type of loading will result in relative motion between the tibia and femur that strains the ACL. The objective of the current study was to document the levels of contact pressure developed in the human knee joint during rupture of the ACL via internal rotation of the tibia. Additionally, the study, using isolated human cadaver joints, was designed to investigate the associated injury to articular cartilage and underlying subchondral bone. Radiographic images have previously been used to relate bone marrow edema Patterns with injury mechanisms through their characteristic “footprint” (Sanders et al., 2000). The current study will provide experimental data to help validate the patterns of bone contusions documented in the clinical literature. 99 METHODS Torsion experiments were conducted on seven TF joints from cadavers aged 55.8i5.7 years (Table 4.1). The specimens the opposite limbs for TF compression experiments in Chapter 2 and were stored and prepared with similar methods. A custom, hydraulic, biaxial testing machine was built by mounting a 244 Nm rotary actuator (Model SS-001-1V, Micromatic, Beme IN) onto a linear actuator frame (Model 312.21, MTS Corp., Eden Prairie, MN) with a vertically oriented actuator (Model 204.52, MTS Corp.). The actuators had separate controllers (Model 458.2 Microconsole for the vertical actuator and Model 442 Controller for the rotary actuator, MTS Corp). The torque was programmed by a waveform generator (Model 458.91 Microprofiler, MTS Corp.) that generated a haversine waveform with a 125 ms time to peak. The potted tibia was attached to the rotary actuator through a biaxial (torsion-axial) load cell (Model 1261 Axial Torsion Load Cell, Interface, Scottsdale, AZ). The potted femur was attached to a similar fixture and X-Y translational table as in the compression experiments, with the main difference being that the whole device was attached to the rotation-locked, linear actuator (Figure 4.1). The joint flexion angle was also set to 30°, and for all but the first torsion specimen (32057R), the varus/valgus angle was left unconstrained. Compressive preloads were applied through the femur prior to the application of the internal torque on the tibia. In the first four specimens a 50 N preload was used and in the final three specimens a 1000 N preload was used to stabilize the joint. Repeated, increasing levels (10 Nm increments) of internal torque were applied to each specimen until catastrophic injury of the joint. Similar to the compression experiments, pressure sensitive film packets were inserted into the medial and lateral compartments of the TF joint to record 100 the contact area and pressure (Figure 4.2). For comparison purposes only the medium film contact areas and average pressures were reported, except when the medium film area was zero, and then the average pressure from the low pressure film was used. Additionally, the contact area over 25 MPa was computed to show the region of joint contact that was at the most risk of articular cartilage and underlying subchondral bone damage. Pressure film data from the experiment immediately prior to failure and after the failure experiment. Table 4.1 Torsion specimen information and test parameters. Specimen Age ex Height Weight Preload Rate (HS) ("1) (kg) (N) (HZ) 32057 R 50 M 1.67 64 200 4 32058 L 64 F 1.73 73 50 4 i . 32081 L 52 F 1.70 75 50 4 32087L 54 M 1.93 114 50 4 '1 | 32153 L 59 M 1.78 100 1000 0.5 | 32181 R 54 M 1.78 84 1000 0.5 i 32184 R 62 F 1.75 52 1000 0.5 i ‘ ‘Linear Actuator I} "‘1=ll i X-Y Platel r T ] 30° Flexion! Angle. L—lBiaxial Load Cell I 1 Rotary L____J Actuator Figure 4.1 Diagram of the torsion testing fixture. 101 Figure 4.2 Knee specimen attached to the torsion testing fixture with pressure film inserted into the medial and lateral compartments. Following the experimental testing, each knee was scanned using MRI, soaked for 24 hours in PBS and rescanned, as described in Chapter 3. Damaged or fibrillated cartilage has previously been shown to swell excessively when placed in physiological saline (Grushko et al., 1989). The average cartilage signal intensity was measured from sagittal slices using the region of interest selection tool of an open source DICOM viewer and analysis tool (OsiriX version 2.7.5, Open Source General Public License). Three sagittal slices were analyzed from both the medial and lateral tibial plateaus; the first slice was oriented towards the exterior of the joint where the meniscus covered the whole surface and the second and third slices were positioned consecutively more interior in the joint and had a center region that was not covered by meniscus. A single polygon region of interest comprising the tibial plateau cartilage was created, ending near the anterior and posterior sides before the edge of the meniscus (Figure 3.1). This polygon was then 102 subdivided into three sections; anterior, center, and posterior, depending if the cartilage was covered or uncovered by meniscus (or, for the first slice, the center was the thinnest region of meniscus). Additionally, the signal intensity was recorded from a circular region of interest in the relaxing solution. The signal intensity from each cartilage subdivision was scaled relative to the reference signal intensity of the relaxing solution from each slice. Finally, the percentage change in signal intensity was found by dividing the pre-soak and post-soaked difference in relative cartilage signal intensities by the pre- soaked scan intensity. All soft tissue injuries and bone fractures were documented during a careful gross dissection of the joint, while separating the tibia, removing the menisci, and highlighting fissures with India ink. Histological slides were prepared as described in Chapter 3. The integrity of the cartilage and subchondral bone was documented based on an established scoring system. Specific microdamages of interest were horizontal and vertical microcracks along the cartilage-subchondral bone interface, superficial and deep zone cartilage damage, and cartilage compression lines along the watermark. Statistical t-tests were used to test for significant differences (p<0.05). Two-way repeated measure ANOVAs were used to compare histological damage with the factors of experimental loading and location. Comparisons were made using paired specimens between the internal tibial torsion specimens and the opposite limbs used in Chapters 2 and 3. RESULTS The average number of torsion tests performed on each specimen prior to gross failure of the knee joint was 5 (range 1-8). In the pre-failure tests the peak valgus rotation had a mean value of 1116.0°, and in most specimens it was closely related to the level of 103 applied internal rotation (Table 4.2). Lateral motion of the femur was 5.4:t5.l mm, and it was closely related to the degree of valgus rotation. The peak posterior femur displacement had a mean value of 9.0133 mm, and this generally occurred early during the rotation. In some specimens there was a decrease later in the test, as valgus rotation and lateral motion peaked. Table 4.2 Data from internal tibial torsion pre-failure experiments on each specimen. ML indicates medial (+) and lateral (-), PA indicates posterior (+) and anterior (-). W indicates valgus (+) and varus (-). ‘\ Q s a“ x 6“ s (000 Q8193) &0&®\®6‘\ gig” «062$ «figs (($369 (9900 69‘: 00 3'50 0‘6? @Qi-«O‘Q 6S3§¢0 ”‘QYeCP‘Q *§@g\°° Q 3 Q ’3 «8°qu Q0699 Qog-eR‘ Q00 99 339,528.10.” 4. 59.2 ., 50 5:41.40? - 910259912 32053 0.15. L 24.? -._--29 2-5-1L--- 11.1 61L ._§_29§1£:_0-_18_._30-1 ,.,,-6z_ .L3m_‘1-2 _ ,___29._7_ _32087L - 0:191 g 46.8 , 62 6.7 9.2 7.6 32153L 0.24 23 M, “.31 _ 1.7 8 14 32181R 1 31.1 32 -o.3 12.7 7.7 fl(SD) 36(14) 46416) 8.4(3Q 6.3(5.1) 11(6.0) Failure tests showed a 13° increase in internal rotation to a maximum value of 58i19° (Table 4.3). The peak torque in the failure test was higher than in the pre-failure test (Figure 4.3). At the point of failure there was a sharp drop in the torque (Figure 4.4). There was also a significant increase between the failure and pre-failure tests for the peak valgus rotation of 20i5.7°, but only slight increases in the peak lateral femur displacement of 11324.2 mm and the peak posterior femur displacement of 10:I:4.I mm. All torsional specimen failures involved the ACL, but there were more frequent incidences of injury to other structures than in compression tests. Three of six specimens suffered posterior-lateral bundle midsubstance ACL rupture near the femoral insertion, while the remaining joints suffered ACL avulsions (Table 4.4). The peak internal torque 104 for ACL ruptures was 353:5 Nm, while the peak torque for avulsion fractures tended to be slightly lower at 30i19 Nm. The posterior femur displacement was slightly different between these two types of joint failure from 8.0:Iz4.4 mm for midsubstance ruptures to 13i2.4 mm for avulsion failures. 40 - Pre-failure 35 « —-Fai|ure Time (see) Figure 4.3 Representative (32081 L) torque versus time plots for pre-failure and failure tests. - - - - Posterior Femur Displacement (mm) ---—'--- Axial Tibia Rotation (deg) 90 . — - Lateral Femur Displacement (mm) -— Internal Tibia Torque (Nm) Valgus Femur Rotation (deg) .43....” f _ 7o - ' E 60- _ 5 50 - Failure 4o - '1 : " j 409 50 " ' ' ' ' 100 150 200 250 Time (msec) Figure 4.4 A representative torque and motion versus time plot during a failure test (32081 L). The peak internal torques and the corresponding axial tibia rotation occurred at the failure time point as marked. All other motions were measured at their peak values, indicated by #. 105 Table 4.3 Data from internal torsion failure experiments on each specimen. #Significant difference between pre-failure and failure tests. ML indicates medial (+) and lateral (-), PA indicates posterior (+) and anterior (-), W indicates valgus (+) and varus (-). \, \ \ . 0 0&0 Q0#&C’\ é°®$<® “0000\609 Qo‘oifioé‘ QO&:\& QQCO‘QOQ \ 0° ‘9 03> \& oak Q5? 0‘6? \X‘\’ <80 QT (90 3‘ 9800 «o 66‘ 69 00‘ 06b 0‘0 \‘b ~15 6* 0C? 6* 0c?" ‘5" 6"p (.990 ‘\\ & Q ’\ «\‘OQQQ Q6 (98 <20 <93 0 Q- 0 O 32057R 0.17 64.6 60 6.7 14.3 0 (locked) 7432058L 0.15 31.5 38 12.9 11.5 9.8 32081L 0.17 36.9 75 13.4 5 22.3 . 32087L 0.18 49.2 85 (“13.2 12.2 18.1” 32153L 1.1 29.7 60 6.5 6 g 22 - 32181R 0.87 39.2 39 7 13 _ 13 F” 23.5 32184R 1.1 10.5 49 4.1 15.9 25.9 Alg (SD) 37 (17)# 58 (18)# 10 (4.1) 10 (4.1) 20 (5.7)# Table 4.4 Dissection documentation of soft tissue injury following failure tests. I Specimen MCL ACL PCL LCL Menlsci , 3205., R Complete tear near Postenor-lateralnear lnta ct htact Normal tibial insertion femoral insertion 32058 L intact Aw‘smgggcwe at intact htact Normal , Complete tear near Posterior-lateral near ‘ Posterior-medial I #32081 L tibial insertion femoral insertion Intact Intact detached I 32087 L Partial tear off Avulsron fracture at Intact Avulsmn fracture Normal femur tibia atfibula I - _ : 32153 L intact POStem' famine“ intact intact Normal I femoral insertion : 32181 R intact P°Ste"°"'f""e'a'.”ear intact intact Normal t_ femoral insertion i 32184R Intact Avul5ion fracture at Avusion from Avul5ion fracture Lateral detached L . tibia tibia at tibia Three of the torsion specimens were tested at 4 Hz with compressive preloads of 50 N, while the other three specimens were tested at 0.5 Hz with 1000 N of preload. Although the small group sizes for these test conditions limited the statistical power, there was significantly less medial femur displacement in pre-failure tests and a slight 106 trend for more valgus femur rotation and less medial femur displacement in failure tests for the lower speed, more pre-loaded specimens. For subfailure (Table 4.5) and failure (Table 4.6) tests, there were no significant differences in the average contact area or pressure between the lateral and medial compartments. During subfailure tests, the maximum pressure in the lateral compartment was 47% higher for compression than for torsion experiments. During failure tests, the maximum pressures showed even larger differences between compression and torsion experiments with 26% and 52% increases in the medial and lateral compartments, respectively. This difference was statistically significant in the lateral compartment. Additionally, the contact areas in both compartments were significantly higher for compression versus torsion experiments. In the medial compartment, the pressure distribution varied in location between compression and torsion experiments, with the region over 25 MPa (Figure 4.5) occurring posterior for compression and anterior for torsion. This variation was not as pronounced in the lateral compartment, with the location of highest pressures occurring posterior. The pressure distributions were also similar between the subfailure and failure tests. .\nlcrior 3 \ll’zl 35 \ll’n Posterior Figure 4.5 Representative shape and magnitude of the pressure distribution on the tibial plateau of failure experiments (specimen 32181). 107 Table 4.5 Pressure film data from subfailure experiments. *Significant difference from paired compression experiments in Chapter 3. _ __ Medial Compartment Lateral Compartment 1 Area > Av ; Max ‘ Area > Av i Max Area I 25 MPa Pressire ' Pressure Area ! 25 MPa Preseire , Pressure , _— ____ ( mmi)" “ "T (Mear' _ (Mm—=17" ” -"w(_MP_a) 132951. __ 183...: ...?L 26- __ 50____ __ __9 |_-__0_-__ ____0__ 1-310-... ngqgafiw 85 1 "“12 27 _._9 _i _ 0.. _ 6 i _ 10 i339€fl_._-_9___ 0 6 .10_ _ ___9 _ __ 0__ ._ 6 1 10.1 @1231 .____49 __ 2 __14 -22, _§9__ 0 _ .--..12 :_-__29__ $59330 __ 0 1_6 .19 $3.6 __ 0 _ .12 39_ 32181 133___5__ __ ..1__ _ _14 _33Z_ 113 g 13 24 32184 0 0 8 l 10 70 0 14 24 Av SD 45 (56) 1(1) 10(4) 1 20 (12) 56(46) 0(0) 11(3) 18(6)* Table 4.6 Pressure film data from failure experiments. *Significant difference from paired compression experiments in Chapter 3. _fi _ Medial Compartment Lateral Compartment . Area > Av | Max Area > Av Max I Area 25 MPa Pressire iPreseure Area 25 MPa Prague Pressure ‘ (mmfiw fl”: (MPa) (mm') (MPa) 32057 245_i__ _78__” _25 _ 1.3503 0 J 0 0 0 32058 117 V __3, .15 ' 313 _0 L- __o. __ _§_ .._1_0____ 322a; ’ o ’ 6 103 -39 = ___0___ 6_ 19___ 32087 80 _ - 5 _,. -17 34_____ __9___-___o____6__1___12__. 32153 33 __ _4 _16 32“ ”_61 __ 0__ _15“ ”24 32181 107 1__-3-_i- _13__ _3_1__-+*_9_5 0 12 20 32184 0 ' 0 8 10 67 0 13 23 Av SD 56(52)* 3(2) 12(5) ;25 (11) 37 (42)* 0(0) 10 (4) 16(7)* The relative MRI signal intensity from the cartilage increased from the pre to post-soaked scans in all specimens, with average increases of 15% to 30% depending on the region. Significant differences between the relative signal intensity increases were documented between compression and torsion specimens in the center, and posterior regions of the medial tibial plateau (Figure 4.6). In the lateral compartment, there were similar amounts of increase in signal intensity across the articular surface for both types of loading. 108 60% - (J1 ‘3 o\ L 40% 1 30°/o - 20°/o - Change in Relative Signal _3 O. s: o\ o\ I I Anterior Central Posterior Anterior Central Posterior Medial Compartment Lateral Compartment Figure 4.6 The relative increase in signal intensity of the tibial plateau cartilage between pre-soak and post-soak MRI scans. *Significant difference from paired compression experiments in Chapter 3. Analysis of the histological samples revealed occult injuries for all specimens following torsion experiments (Figure 4.7). There was significantly more damage in the anterior medial region and slightly more in the posterior lateral region. In addition to horizontal microcracks in the lateral compartment, the torsion specimens also had cartilage damage in the central lateral and central and anterior medial regions (Figure 4.8). The damage in the torsion specimens was not as extensive as in the paired specimens from the series 1 compression experiments in Chapter 3. In fact, there was significantly more damage in the posterior regions of both facets for TF compression experiments, but more damage in the anterior medial region for torsion experiments. 109 Anterior . Central . \ Lateral Medlal Figure 4.7 Average 1 SD histological scores for regions within the medial and lateral facets. *Signiflcant differences from the anterior and central regions. #Significant difference from the central lateral region and anterior medial region. Figure 4.8 Gross surface damages after staining with India ink. DISCUSSION Although commonly suggested as the injury mechanism in clinical ACL ruptures, the current study is the first to consistently produce ACL injuries via excessive internal tibia rotation using cadaver knee joints (Figure 4.9). Kennedy et al. (1974) tested five cadaver knees in internal rotation (at 15-20° of flexion) and described the ACL becoming very taut, but they generated premature bone fracture at the clamps. The ACL resists internal rotation by its orientation in the axial plane, where it attaches slightly medial on the anterior tibial plateau and slightly lateral in the femoral notch (Amis and Dawkins, 1991; Amoczky, 1982). An important factor in predicting the tension in the ACL is the 110 location of the axis of tibia rotation in the frontal plane. In the current study the relative axis of rotation between the tibia and femur was not constrained by the experimental fixture. Since posterior displacement of the femur and posterior displacement of the medial tibial plateau (induced by the internal tibia rotation, Figure 4.10) occurred simultaneously, we concluded that the effective center of rotation was located medial of the ACL (Figure 4.11). In addition, these motions occurred on an inherent posteriorly sloped tibia and produced a coupled motion between internal rotation of the tibia and valgus rotation of the femur (Matsumoto et al., 2001). Coupled internal tibial and valgus femur rotations have been previously documented for a 12.5 Nm val gus moment (Matsumoto, 1990). During those experiments, the axis of rotation was located near the MCL due to tension in that ligament as the primary restraint for valgus bending moments. In the current study, valgus rotation of the femur may have also played a role in creating ACL tension and was significantly increased in magnitude after ACL failure. On the other hand, even with approximately 20° of valgus rotation of the femur, only two specimens had MCL damage. Initial Position ,, Torsion ACL failure “- position Figure 4.9 Sagittal view, for a fixed tibia, of the relative TF joint motion during torsion experiments, resulting in ACL failure. 11] Medial Compartment Lateral Compartment Posterior Posterior Drawer Drawer High Slope Anteriorly I Internal Rotation High Slope Posteriorly Internal Rotation 1. Figure 4.10 Sagittal view of the relative TF joint motion in the medial and lateral compartments of the knee during torsion experiments. Valgus Rotation of the Femur PCL ACL 0 3 ..I a MCL ". ' : LCL Tension'. ' E Medlal / Axis of . Ratatlon Lateral : ‘ Anterior 3 ‘ Translation Tibial Internal Rotation Figure 4.11 Coupled internal rotation of the tibia and valgus rotation of the femur. 112 These experiments produced relatively smaller and less severe contact regions posteriorly on the lateral plateau and anteriorly on the medial plateau than were seen for TF compression. The contact pressures in this case were indirectly caused by a balance of forces due to high tensile forces generated in the AC L and other ligamentous structures of the knee. Osteochondral microdamages were also documented in torsion specimens. These microdamages were usually in the form of horizontal microcracks on the lateral tibial plateau and could be due to excessive shear stress. Since TF compression did not result in as many horizontal microcracks, this suggests that alternate loading conditions may have influenced the type of microdamage generated in the subchondral bone. An additional factor for the location of subchondral and cartilage damage may be the intrinsic structural and material properties of the tissues. Contact in the posterior lateral compartment of torsion specimens occurred as the lateral femoral condyle was displaced towards the posterior-center of the tibial plateau. The central area of the tibial plateau has material properties that can be orders of magnitude lower than the lateral plateau (Goldstein et al., 1983), and this infrequent location for joint contact may have produced microdamage at a relatively low level of force in the current study. In the current study, MRI was used to measure the relative fluid uptake of the cartilage after experimental testing and soaking for 24h in PBS. This provided an indirect method of evaluating the amount of functional matrix damage to the cartilage (Grushko et al., 1989). While the cartilage from all specimens underwent some increase in signal, in the center and posterior medial plateau, the cartilage from the torsion specimens had significantly less fluid uptake than the compression specimens. This occurred in a region 113 of the cartilage where the TF compression specimens developed high contact pressures, but the torsion specimen had very low or negligible contact pressures. In one specimen where varus-valgus rotation was prevented (32058R), there were high contact pressures developed on the medial plateau, and the lateral aspect of the knee joint separated with negligible contact pressure. In the subsequent, unconstrained specimens valgus rotation occurred and the contact pressures were distributed evenly between the medial and lateral compartments. An important aspect of the current study was that the center of rotation was not constrained by the experimental fixture. Instead the applied torque produced forces in the knee joint ligaments as they might occur under normal circumstances, and relative displacement between the tibia and femur appeared to be a complex series of translations and rotations that need to be documented in future studies. The current study provided contact pressure data for the internal tibial torsion ACL injury mechanism, as well as secondary injury information associated with that particular mechanism. This information may be useful to better diagnose occult soft tissue, cartilage and subchondral bone damage when knowledge of the injury mechanism is provided. Conversely, if the injury mechanism is unknown, 8 specific pattern of secondary injury may be helpful in ruling out some joint loading scenarios. Patterns of bone contusion have been linked to certain mechanisms of ACL injury (Sanders et al., 2000). However, the current data support internal tibia and valgus femur rotations as the mechanism for pivot shift injuries (Matsumoto, 1990), not external tibia rotation as was documented previously (Sanders et al., 2000). 114 In summary, the results of this study suggest that damage to articular cartilage overlying MRI detected bone bruises in patients with ACL tears may be due to high contact forces generated in the joint during rupture of the ACL, that may form the basis for the subsequent development of OA in the ACL-injured patient (Fang et al., 2001). ACL rupture occurred in isolated human knee joints via excessive internal tibial torsion in this study. The peak torques were in the range that could regularly be generated during SRE activities. However, there were fewer microcracks in the subchondral bone and less articular cartilage damage in torsion than compression experiments. Therefore, the damage to the articular cartilage and underlying subchondral bone during rupture of the ACL are a function of the mechanism of ACL rupture. 115 REFERENCES Amis A, Dawkins G. Functional anatomy of the anterior cruciate ligament: fibre bundle actions related to ligament replacements and injuries. J Bone Jt Surg. 1991;73:260- 267. Arnoczky S. Anatomy of the anterior cruciate ligament. Clin Orthop Rel Res. 1923;172:19-25. Arnold J, Coker T, Heaton L, Park J, Harris W. Natural History of Anterior Cruciate Tears. Am J Sports Med. 1979;72305-313. Blankevoort L, Huskies R. A mechanism for rotational restraints in the knee joint. J Orthop Res. 1996;14:676-679. Boden B, Dean G, Feagin J, Garrett W. Mechanisms of Anterior Cruciate Ligament Injury. Orthop. 2000;23:573-578. Ettlinger C, Johnson R, Shealy J. A method to help reduce the risk of serious knee sprains incurred in alpine skiing. Am J Sports Med. 1995;23:531-537. Fauno P, Wulff Jakobsen B. Mechanism of Anterior Cruciate Ligament Injuries in Soccer. Int J Sports Med. 2004;27:75-79. Fleming B, Renstom P, Beynnon B, Engstrom B, Peura GD, Badger GJ, Johnson RJ. The effect of weight-bearing and external loading on anterior cruciate ligament strain. J Biomech. 2001;34:163-170. Goldstein S, Wilson D, Sonstegard D, Matthews L. The mechanical properties of human tibial trabecular bone as a function of metaphyseal location. J Biomech. 1983;16:965-969. Grushko G, Schneiderman R, Maroudas A. Some biochemical and biophysical parameters for the study of the pathogenesis of osteoarthritis: A comparison between the processes of ageing and degeneration in human hip cartilage. Connective Tissue Res. 1989;19:149-176. Hame S, Oakes D, Markolf K. Injury to the Anterior cruciate ligament during alpine skiing; A biomechanical analysis of tibial torque and knee flexion angle. Am J Sports Med. 2002;30:537-540. Kennedy J, Weinberg H, Wilson A. The anatomy and function of the anterior cruciate ligament: As determined by clinical and morphological studies. J Bone Jt Surg. 1974;56-A2223-235. Li G, Rudy T, Allen C. Effect of combined axial compressive and anterior tibial loads on in situ forces in the anterior cruciate ligament: a porcine study. J Orthop Res. 1998;16:122-127. 116 Markolf K, Burchfield D, Shapiro M, Shepard M, F inennan G, Slauterbeck J. Combined knee loading states that generate high anterior cruciate ligament forces. J Orthop Res. 1995;13:930-935. Matsumoto H. Mechanism ofthe pivot shift. J Bone Jt Surg. 1990;72—Bz816-821. Matsumoto H, Suda Y, Otani T, Niki Y, Seedhom B, F ujikawa K. Roles of the anterior cruciate ligament in preventing valgus instability. J Orthop Sci. 2001;6228-32. McNair P, Marshall R, Matheson J. Important features associated with acute anterior cruciate ligament injury. N Zealand Med J. 1990;103:537-539. Olsen 0, Myklebust G, Engebretsen L, Bahr R. Injury mechanism for anterior cruciate ligament injuries in team handball: A systematic video analysis. Am J Sports Med. 2004;32:1002-1012. Sanders T, Medynski M, Feller J, Lawhom K. Bone contusion patterns of the knee at MR imaging: Footprint of the mechanism of injury. RadioGraphics. 2000;20:Sl35-S151. Seering W, Piziali R, Nagel D, Schurman D. The Function of the Primary Ligaments of the Knee in Varus-Valgus and Axial Rotation. J Biomech. 1980;13:785—794. Shoemaker S, Markolf K, Dorey F, Zager S, Namba R. Tibial torque generation in a flexed weight-bearing stance. Clin Orthop Relat Res. 1988;228:164-70. Speer K, Spritzer C, Basset F, F eagin J, Garrett W. Osseous injury associated with acute tears of the anterior cruciate ligament. Am J Sports Med. 1992;20:382-389. 117 CHAPTER FIVE ANTERIOR CRUCIATE LIGAMENT INJURY INDUCED BY HYPEREXTENSION AND THE ASSOCIATED OSTEOCHONDRAL MICROTRAUMA IN REGIONS OF HIGH CONTACT PRESSURE ABSTRACT Knee hyperextension has been described as a clinical mechanism of isolated ACL tears and is associated with a specific “footprint” of bone bruising on MRI scans. The hypothesis of the current study was that isolated hyperextension moments that induce ACL rupture in human cadaver knees will create regions of contact pressures which exceed a critical threshold and induce osteochondral microtrauma in the TF joint. Six human knee pairs were loaded in hyperextension until gross injury and the moments, rotations and displacements at the knee joint were recorded. Pressure sensitive film documented the magnitude and location of TF contact. Microtrauma in the tibial plateau cartilage and subchondral bone was quantified from histological sections. All failures involved the ACL, although frequently combined with other soft tissue injuries. The peak bending moment at failure was 108i46 Nm and more extension occurred in the femur than the tibia, which created anterior subluxation of the tibia. Hyperextension produced regions of contact on the medial and lateral, anterior tibial plateau with peak pressures of approximately 34 MPa and 24 MPa, respectively. Osteochondral microtrauma was increased in those regions of high pressure. Excessive TF compressive forces generated in the joint during a severe knee hyperextension incident, even without ACL rupture like in the pre-failure tests of the current study, may put athletes at risk of associated injuries to articular cartilage and underlying subchondral bone. 118 INTRODUCTION Isolated ACL injury occurs due to hyperextension approximately 6-25% of the time in sports (Arendt and Dick, 1995; Boden et al., 2000; Lohmander et al., 2004; Olsen et al., 2004). Anterior dislocation of the knee (bicruciate injury) is also caused by hyperextension but occurs much less frequently than isolated ACL rupture (Brautigan and Johnson, 2000; Heinrichs, 2004; Meyers and Harvey, 1971; Shelboume and Klootwyk, 2000). A few case studies have described the scenario involved in isolated ACL and bicruciate injuries from hyperextension, including noncontact jumping on a trampoline (Kwolek et al., 1998) and player to player contact during football (Wang et al., 1975). Biomechanical studies have measured the tension in the ligaments of the knee for low hyperextension moments and show significantly more tension in the ACL than the PCL at 10 Nm (Washer et al., 1993). Histological microtrauma of the subchondral bone and cartilage damage have been documented in regions of high contact pressure during isolated rupture of the ACL in cadaver (Meyer et al., 2008) and porcine (Yeow et al., 2008) knees. These experiments have demonstrated significant differences in the magnitude and distribution of joint contact pressures between two types of joint loading, namely tibia rotation and TF joint compression. Similarly bone bruise patterns have been identified as “footprints” of the mechanism of ACL injury in clinical studies (Kaplan et al., 1992; Sanders et al., 2000; Shelboume and Klootwyk, 2000; Terzidis et al., 2004). Few studies have documented the joint contact force or relative joint displacements at failure levels under controlled loading of the knee joint. The objective of the current study was to load knee joints with a hyperextension bending moment until 119 ACL rupture and assess the osteochondral microtrauma that occurs in regions of high TF contact pressure. A hypothesis of the study was that significant joint contact pressure would be generated in the anterior regions on the tibial plateau during failure experiments. These pressure distributions from hyperextension would have similar magnitudes as previous compression experiments and therefore also induce osteochondral microtrauma, but their location would be different. These data will help to experimentally verify the mechanism associated with the specific pattern of bone bruises documented in the clinical literature for hyperextension knee ligament injuries. METHODS Hyperextension experiments were conducted on TF joints from six male cadavers (aged 53i12 years) (Table 5.1). The joints were stored and prepared with similar methods as the previous studies described in Chapter 2. Hyperextension was applied via four-point bending, with the moment applied to the potting cups (Figure 5.1). The femoral cup was mounted on an XY translational table to allow medial/lateral (ML) and proximal/distal (PD) motion. These motions were recorded with linear encoders (Model #XOOZOIA, Renishaw, Hofinan Estates IL). In addition, varus/valgus (VV) angular rotation of the femur was unconstrained and recorded with a rotary encoder (Model #RCH25D-6000, Renco Encoders Inc., Goleta CA), while axial tibia rotation was fixed in its neutral position. A hydraulic materials testing machine with a linear actuator (Models 312.21 and 204.52, MTS Corp., Eden Prairie, MN) was used to apply the bending moment with a time to peak load of 250 ms. Extension was applied to the joint through a four-point bending moment arm via repeated, increasing magnitudes of actuator displacement (in 5 mm increments) until gross injury of the joint. Motion of the femur and tibia was also 120 measured during hyperextension using a Vicon motion capture system (Oxford Metrics Ltd., Oxford, United Kingdom). Reflective markers were attached to the femur and tibia fixtures, and four marker arrays were screwed into each bone. For the current study only femur and tibia hyperextension angle was analyzed with the motion analysis system. The peak anterior/posterior translation of the femur relative to the tibia was calculated from the difference between the angles of rotation and the length of each bone. Table 5.1 Hyperextension specimen information and test parameters. Specimen 83:) Sex Hight W(:igg)ht ‘ [23:28) 32416 R 34 M 1.83 68 2.5 32284 L 53 M 1.83 111 ' 2.5 32273 R 54 M , 1.75 78 2.5 32302 R 59 1 M 1.78 i 86 2.5 32516 L 47 i M 1.47 i 113 T 2.5 32388 R N/A M N/A . N/A 1 2.5 Linear Actuator l 4-Point Bending Load Cell / Moment Arm 5] Tibia o ‘ :IA Varus-Valgus Free 7’ Rotation X-Y Plate r I Figure 5.1 Diagram of the hyperextension testing fixture. 121 During each test, pressure film was inserted into the medial and lateral compartments of the TF joint, using the same method as described in Chapter 3 (Figure 5.2). For comparison purposes only the medium film contact areas and average pressures were reported. Additionally, the area over 25 MPa was computed to show the region of joint contact that was at the most risk of microdamage. Pressure film data from the test immediately prior to failure and during the failure test were compared. 31" 4 4:8 -: m1 ‘ - 7‘ "n. - ‘ ' Figure 5.2 Knee specimen attached to the hyperextension testing fixture. .~ ~¥. ~14»- ,. While separating the tibia and removing the menisci, all soft tissue injuries and bone fiactures were documented during a careful gross dissection of the joint. Surface fissures on the articular surfaces were highlighted for photographic documentation by wiping India ink across the surfaces. Histological sections were prepared from blocks 122 taken across the lateral and medial facets of the tibial plateaus. The samples were processed using the same procedure as described in Chapter 3. Each specimen was divided into medial and lateral compartments and by thirds across the anterior, central and posterior regions of each facet. The integrity of the cartilage and subchondral bone was documented with an established scoring system. Specific microdamages of interest were horizontal and vertical microcracks along the cartilage-subchondral bone interface, superficial and deep zone cartilage damage, and cartilage compression lines along the tidemark. The peak hyperextension moment and motions of the tibia and femur were documented for the pre-failure test and from the test when failure occurred. Relative displacements between the tibia and femur were measured at the time corresponding to the peak moment. During failure tests the displacements may have increased after the load was released due to gross failure, particularly the posterior displacement of the femur. One way, repeated measure ANOVAs were used to compare peak moments and the corresponding displacements between the pre-failure and failure tests, except for the femur and tibia extension angles which used a two-way, repeated measure ANOVA with factors of bone and test. SNK-post hoc tests were used where appropriate for multiple comparisons, and significance was indicated for p<0.05. RESULTS All failures involved the ACL, although there were frequently other ligamentous and/or capsular damage, including five of the six specimens which had a partial tear of the posterior cruciate ligament (PCL) (Table 5.2). The hyperextension moment was 108:l:46 Nm at failure with a total extension angle of 34:1:11°. The failure tests had 123 significantly higher moments than the pre-failure tests, and they exhibited a sudden drop in the force versus time plot (Figure 5.3) and a “popping” sound at the time of injury. There was an unequal sharing of the hyperextension rotation between the femur and tibia (Figure 5.4), which produced an anterior subluxation of the tibia in all specimens (Figure 5.5). There was minimal lateral or valgus motion of the femur in any test. Table 5.2 Peak force during pre-failure and failure tests and knee joint injury documented during dissection. Posterior-lateral bundle (PLB), Anterior-medial bundle (AMB), Posterior-medial bundle (PMB), Anterior-lateral bundle (ALB), Complete tear (X), Partial tear (l), Avulsion (/\). #Significant difference between prefailure and failure tests. . Peak Moment (Nm) ACL PCL SpeClmen —-------- ~ _ - 4 Pre-fallurei Failure PLB AMB PMB ALB 32416R 172 195 X I I 32284L 84 101 l I 32273R 101 119 A (T) I 32302R 73 80 I I 32516L__ 63 70 I\ (F) 32388R 80 83 I I Avg (so) 96 (40) 108(46)#, ‘40 - Failure A - P -f 1 £120 re aiure E 100 . § 2 80 7 S '3 60 — i 2 4° ‘ 0 g 20 - l 0 4 ~ ~ 4 - _ 0 0.1 0.2 0.3 0.4 0.5 Time (see) Figure 5.3 Representative (32273) bending moment versus time plot for pre- failure and failure tests. 124 El Pre-failure # I Failure 01 O r .h 0 r N O r _a. O 1 Extension Angle (deg) 0- Femur Tibia Total Figure 5.4 Rotation of the tibia and femur and total knee hyperextension in pre- failure and failure tests. #Significant difference between pre-failure and failure tests. *Significant difference between femur and tibia rotations. Hyperextension experiments generated slightly higher contact area (medium film only) in the medial compartment than in the lateral compartment (Figure 5.6). The location of the contact area was anterior with respect to the tibial plateau due to the combined rolling and sliding of the femoral condyles during hyperextension. Additionally, both the average contact pressure and maximum pressure were higher in the medial and lateral compartment (Tables 5.3 and 5.4). 12- El Pre-failure 10 - . I Failure 3 .4 6 .. 4 . 2‘ Ti 1 i 0 . , , Posterior Displacement Lateral Displacement Valgus Rotation (deg) (mm) (mm) Figure 5.5 Joint displacements for pre-failure and failure tests. 125 Lateral Figure 5.6 Representative contact pressure distributions (from low and medium pressure films combined) for failure experiments. Medial \nlerior 3 \ll’n 35 \I I’ll Posterior Table 5.3 Pressure film data, average (SD), from pre-failure tests. *Significant difference between medial and lateral compartments. Medial Compartment Lateral Compartment Area Arell:P>a25 Pressure Prrsasxlre Area Aresgazs Pressure PrzeZIlre (mm’) (MPa) (mm’) (MPa) 32416R 65 9 19 37 45 0 12 19 32284L 47 0 13 23 88 0 15 22 32273R 101 0 17 29 26 0 12 19 32302R 71 2 17 29 66 9 19 35 32516L 35 6 19 41 65 0 14 21 32388R 119 0 13 23 1 0 12 15 we) 73 (32) 3(4) 16 (3)* 30 (7)* 48(31) 1(3) 14(3) 22 (7) Table 5.4 Pressure film data, average (SD), from failure tests. *Significant difference between medial and lateral compartments. Medial Compartment Lateral Compartment Area Area > Pressure Max Area Area > 25 Pressure Max 25 MPa Pressure MPa Pressure (mm’) (MPa) (mm’) (MPa) 32416R 80 28 22 44 106 0 15 24 32284L 54 0 17 25 127 0 16 25 322738 155 49 21 38 48 0 12 18 323028 79 9 19 34 68 3 17 37 32516L 44 7 18 40 68 0 14 23 32388R 130 0 13 23 7 0 12 17 M (SD) 90 (44) 15 (19) 18 (3)* 34 (8)1‘ 71 fiZL 1 (1) 14 (2) 24 UL 126 Analysis of the histological samples revealed occult injuries for all specimens following torsion experiments (Figure 5.7). Microdamage varied central to anterior across each facet. In the lateral facet there were no differences between the microdamage score documented in the anterior, central, and posterior regions. In the medial facet there was significantly more microdamage recorded in the central region than in the anterior and posterior regions. The anterior region also had significantly more microdamage than the posterior region. Hyperextension specimens also had cartilage damage in the central and anterior regions on both facets (Figure 5.8). The damage in the hyperextension specimens was as severe as in the paired specimens from the series 2 compression experiments in Chapter 3. In fact, there was significantly more damage in the anterior regions of both facets for hyperextension experiments, but less damage in the posterior regions than TF compression experiments. Figure 5.7 Average :1: SD histological scores for regions within the medial and lateral facets. *Significant difference between medial and lateral facets. #Significant difference between anterior, central and posterior regions. 127 Posterior Figure 5.8 Gross surface damages after staining with India ink. DISCUSSION The hyperextension mechanism of ACL rupture has been associated clinically with bone bruises in the anterior aspect of the tibial plateau and anterior aspect of the femoral condyle (Sanders et al., 2000; Viskontas et al., 2008). This clinical “footprint” compares well to the current results with contact pressure and osteochondral microdamages in the central and anterior tibial plateau. An important finding of this study was that the hyperextension moments needed to produce ACL rupture generated similar magnitudes of contact pressure to those produced during ACL rupture under TF compressive force. The increase in average contact area, average contact pressure, and maximum pressure in pre-failure to failure experiments was relatively low, suggesting that contact pressures needed to create bone bruises may be present at pre-failure levels as well. In fact, bone bruises can occur in clinical patients during severe knee hyperextension resulting in either bicruciate dislocations (Crotty et al., 1998), isolated ACL rupture (Remer et al., 1992), isolated PCL rupture (V anhoenacker and Snoeckx, 2007), or without any associated ligamentous damage (Wright et al., 2000). This is in contrast to the pattern of contact pressures documented in pure compression specimens 128 where there is a much larger increase in pressure level for pre-failure tests than tests that produce ACL failure (Meyer et al., 2008). The knee hyperextension moments to cause gross failure in the human knee under four point bending have not been previously documented. Of the four experimental hyperextension studies conducted to date by others, the most similar results to the current study were produced by Kennedy (1963). In that study the authors applied an anterior directed force to the proximal femur with the tibia rigidly constrained. They described tearing in the posterior capsule at approximately 30° of hyperextension, followed by complete anterior dislocation of the knee joint (Kennedy, 1963). After cruciate injury the femur displaced posteriorly relative to the tibia, similar to the current study. Additionally in 8/10 specimens the PCL was torn, which correlates well with 5/6 specimens with PCL injury in the current study. In contrast the other previous experimental studies did not produce ACL injury (Bizot et al., 1995), or used three-point bending with fixed pivot points on the ends of the tibia and femur that limited the TF contact as the knee was hyperextended (Fomalski et al., 2008; Schenck et al., 1999). Bizot et al. (1995) likely produced posterior capsule and PCL injuries, without any ACL ruptures because their method of applying 4-point bending consisted of displacing the tibia and femur equally in the posterior direction. Instead, the current study applied equal bending moments to the tibia and femur, and did not constrain the extension angles to be the same for the two bones. This led to higher rotation of the femur than the tibia in all specimens which produced anterior subluxation of the tibia. Since the ACL is the primary restraint for anterior subluxation of the tibia, this motion was an important component of the ACL injuries produced from hyperextension in the current study and by Kennedy (1963). 129 Kennedy (1963) also constrained axial rotation of the tibia. In the current study we followed that procedure, but allowed valgus bending and lateral displacements of the femur. The bending strength of the knee has been measured previously under lateral- medial bending moments that simulate automotive impacts onto pedestrian lower limbs (Kajzer et al., 1999; Kerrigan et al., 2003). Although the injuries produced in those studies are different than the current study, Yamada (1970) predicted that lateral-medial and anterior-posterior bending strengths would be similar. In fact, the current value of 108:l:46 Nm for joint failure in A-P bending compared well with three lateral-medial bending failures that occurred between 130 and 142 Nm in a previous study (Kerrigan et al., 2003). Slight differences in the failure moments may have been due to the high rates of loading used in the previous L-M experiments. The current experiments were conducted at a much lower loading rate (250 ms to peak), which might better simulate the loading rate that occurs in sports. Hyperextension of the knee joint produced bicruciate ligament injuries in all but one specimen (ACL only). Additionally, the posterior capsule was damaged in all specimens. In a previous biomechanical study by another laboratory, ACL tension was significantly increased at 5° of hyperextension with the addition of either an internal torque or a varus moment, but was not increased with an external torque on the tibia or a valgus moment on the femur (Markolf et al., 1995). In sports injury scenarios, hyperextension moments are likely combined with internal tibia torque (Lohmander et al., 2004) and/or TF compression (von Porat et al., 2004) that may produce isolated ACL injury. In noncontact injury scenarios, it would be difficult to produce a large 130 hyperextension moment without a significant TF compressive load occurring simultaneously (either from weight bearing or muscle contraction force). This may therefore suggest another limitation of the current study, where pure hyperextension moments were applied and produced combined injuries in the joint. Future studies should be considered that incorporate axial compressive loads in the knee during application of hyperextension moments. This may better simulate the in vivo situation and produce a higher frequency of isolated ACL rupture. In summary, the study provided contact pressure distributions for an ACL injury mechanism, and documented the associated osteochondral microtrauma in the knee. The results of this study suggest that damage to articular cartilage overlying MRI detected bone bruises in patients with ACL tears may be due to high contact pressures generated in the TF joint and form the basis for subsequent development of OA (Johnson et al., 1998; Mink and Deutsch, 1989). The current experimental study verified the clinical “footprint” bone bruise pattern suggested for hyperextension injuries of the knee. In fact, little change between pre—failure and failure contact pressures during the hyperextension experiments may indicate that athletes who undergo this type of loading, even without ligament rupture, could be at risk for associated injuries to articular cartilage and underlying subchondral bone that may require treatment in order to prevent long-term joint disease. 131 REFERENCES Arendt E, Dick R. Knee injury patterns among men and women in collegiate basketball and soccer: NCAA data and review of literature. Am J Sports Med 1995;23:694-701. Bizot P, Meunier A, Christel P, Witvoet J. Experimental passive hyperextension injuries of the knee: biomechanical aspects and their consequences (French). Revue de Chirurgie Orthop. 1995;81:21 ‘1-220. Boden B, Dean G, Feagin J, Garrett W. Mechanisms of anterior cruciate ligament injury. Orthop. 2000;23:573-578. Brautigan B, Johnson DL. The epidemiology of knee dislocations. Clin Sports Med. 2000;19:387-397 Crotty JM, Snow RD, Brogdon BG, DeMouy EH. Magnetic resonance imaging of trauma patterns in the knee. Emergency Rad. 1998;52237-244. Fomalski S, McGarry MH, Csintalan RP, Fithian DC, Lee TQ. Biomechanical and anatomical assessmendt after knee hyperextension injury. Am J Sports Med. 2008;36:80-84. Heinrichs A. A review of knee dislocations. J Ath Train. 2004;39:365-369. Johnson D, Urban W, Caborn D, Canarthos W, Carlson C. Articular cartilage changes seen with magnetic resonance imaging-detected bone bruises associated with acute anterior cruciate ligament rupture. Am J Sports Med. 1998;26:409-414. Kajzer J, Matsui Y, Ishikawa H, Schroeder G. Shearing and Bending Effects At the Knee Joint At Low-Speed Lateral Loading. SAE Paper. 1999-01-0712. Kaplan P, Walker C, Kilcoyne R, Brown D, Tusek D, Dussault R. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: Assessment with MR imaging. Musculoskeletal Rad. 1992;183:835-838. Kennedy JC. Complete dislocation of the knee joint. J Bone Jt Surg. 1963;45:889-904. Kerrigan JR, Bhalla KS, Madeley J, Funk JR, Bose D, Crandall JR. Experiments for establishing pedestrian-impact lower limb injury criteria. SAE Paper. 2003-01-0895. Kwolek CJ, Sundaram S, Schwarcz TH, Hyde GL, Endean ED. Popliteal artery thrombosis associated with trampoline injuries and anterior knee dislocations in children. Am Surg. 1998;64:1183-1187. Lohmander L, Stenberg A, Englund M, Roos H. High prevalence of knee osteoarthritis, pain, and functional limitations in female soccer players twelve years after anterior cruciate ligament injury. Arth Rheum. 2004;50: 3145-3152. 132 Markolf KL, Burchfield DM, Shapiro MM, Shepard MF, Finerman GAM, Slauterbeck JL. Combined knee loading states that generate high anterior cruciate ligament forces. J Orthop Res. 1995;13:930-935. Meyer, E., Baumer, T., Slade, J ., Smith, W., Haut, R.. Tibiofemoral contact pressures and osteochondral microtrauma during ACL rupture due to excessive compressive loading and internal torque of the human knee. Am J Sports Med. 2008;36(10):1966- 1977. Meyers M, Harvey P. Traumatic Dislocation ofthe knee joint. J Bone Jt Surg. 1971;53- A:16-29. Olsen 0, Myklebust G, Engebretsen L, Bahr R. Injury mechanism for anterior cruciate ligament injuries in team handball: A systematic video analysis. Am J Sports Med. 2004;32:1002-1012. Remer EM, Fitgerald SW, Friedman H. Anterior cruciate ligament injury: MR imaging diagnosis and patterns of injury. RadioGraphics. 1992;12:901-915. Sanders T, Medynski M, Feller J, Lawhom K. Bone contusion patterns of the knee at MR imaging: Footprint of the mechanism of injury. RadioGraphics. 2000;20:Sl35-Sl 51. Schenck RC, Kovach IS, Agarwal A, Brummelt R, Ward RA, Lanctot D, Athanasiou KA. Cruciate injury patters in knee hyperextension: A cadaveric model. Arthroscopy. 1999;15:489-495. Shelboume KD, Klootwyk TE. Low—velocity knee dislocation with sports injuries: Treatment principles. Clin Sports Med. 2000;19:443-456. Terzidis I, Christodoulou A, Ploumis A, Metsovitis S, Koimtzis M, Givissis P. The appearance of kissing contusion in the acutely injuried knee in the athletes. Br J Sports Med. 2004;38:592-596. Torzilli PA, Deng X, Warren RF. The effect of joint-compressive load and quadriceps muscle force on knee motion in the intact and anterior cruciate ligament-sectioned knee. Am J Sports Med. 1994;22:105-112. Vanhoenacker FM, Snoeckx A. Bone marrow edema in sports: General concepts. Euro J Rad. 2007;62:6-15. Viskontas DG, Giuffre BM, Duggal N, Graham D, Parker D, Coolican M. Bone bruises associated with ACL rupture: Correlation with injury mechanism. Am J Sports Med. 2008;36: 927-933. von Porat A, Roos EM, Roos H. High prevalence of osteoarthritis 14 years after an anterior cruciate ligament tear in male soccer players: A study of radiographic and patient relevant outcomes. Ann Rheum Dis. 2004;63:269-273. 133 Wang JB, Rubin RM, Marshall JL. A mechanism of isolated anterior cruciate ligament rupture. J Bone Jt Surg. 1975;57A:4l 1-413. Washer DC, Markolf KL, Shapiro MS, Finerman GA. Direct in vitro measurement of forces in the cruciate ligaments. J Bone Jt Surg. 1993;75Az377-386. Wright RW, Phaneuf MA, Limbird TJ, Spindler KP. Clinical outcome of isolated subcortical trabecular fractures (bone bruise) detected on magnetic resonance imaging in knees. Am J Sports Med. 2000;28:663-667. Yamada H. Strength of biological materials. Baltimore, MD: The Williams and Wilkings Co. 1970. Yeow CH, Cheong CH, Ng KS, Lee PVS, Goh SCH. Anterior cruciate ligament failure and cartilage damage during knee joint compression: A preliminary study based on the porcine model. Am J Sports Med. 2008;36:934-942. 134 CHAPTER SIX TIBIOFEMORAL CONTACT PRESSURES THAT INDUCE CARTILAGE DAMAGE AND SUBCHONDRAL BONE MICROCRACKS DURING VALGUS BENDING OF THE KNEE ABSTRACT Valgus bending of the knee is promoted as an ACL injury mechanism and is associated with a characteristic “footprint” of bone bruising seen clinically. The hypothesis of this study was that during ligamentous failure caused by valgus bending of the knee, high TF contact pressures would induce acute osteochondral damage primarily in the lateral compartment of the knee. Four knee pairs were loaded in valgus bending until gross injury with or without a TF compression pre-load. The peak moment and valgus rotation of the knee joint causing gross failure were recorded. Pressure sensitive film documented the magnitude and location of TF contact. Cartilage fissures were documented on the articular surface of the tibial plateau, and microcracks were documented from microCT scans of the subchondral bone. Failures involved the ACL, MCL or both. The peak bending moment at failure was 107 Nm. Valgus bending produced regions of contact pressure on the lateral plateau over 30 MPa. Osteochondral microtrauma was observed in regions of high contact pressure. Combined valgus bending and TF compression produced higher TF contact pressures, but did not alter the gross injury pattern from isolated valgus bending experiments. The pattern of damage in the lateral facet was specific for this mechanism of loading. Athletes who sustain a severe valgus knee bending injury may be at risk of acute osteochondral damage especially if the loading mechanism occurs with a significant TF compression component. 135 INTRODUCTION Valgus bending of the knee is one of the most commonly referenced loading mechanisms for ACL rupture in athletes. This type of motion is described in over 60% of non-skiing ACL injuries (Boden et al., 2000). In basketball, approximately 37% of non- contact ACL injuries were termed “valgus collapse” (Krosshaug et al., 2007). Valgus bending of the knee is affected by many types of external forces, and associated with other motions of the joint. Specifically, there is a strong coupling between valgus bending and axial tibial rotation (Inoue et al., 1987; Matsumoto et al., 2001). In experiments that allow motion in five out of the six possible degrees of freedom (all except knee flexion/extension), ACL sectioning significantly increases valgus laxity, while MCL sectioning does not (Inoue et al., 1987). The importance of the ACL in restraining valgus bending was also demonstrated by a large force in the ACL at 30° of knee flexion (Fukuda et al., 2003). Other biomechanical studies, however, have shown significantly more restraint from the MCL than the ACL during valgus bending due to its anatomical location and orientation in the knee joint (Seering et al., 1980; Shapiro et al., 1991). Although valgus bending is frequently identified in video analysis of ACL injuries in sports, it is not clear if this motion induces the injury or occurs as a result of the ACL being torn (Olsen et al., 2004). Post-traumatic OA development in patients with ligament tears may be caused by acute damage to the articular cartilage and subchondral bone due to excessive compressive forces generated in the joint at the moment of injury (Fang et al., 2001; Frobell et al., 2008). In over 80% of ACL cases and 50% of MCL cases, there is a characteristic osteochondral lesion in the tibial plateau and/or the femoral condyle 136 (Atkinson et al., 2008). Geographic bone bruises, in particular, are also a sign of cartilage softening, fissuring or overt chondral fracture in regions overlying bone bruises (Vellet et al., 1991; Johnson et al., 1998). Osteochondral microdamage is produced as a “footprint” of the pattern of the joint contact at the moment of ACL injury (Sanders et al., 2000; Viskontas et al., 2008). These “footprints” were shown to be specific to the mechanism of ACL injury due to high TF contact pressures produced by TF compression and hyperextension loading mechanisms in the preceding chapters. Few studies have documented the forces or relative joint displacements at failure levels under controlled loading of the knee joint. The objective of the current study was to apply failure level valgus bending moments to the knee and document the soft tissue injuries and osteochondral microdamage that occur during this event. The study was designed to measure the contact pressure occurring in the knee joint during failure level valgus bending moments. Additionally, in the current study valgus bending and TF compression were combined to better simulate an off balance jump landing. It was hypothesized that valgus bending would result in ligament failure and that during gross ligamentous injury to the knee valgus loads would generate high contact pressures in the lateral plateau causing acute osteochondral microdamages. These damaged regions may have the potential for development of post-traumatic OA, even after surgical reconstruction of the ligamentous injury. METHODS Valgus bending experiments were conducted on paired TF joints from four male cadavers aged 40i15 years (Table 6.1). The joints were stored and prepared as described in Chapter 2. One side was randomly selected for isolated valgus bending experiments, 137 while the opposite limb was used for combined loading experiments with a valgus bending moment and a TF compression pre-load. Val gus moments were applied via four- point bending, with the moment applied to the potting cups (Figure 6.1). The femoral cup was mounted on an XY translational table to allow anterior/posterior and proximal/distal motion. These motions were recorded with linear encoders (Model #XOOZOIA, Renishaw, Hofman Estates, IL, USA). The flexion angle was fixed at 30°, while intemal/external rotation of the femur was unconstrained and recorded with a rotary encoder (Model #RCH25D-6000, Renco Encoders Inc., Goleta, CA, USA). The tibia rotation was fixed in all directions of motion except the applied valgus bending. A hydraulic materials testing machine with a linear actuator (Models 312.21 and 204.52, MTS Corp., Eden Prairie, MN, USA) was used to apply the bending moment with a time to peak load of 250 ms. Valgus bending was applied to the joint through the four-point bending moment arm via repeated, increasing magnitudes of actuator displacement (in 5 mm increments) until gross injury of the joint. In the combined valgus bending and TF compression experiments the TF compression pre-load was applied immediately before each valgus bending test by hand using a lever arm that was connected to the femoral fixture through extension springs. Two springs (Part # 9630K33, McMaster-Carr, Atlanta, GA, USA) with a stiffness of 161 N/cm each were arranged in parallel and displaced 7 cm from their nominal spring length in order to produce approximately 2 times body weight (x BW) of TF compression in the first three specimens. In the last specimen, four springs were used to produce approximately 4 x BW of TF compression. Motion of the femur and tibia were also measured during valgus bending using a Vicon motion capture system (Oxford Metrics Ltd., Oxford, UK). Reflective markers were 138 attached to the femur and tibia fixtures and four marker arrays were screwed into each bone. For the current study, only the femur and tibia valgus angle was analyzed with the motion analysis system. Relative displacements between the tibia and femur were measured at the time corresponding to the peak moment. Linear Actuato I 3-1.9,. M 4-Point Bending 42"“ Load Cell Moment Arm .,; *1" '- Axial Femur .. Free Rotation 'I—.. I :1 B! , “ - Lever Arm to . . h) ‘v ‘54 lela 934:." _ Apply TF - § . _[e- o X-Y Plate Spring If . "-1 Displacement ; Figure 6.1 Knee specimen, potted and attached to the valgus bending test fixture. Table 6.1 Valgus bending specimen information and test parameters. S ecimen Sex ' Height (m) ' e :Compressive Rate 9 {Age (yrs) (Weight(kg)s 9 Preload (N) . (Hz) " ’ ‘ . , i 32462 -——-M—-— «137—5 4-3-..- __03 _--_2_-_5._. - _ _52_ 105 _ _L_“__ 2250 ‘33.-. 32498 M 1.78 L l _ 0 _-. _ 2.5 f_ ,_ 3-117-- _ -94 ,_ -.R -, - 82559- .33 32489 — M 4 - 1'78 _ R - 0 . .99 _ _-_V_____,_ 49._ _ _, 1.14. 3 ': -2320--- 3:? 32532 M 1'88 L _ 0-- _ .255 19 86 R 3500 j 2.5 139 Additionally, as described in Chapter 3, pressure film packets were inserted into the medial and lateral compartments of the TF joint to record the contact area and pressures (Figure 6.2). For comparison purposes only the medium film contact areas and average pressures were reported, except when the medium film area was zero, and then the average pressure from the low pressure film was determined. Additionally, the area over 25 MPa was determined to show the region of joint contact that was at most risk of microdamage. Figure 6.2 Knee specimen attached to the valgus bending testing fixture with pressure film packets inserted in the medial and lateral compartments. All soft tissue injuries and bone fractures were documented and photographed during a careful gross dissection of the joint, while separating the tibia, removing the menisci, and highlighting fissures with India ink. After the failure experiments, the tibial plateaus were scanned using a GE Explore Locus microCT system (General Electric, F airfield, CT, USA) at a voxel resolution of 90 um obtained from 400 views. The beam angle of increment was 0.5, and the beam strength was set at 80 kvp and 450 tiA. The 140 volumetric image data was visualized with the GE Healthcare MicroView software application to document subchondral bone microdamage in the medial and lateral compartments. Potential damage was identified in the anterior, central and posterior regions of each facet. Two-way, repeated measure ANOVAs with factors of loading type (isolated valgus bending, combined valgus bending and TF compression) and location (medial, lateral) were used to compare the contact area, average pressure and maximum pressure. One-way, repeated measure ANOVAs with a factor of loading type (isolated valgus bending, combined valgus bending and TF compression) were used to compare peak moments and the relative joint displacements. SNK-post hoe tests were used where appropriate for multiple comparisons, and significance was determined for p < 0.05. RESULTS The application of a valgus bending moment resulted in internal femur rotation and valgus rotation of the knee joint (Table 6.2). No significant differences in the amount of valgus rotation or internal femur rotation were documented between the isolated valgus bending group and the combined valgus and TF compression group. The average valgus rotation at failure was 29i7.2 degrees, and the average internal femur rotation was 38:1:14 degrees. The data presented in Table 6.2 was recorded only during the application of the valgus bending moment. The initial application of the TF compression pre-load in the combined valgus bending experiments resulted in approximately 25° of varus knee joint rotation, 5° of external femur rotation and 4 mm of posterior femoral subluxation. No posterior displacement was noted during the application of the valgus bending moment. 14] All failures, except for one specimen, involved the MCL. There were also three cases of partial rupture or avulsion of the ACL (Table 6.2). The average failure valgus bending moment was similar between the isolated valgus bending and the combined valgus bending and TF compression experiments, yielding an average of 1073:52 Nm (Table 6.2). Table 6.2 Maximum valgus bending moment, knee motion at peak moment and knee joint injuries documented during dissection. Complete tear (X), partial tear (I), avulsion (/\), tibia (T), fibula (F). ’ internal (+) I Peak Valgus (+) I Injuries : Age . ' Femur — . (is) Moment Rotation Rotation . y (Nm) (deg) MCL lACL PCL LCL (deg) 1 4, 2,44____52__ 31811....31153 ,1 ----M52 g 32498R 47 96 36 59 i / I . g 32439L 40 63 20 23 x . i 0 32532R 19 93 1 24 i 31 I x L/\(‘I) Avg (SD) 109 (52) 27 (6.8) I 42 (17) I E>~ 32462R 52 39 i 36 L 48 / I a 32498L 47 225 20 ' 22 x ‘ , 34 732439.; 40 91 . 35 32 x — _ --.._,.___ -3... -+ - —l _ __-___+_._..- ___—___ g 32532L 19 68 , 35 1 35 x / . Avg (s0) 106(82h 32 (7.7) 1 34(11) , Both types of experiments generated high contact pressures in the lateral compartment (Figure 6.3). In the isolated valgus bending experiments there was no pressure recorded by the medium pressure film in the medial compartment for any specimen. The low pressure film was then considered, but two specimens still exhibited no medial pressure (Table 6.3). The average and maximum pressure in the lateral compartment was significantly higher compared to the medial compartment for isolated valgus bending experiments. In the combined valgus bending and TF compression I42 experiments, pressure was recorded in the medial compartment, but the contact area, average pressure and maximum pressure were significantly less than in the lateral compartment (Table 6.3). .\ nle rio r 3 \l I’a M 35 \ll’a Posterior Figure 6.3 Representative (specimen 32532) contact pressure patterns for an isolated valgus bending experiment (Lt) and a combined valgus bending and TF compression experiment (Rt). Lateral (L), medial (M). Table 6.3 Data from maximum contact pressure distribution. *Significant difference between medial and lateral compartments. #Significant difference between isolated valgus bending and combined valgus bending and TF compression groups. Medial Plateau Lateral Plateau Area Area Area Over 25 Pretfure Przsifire Area Oxer 25 Pre:fure Prgzre MPa MPa (mm’) (MPa) (mm’) (MPa) 3 32462L 19 o 11 15 257 63 20 39 .g 32498R_ 96 o 12 20 167 1 15 26 5 32489L 130 o 15 22 345 17 16 34 32532R 12 o 11 20 536 15 16 32 Avg (30) 64(58) 0(0) 12(2)# 19(3)# 326 (158)*# 24 (27) 17(2)* 33 (6)* g 32sz o o o o 56 o 11 15 2 32498L o o o o 245 11 16 33 g 32489R o o 5 7 166 16 16 32 > 32532L o o 5 9 104 0 13 22 Avg (SD) 0 (0) o (0) 3 (3) 4 (5) 143 (82) 7 (8) 14 (2)* 26 (8)* 143 The articular cartilage exhibited fissures in all specimens (Figure 6.4), except for one specimen from the isolated valgus bending group (Table 6.4). A majority of the articular cartilage damage was noted on the lateral facet, with only two specimens also displaying medial fissures and both of those were from the combined valgus bending and TF compression group. The subchondral bone damage was documented using microCT (Figure 6.5). Microcracks under the lateral facet of the tibial plateau were identified in all four of the specimens tested in combined valgus bending and TF compression. Lateral microcracks were also identified in two of the four specimens tested in isolated valgus bending. The two specimens with no tibial plateau microdamage experienced maximum pressures less than 25 MPa. Table 6.4 Maximum pressure and articular cartilage (AC) and subchondral bone (SB) damage. Maximum Pressure Gross AC Damage Micro-CT SB Damage (MPa) Med Lat Med Lat Med Lat 32462L ii _. _.‘ . i E 32498R 1 IE ‘- ' ' g 7.3.,“ 3.; o 32489L , , ] 32532R 32462R s O 32498L m a a 32489R > 32532L . 144 Anterior Figure 6.4 Gross surface fissures stained with India ink for representative Cartllage Surface Mlcro-CT Flgure 6.5 Gross cartilage damage and corresponding subchondral bone damage on the lateral tibial plateau of specimen 32498L. Reference lines in microCT images indicated the relative slice location between planes. Lateral (L), medial (M), anterior (A), posterior (P) 145 DISCUSSION The valgus bending mechanism of ACL rupture is associated with bone bruises in the lateral aspect of the tibial plateau and lateral femoral condyle (Hayes et al., 2000; Sanders et al., 2000). This clinical “footprint” compares well with the current results showing cartilage damage and subchondral bone microcracks in cadavers. All of the combined valgus bending and TF compression specimens had cartilage damage and subchondral bone microcracks, and three of the isolated valgus bending specimens had cartilage damage with two of those also having subchondral bone microcracks. These osteochondral microdamages were located in the lateral tibial plateau. Bone bruises occur in the lateral compartment 3-5 times more frequently than in the medial compartment (Speer et al., 1992; Rosen et al., 1991; Spindler et al., 1993). One clinical injury mechanism that likely produces bone bruising only in the lateral compartment is a forced valgus moment from contact with another player who impacts the lateral side of the knee (Sanders et al., 2000, Viskontas et al., 2008). This loading scenario may be similar to the valgus bending only experiments in the current study. On the other hand, the mechanism of injury associated with a non-contact jump landing also produces bone bruises in the lateral facet, but there is an increase in the prevalence of bone bruises in the medial facet as well (Hayes et al., 2000; Viskontas et al., 2008). This non-contact scenario may compare to the combined valgus bending and TF compression experiments of the current study. Based on these results, while the risk of osteochondral damage in the lateral compartment may be slightly reduced for an injury mechanism without a substantial axial load, the likelihood of medial compartment damage may be significantly increased in cases with a combined axial tibial compression component. 146 Previously TF compression experiments have shown regions of high contact pressure (>30 MPa) on both the medial and lateral facets of the tibial plateau (Meyer et al., 2008). It was interesting that both types of valgus bending experiments also produced an average maximum contact pressure of approximately 30 MPa. However, in the current study, isolated valgus bending experiments resulted in TF contact pressures located only on the lateral facet of the tibial plateau. In the combined valgus bending and TF compression experiments the contact was distributed slightly more to the medial side, but remained significantly higher on the lateral facet than the medial facet. There was a statistically significant difference in the maximum pressure on the medial facet between the two experimental groups, but not for the lateral facet. Subchondral bone microcracks were produced in the lateral tibial plateau for all specimens with contact pressures over 25 MPa. Microdamage to the calcified cartilage and subchondral bone have been strongly implicated in the development of post-traumatic OA (Burr and Radin, 2003; Thambyah, 2005), and this type of damage seems to occur at a contact pressure threshold of approximately 25 MPa (Thambyah et al., 2008; Verteramo and Seedhom, 2007). Surface fissures of the tibial articular cartilage were also produced in these regions of high contact pressure. Additionally, in one lateral and two medial tibial plateaus there were cartilage fissures in regions with contact pressures above 20 MPa. Impact loading has been shown to initiate damage to articular cartilage between 11-36 MPa of contact pressure depending on the thickness, rate of loading and location (Repo and Finlay, I977; Haut, 1989; Atkinson et al., 1998). Although histology has been commonly used to document subchondral bone microcracks, this method is somewhat limited by providing only a 2D slice of the tissue I47 in selected regions. MicroCT, on the other hand, has also been previously used to document subchondral bone microcracks after a severe impact (~25 MPa) on the surface of cartilage (Thambyah et al., 2008). In the current study a microCT scan was used to identify microcracks in the lateral facet of six out of eight specimens. In addition to lateral microcracks, there were also two cases of microfractures in the central-medial tibial plateau due to initiation of an ACL avulsion, even though one of these (32498R) had been identified as a partially torn ACL. Articular cartilage fissures were documented over the medial and lateral facets by wiping India ink over the tibial surface. The current study documented good agreement between the location of stained cartilage fissures and microCT subchondral bone microcracks. These data compare favorably to clinical studies showing damage to articular cartilage overlying bone bruises (Vellet et al., 1991; Johnson et al., 1998). Seven out of eight of the valgus bending experiments resulted in MCL rupture, and three out of eight experiments generated ACL injury (one isolated, two combined with MCL injury). There was no difference in the gross injury patterns produced by the two experimental loading conditions. Epidemiological studies also document similar rates for combination MCL and ACL injuries for both non-contact and contact types of injury mechanisms (Fayad et al., 2003). The knee valgus bending moments to cause gross failure in four point bending have been previously documented (Kerrigan et al., 2003; Bose et al., 2008). In one study three knees were tested at full extension with only distal motion of the femur allowed that resulted in MCL rupture at approximately 137 Nm and 12° of valgus bending (Kerrigan et al., 2003). One specimen in that study also had an injury documented in the ACL. Another study tested knee joints in either four point I48 bending or combined bending and shear (Bose et al., 2008). The average four point bending failure moment was l21i26 Nm at a valgus angle of l4:t2°. The injuries were not grouped by the type of experiment, but most of the joints suffered MCL injury and 12.5% also had an ACL injury. The bending moments from those studies compare relatively well with the current study at 107 Nm, but the failure angles were lower than the 29° of valgus bending in the current study. This difference may be attributable to the large amount of axial femur rotation that was coupled with the valgus bending in the current study. In most descriptions of valgus collapse in sports there are similar references to coupled valgus bending, axial rotation, and even knee flexion during the injury (Cochrane et al., 2007; Krosshaug et al., 2007). Slight differences in the failure moment may also have been due to the high rates of loading (5 ms to peak) used in the previous experiments, while the current experiments were conducted at a much lower loading rate (250 ms to peak). A limitation of the study was that the lack of a statistical difference between the failure moments for the isolated valgus bending and combined valgus bending and TF compression experiments may have been due to large variations in the failure moment of some specimens within each group and the limited power of only having four specimens in each group. In summary, the study used a cadaver model to validate the valgus loading “footprint” of osteochondral injury that has been previously documented in clinical cases of ACL rupture. Severe levels of TF contact pressure were documented on the lateral facet and these regions were associated with cartilage damage and subchondral bone microcracks. The results of this study suggest that damage to the articular cartilage overlying MRI detected bone bruises in patients with ACL and MCL tears is likely due to 149 high contact pressures generated in the TF joint, and may form the basis for the subsequent development of 0A in the joint. The lack of isolated ACL injuries without a significant TF compressive load may suggest that TF compression is in fact a primary component of sports ACL injury mechanisms. Future studies should be considered that incorporate higher axial compressive loads in the knee during application of valgus moments. This may better simulate the in vivo situation and produce a higher frequency of isolated ACL rupture. 150 REFERENCES Atkinson P, Newberry W, Atkinson T, Haut R. A method to increase the sensitive range ofpressure sensitive film. J Biomech. 19983;31:855-859. Atkinson T, Haut R, Alterio N. Impact-induced fissuring of articular cartilage: an investigation of failure criteria. J Biomech Eng. l998b;120: 181-187. Atkinson P, Cooper T, Anseth S, Walter N, Kargus R, Haut R. Association of knee bone bruise frequency with time postinjury and type of soft tissue injury. Orthop. 2008;31:440. Boden B, Dean G, Feagin J, Garrett W. Mechanisms of Anterior Cruciate Ligament Injury. Orthop. 2000;23:573-578. 803, D, Bhalla K, Untaroiu C, Ivarsson B, Crandall J, Hurwitz S. Injury tolerance and moment response of the knee joint to combined valgus bending and shear loading. J Biomech Eng. 2008;130:031008 1-8 Burr D, Radin E. Microfractures and microcracks in subchondral bone: are they relevant to osteoarthritis? Rhem Dis Clin North Am. 2003;29:675-685. Cochrane J, Lloyd D, Buttfield A, Seward H, McGivem J. Characteristics of anterior cruciate ligament injuries in Australian football. J Sci Med Sports. 2007;10:96-104. Fang, C., Johnson, D., Leslie MP, Calson, C., Robbins, M., Cesare, P., 2001. Tissue distribution and measurement of cartilage oligomeric matrix protien in patients with magnetic resonance imaging-detected bone bruises after acute anterior cruciate ligament tears. J. Orthop. Res. 19, 634-641. Fayad L, Parellada J, Parker L, Schweitzer M. MR imaging of anterior cruciate ligament tears: is there a gender gap? Skeletal Radiol. 2003;32:639-646. Frobell, R., Roos, H., Roos, E., Hellio Le Graverand, M., Buck, R., Tamez-Pena, J., Totterman, S., Boegard, T., Lohmander, L., 2007. The acutely ACL injuried knee assessed by MRI: are large volume traumatic bone marrow lesions a sign of severe compression injury? Osteo. Cart. 16, 829-836. Fukuda Y, Woo S, Loh J, Tsuda E, Tang P, McMahon P, Debski R. A quantitative analysis of valgus torque on the ACL: A humen cadaveric study. J. Orthop. Res. 2003;21:1107-1112. Haut R. Contact pressures in the patellofemoral joint during impact loading on the human flexed knee. J Orthop Res. 1989;7z272-280. Hayes C, Brigido M, Jamadar D, Propeck T. Mechanism-based pattern approad to classification of complex injuries of the knee depicted at MR imaging. RadioGraphics. 2000;20:8121-134. lSl Hootman J, Macera C, Ainsworth B, Addy C, Martin M, Blair S. Epidemiology of musculoskeletal injuries among sedentary and physically active adults. Med Sci Sports Exercise. 2002;34:838-844. Inoue M, McGurk-Burleson E, Hollis M, Woo S. Treatment of the medial collateral ligament injury. I: The importance of anterior cruciate ligament on the varus-valgus knee laxity. Am J Sports Med. 1987;15:15-21. Johnson D, Urban W, Caborn D, Canarthos W, Carlson C. Articular cartilage changes seen with magnetic resonance imaging-detected bone bruises associated with acute anterior cruciate ligament rupture. Am J Sports Med. 1998;26:409-414. Kerrigan J, Bhalla K, Madeley J, Funk J, Bose D, Crandall J. Experiments for establishing pedestrian-impact lower limb injury criteria. SAE Paper. 2003-01-0895. Krosshaug T, Nakamae A, Boden B, Engebretsen L, Smith G, Slauterbeck J, Hewett T, Bahr R. Mechanisms of anterior cruciate ligament injury in basketball: Video analysis of 39 cases. Am J Sports Med. 2007;35:359-367. Majewski M, Susanne H, Klaus S. Epidemiology of athletic knee injuries: A 10-year study. Knee. 2006;13:184-188. Matsumoto H, Suda Y, Otani T, Niki Y, Seedhom B, Fujikawa K. Roles of the anterior cruciate ligament in preventing valgus instability. J Orthop Sci. 2001;6z28-32. Mink J, Deutsch A. Occult cartilage and bone injuries of the knee: Detection, classification, and assessment with MR imaging. Radiology. 1989;170:823-829. Meyer E, Baumer T, Slade J, Smith W, Haut R. Tibiofemoral contact pressures and osteochondral microtrauma during ACL rupture due to excessive compressive loading and internal torque of the human knee. Am J Sports Med. 20083;36:1966- 1977. Olsen 0, Myklebust G, Engebretsen L, Bahr R. Injury mechanism for anterior cruciate ligament injuries in team handball: A systematic video analysis. Am J Sports Med. 2004;32:1002-1012. Repo R, Finlay J. Survival of Articular Cartilage after Controlled Impact. Am J Bone Jt Surg. 1977;59:1068-1076. Rosen M, Jackson D, Berger P. Occult osseous lesions documented by magnetic resonance imaging associated with anterior cruciate ligament ruptures. Arthroscopy. 1991;7245-51. Sanders T, Medynski M, Feller J, Lawhom K. Bone contusion patterns of the knee at MR imaging: Footprint of the mechanism of injury. RadioGraphics. 2000;20:Sl35-SlSl. 152 Seering W, Piziali R, Nagel D, Schurman D. The function of primary ligaments of the knee in varus-valgus and axial rotation. J Biomech. 1980;13:785-794. Shapiro M, Markolf K, Finerman G, Mitchell P. The effect of section of the medial collateral ligament on the force generated in the anterior cruciate ligament. Am J Bone Jt Surg. 1991;73:248-256. Speer K, Spritzer C, Basset F, F eagin J, Garrett W. Osseous injury associated with acute tears of the anterior cruciate ligament. Am J Sports Med. 1992;20:382-389. Spindler K, Schils J, Bergfeld J, Andrish J, Weiker G, Anderson T, Piraino D, Richmond B, Medendorp S. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. AmJ Sports Med. 1993;21:551-557. Thambyah A. A hypothesis matrix for studying biomechanical factors associated with the initiation and progression of posttraumatic osteoarthritis. Med Hypoth. 2005;64:1157-1161. Thambyah A, Shim V, Chong L, Lee V. Impact-induced osteochondral fracture in the tibial plateau. J Biomech. 2008;41:1236-1242. Vellet A, Marks P, Fowler P, Munro T. Occult posttraumatic osteochondral lesions of the knee: Prevalence, classification, and short-terrn sequelae evaluated with MR imaging. Radiology. 1991;178:271-276. Verteramo A, Seedhom B. Effect of a single impact on the structure and mechanical properties of articular cartilage. J Biomech. 2007;40:3580-3589. Viskontas D, Giuffre B, Duggal N, Graham D, Parker D, Coolican M. Bone bruises associated with ACL rupture: Correlation with injury mechanism. Am J Sports Med. 2008;36:927-933. 153 CHAPTER SEVEN TIBIOFEMORAL CONTACT PRESSURES GENERATE CARTILAGE AND SUBCHONDRAL BONE DAMAGE DURING ACL RUPTURE: A F INITE ELEMENT ANALYSIS ABSTRACT The contact pressure distribution in the TF joint has been measured and computed for many loading mechanisms at physiological force levels. In the experimental loading mechanisms of the current study, severe magnitudes of contact pressure were documented. The final objective of this dissertation was to build a anatomically-correct computational model, using literature-based material properties to predict the experimentally-produced osteochondral microdamages. Representative contact pressure distributions from pre-failure tests were applied to the articular surface of the tibial plateau. Maximum shear stress (Tresca) damage criteria were applied to the articular cartilage, subchondral bone and trabecular bone, based on regional variations recorded in the literature. All of the loading conditions predicted articular cartilage damage, except internal tibial torsion. The TF compression model predicted damage to cartilage, subchondral bone and trabecular bone on both the medial and lateral tibial plateau. The hyperextension model predicted a region of damage on the anterior-medial facet, and the valgus bending model predicted damage on the central-lateral facet. These patterns of predicted damage compared well with the patterns of osteochondral microdamage documented in experimentally loaded cadaveric specimens. 154 INTRODUCTION Contact pressure distributions in the TF joint have been measured (Ahmed and Burke, 1983; Thambyah et al., 2005) and computed by finite element analysis (Li et al., 2001; Haut-Donahue et al., 2002; Bendjaballah et al., 1.997; Bendjaballah et al., 1998; Jilani et al., 1997) for a number of loading mechanisms. One limitation with these previous studies is that the contact pressures are from physiological levels of loading and therefore not relevant for injury level forces/moments. In addition, these computational models of the knee generally consist of rigid bones that cannot predict when and where microdamage will occur to the articular cartilage and subchondral bone. For high levels of contact pressure, a rigid subchondral bone would also cause concentrated stress in the cartilage. In the previous experiments, each loading mechanism produced a distinct contact pressure distribution in the TF joint (Figure 7.1). There were similarities between the maximum pressures developed in three of the loading mechanisms, namely TF compression, hyperextension and valgus bending. The maximum pressures in those experiments were greater than 30 MP8 for at least one location on the tibial plateau. TF compression had the most distributed pattern of severe contact pressure, over the central and posterior regions of the lateral and medial facets. For the hyperextension specimens the severe pressures were located in the central and anterior regions of the lateral and medial facets. In valgus, the maximum contact pressure was in the central and posterior regions of the lateral facet. Internal tibial torsion experiments, on the other hand, had maximum pressures of only 25 MPa on the anterior-medial facet and they were even lower on the posterior-lateral facet. 155 Anterior Posterior l leI (ll Medial C Figure 7.1 Representative contact pressure patterns for TF compression (A), internal tibial torsion (B), hyperextension (C), and valgus bending (D). The final objective was to build an anatomically-correct 3D finite element model using literature-based properties to predict the osteochondral damage caused by the TF contact pressures in each of the previous experimental loading conditions. The material properties of subchondral bone have been shown to vary widely across the medial and lateral plateau (Hurwitz, 1998), as well as from anterior to posterior compartment (Goldstein, 1983). Therefore this layer will be modeled with an inhomogeneous Young’s modulus that is a function of the bone mineral density. The overlying cartilage layer also varies in thickness. Although cartilage is typically modeled as a biphasic material (Shepard, 1999), the viscous relaxation time is relatively large (1500 see) so that it can be reduced to an elastic material for the short loading times in the current model (SO-250 msec) (Eberhardt, 1990). 156 Load induced articular cartilage damage occurs either with or without simultaneous disruption of the underlying bone (Atkinson et al., 1998). Cartilage damage that occurs without subchondral bone damage includes fissures that begin at the cartilage surface and extend downwards at approximately 45°. Damage that includes the subchondral bone occurs as microcracks at the interface between cartilage and bone, also called the tidemark (Atkinson and Haut, 1995). Both types of damage have been explained using a Tresca, or maximal shear stress criterion (Armstrong et al., 1985; Askew and Mow, 1978; Ateshian et al., 1994; Eberhardt et al., 1991; Wilson et al., 2003). The high TF contact forces developed during ACL rupture have been linked with acute trauma to the articular cartilage and underlying subchondral bone in cadaver and clinical studies. The hypothesis of this chapter was that the contact pressure that occurs in the TF joint during ACL rupture will induce regions of stress in the articular cartilage and subchondral bone that exceed each tissue’s respective damage criteria. These regions of predicted damage will be consistent with the patterns of osteochondral microdamage documented in previous experiments. How these contact pressure distributions affect the tissues will depend on the structural and material properties of the tissues themselves. The lateral tibial plateau has a thicker layer of cartilage (Muhlbauer et al., 2000), but lower subchondral bone material properties than the medial plateau (Hurwitz et al., 1998). Even when there is an equal distribution of pressure between the medial and lateral plateaus, for example, the stress in the cartilage may be reduced and the stress in the subchondral bone may be increased on the lateral versus the medial side due to these differences in tissue properties. This effect may help explain the frequent occurrence of bone bruises in the lateral tibial plateau. Yet because higher physiological loads pass 157 L: E: ii .2. .1. '7 3 . . in” ... ,.,.1.. .....n, .v ...:.._..‘.‘.:.. ...”. ,1 3...: f. _ .15.; , , '. . C . I ‘, ,' -. '. -;—: ..-§ -21.} - ~ . - .. .. ._ _‘ _. ';':' . -. . . L...‘ . “ ' - . . h... " ' J'Mmmmmrfi-nfirwmfifihw166-16"""t'r"renw’lfi'liizzlluWS!‘”'irv-rsm.lhi.’tf.Ii!“3??“l°"'*~“~m5~'u . owing; .':.: 1. 1.;1‘ - .. ‘ n» 1,- (u.- ' ~ _ ‘ .' ,- it :5 . _ A‘.‘ . through the medial compartment, the clinical literature reports a higher rate of post- traumatic OA in the medial tibial plateau following AC L rupture. With validation of the current experimental data from the model, future parametric studies of even different loading patterns on the knee may help explain the potential for post-traumatic 0A in sports. METHODS The model geometry used in the current study was based on a subject—specific CT scan of a tibia plateau that most closely represented a healthy, middle-aged individual- The left tibial plateau of a 54 year old male (32273) with a height of 69 inches and weight of 171 lbs was chosen for the study. The subject had no history of lower extremity injury or arthritis. The knee joint was procured through university sources (see Acknowledgement) and had been stored at —20°C prior to testing in a previous study. An axial CT scan of the knee was acquired prior to testing with an in-plane resolution of .33 mm and slice thickness of .625 mm (Figure 7.2). The tibia had been potted in room temperature curing epoxy to within 7 cm of the knee joint. The potting material was included in the CT scan. To create a computational model of this geometry, the DICOM images were imported into Mimics (v12.11 Materalise, Leuven, Belgium) and a standard CT bone density threshold value (>226 Hu) was used to create a cortical bone mask. Then the trabecular bone and intennedulary canal were filled in using the graphical selection tools. These masks were used to create 3D surface objects of the tibia and femoral bones. Next the remesh option was selected to smooth the bones’ surfaces of irregularities and make sure all triangular elements had a shape measure above 0.3 (based on the height, base and angle magnitude, and an equilateral triangle is considered 1.0). In 158 addition, the subchondral and cortical bone shell and trabecular bone interiors were combined using the non-manifold part option. This assured that the final mesh would include a clearly defined subchondral plate. The quality-preserving triangular reduction algorithm was used to further simplify and smooth the surface mesh before it was converted into a volume mesh of tetrahedral elements. The resultant bone models were exported into ABAQUS CAE (v 6.8 Hibbit, Karlsson & Sorensen Inc., Pawtuchet, R1) for finite element analysis. Figure 7.2 Knee joint geometry from A) CT volume representation, B) thresholded surface model and C) finite element model. The layer of cartilage was created in ABAQUS by selecting the articular surface and creating additional elements to get the desired thickness (Figure 7.3). Since cartilage is not clearly distinguishable in CT scans, these thicknesses were based on documented values from the literature (Muhlbauer, 2000) and vary between 1.3 and 4.5 mm across the medial and lateral plateaus (Figure 7.4). 159 Figure 7.3 Layer of articular cartilage created on the medial and lateral surfaces of the tibial plateau. The material properties of cartilage and bone were assumed to be linearly elastic with moduli and Poisson’s ratios (v) as given in Figure 7.4 (Ashman et al., 1984; Cuppone et al., 2004; Knets and Malmeister, 1971; Kuhn et al., 1989; Shepard and Seedhom, 1999). Trabecular bone regions were separated and the Young’s modulus (E) was based on bone mineral density (Ciarelli et al., 1991). The damage initiation criterion thresholds were based on the relationship between shear strength and bone mineral density from the literature given in Figure 7.4 (Stone et al., 1983; Ciarelli et al., 1991). Subchondral bone regions were divided into medial and lateral plateau, and areas covered or uncovered by the meniscus. The Young’s modulus and shear stress damage criterion were scaled between trabecular bone and cortical bone based on a linear relationship with bone mineral density (Snyder and Schneider, 1991). The threshold for Tresca stress that initiates cartilage fissures is approximately 5.5 MPa. This value was predicted based on impact experiments along with the probability of a fissure occurring at a variety of loads (Atkinson et al., 1998). 160 ' 9 Coronal CT Slice Sagittal CT Slice Tresca Tissue Zone Abb. Thickness E Crlterlon v (mm) (MPa) (MPa) Medial Articular Cartilage MAC 1.5 35 5.5 .49 Lateral Articular Cartilage LAC 2 35 5.5 .49 Calcified Cartilage CC .2 350 4 .3 Medial Subchondral Bone Covered MSBC 3.5 3500 35 .3 Medial Subchondral Bone Uncovered MSBU 3 3250 34 .3 Lateral Subchondral Bone Covered LSBC 2.5 2750 31 .3 Lateral Subchondral Bone Uncovered LSBU 2 2500 30 .3 Medial High Density Trabecular Bone MTB3 3 1250 15 .3 Lateral High Density Trabecular Bone LTB3 4 1250 15 .3 Cortical Bone CB NA 14000 68 .3 High Density Trabecular Bone TB3 NA 1250 15 .3 Medium Density Trabecular Bone T82 NA 750 9 .3 Low Density Trabecular Bone T81 NA 350 4 .3 lntramedulary Canal iMC NA 0 NA N Figure 7.4 Tissue zones and material properties for bone and cartilage. Subchondral bone was divided into regions covered or uncovered by the meniscus. In order to provide a preliminary validation of the model, the first analysis was run as a contact FEM between the tibia and femur. Spring elements were created between the bones to simulate the main ligamentous structures of the knee (Figure 7.5). The ACL and PCL were each created as anterior and posterior bundles, the MCL had a superficial and deep bundle, and the LCL was only one bundle between the femur and fibula. 161 Contact areas were defined over the articular surfaces and tangent interaction was frictionless. The bones were assumed to be rigid in this model and were tied rigidly to the cartilage. The femur was fixed at the potting and an input displacement of 5 mm was applied to the tibia potting along the axis of the tibia (Figure 7.5). The resultant contact pressure distributions between the tibia and femur were compared with the results from the pressure film data presented in Chapter 3. Fixed displacements 5 mm axial displacement Figure 7.5 Contact model with input displacement along the tibial axis. In a second series of analyses, four generalized pressure distributions were applied directly to the tibial plateau representing the different types of loading experiments conducted previously (Figure 7.6). This data was documented with pressure film packets that were inserted into the medial and lateral compartments of the TF joint during the load experiments presented in the previous chapters (Figure 7.1). The location, contact area, maximum pressure, and pressure distribution were matched on both facets for each simulation (Figure 7.7). The tibial diaphyseal region at the potting interface was 162 rigidly constrained. In these analyses the resultant Tresca stress patterns in the cartilage, subchondral bone and trabecular bone were compared. Lateral Medial Posterior Anterior Lateral Plateau Medlal Plateau Tissue Zone Contact Max Pressure Contact Max Pressure Area (mmz (MPa) Area (mmz) (MPa) TF Compression A 125 25 150 30 internal Torsion B 50 20 50 20 Hyperextension C 50 25 75 30 Valgus Bending D 125 25 O 5 Figure 7.6 Generalized contact pressure distribution patterns for each loading condition. 163 Lateral Pressure Medial‘ Pressure Fixed Boundary Figure 7.7 Representative pressure distribution applied to the cartilage surfaces. The maximum Tresca stress in the anterior, central and posterior regions of the medial and lateral articular cartilage were compared with the percentage of experimental specimens with cartilage microdamage in the corresponding regions. The maximum Tresca stress was converted into pass/fail data depending on the condition of the failure criteria (5.5 MPa) was exceeded. Similarly, the maximum Tresca stress in the subchondral bone and percentage of experimental specimens with subchondral bone microcracks were compared. The failure criteria for subchondral bone were region- specific according to Figure 7.4. A Mann-Whitney Rank Sum Test was used to test for significant differences (p<0.05) between the percentage of failed specimens in regions that did or did not exceed the failure criterion thresholds. RESULTS The contact knee model used an axial displacement of 5 mm to simulate the TF compression pre-failure tests. This model revealed maximum contact pressures on the medial and lateral facets of 24 and 32 MPa, respectively (Figure 7.8). These pressures were compared with 22 and 26 MPa for the experiments. The contact area on the tibial plateau was approximately 230 mm2 on the lateral and 130 mm2 on the medial facet. This 164 produced a total contact force between the tibia and femur of 5.4 kN, which had a slightly posterior directed component due to the normals between the contacting surfaces (Figure 7.9). That compressive load matched the 5 .3 kN and the posterior component was expected to produce the anterior tibial subluxation seen in the experiments. Figure 7.9 Contact results for an applied displacement along the tibial axis. The generalized pressure distribution analyses produced a maximum Tresca stress in the articular cartilage of approximately 6 MPa for TF compression, hyperextension and valgus bending and 4.5 MPa for internal tibial torsion. The patterns of Tresca stresses on the articular surfaces are shown in Figure 7.10, where a threshold of 5.5 MPa was considered the failure criterion for cartilage (presented as dark red). In the TF compression analysis there were similar maximum shear stresses produced on each facet, but the area exceeding the damage criterion threshold was slightly larger on the medial cartilage surface than on the lateral cartilage surface. On both facets there was also a distinguishable horseshoe shaped pattern of high Tresca stress that matched the uncovered regions of increased modulus in the underlying subchondral bone. For hyperextension and valgus bending there were clear differences between the maximum 165 Tresca stresses on the medial and lateral facets. Damage was predicted in the anterior- medial hyperextension specimens and the central-lateral valgus bending specimens. :. - Figure 7.10 Cartilage surface Tresca stress (MPa) d)istributions for TF compression (A), internal tibial torsion (B), hyperextension (C), and valgus bending (D). The maximum Tresca stresses in the anterior, central and posterior regions of the medial and lateral articular cartilage were classified as damaged if the failure criteria were exceeded (Table 7.1). The regions of predicted cartilage failure matched the location and rates of experimental cartilage microdamage (Figure 7.11). The mean percentage of experimental specimens suffering damage in the five regions exceeding the failure criteria was 62%. This was significantly higher than 14% in the 19 undamaged regions (p=0.008). 166 Table 7.1 Maximum Tresca stresses (MPa) in the articular cartilage. Highlights indicate that tissue region that exceeded the Tresca damage criterion threshold of 5.5 MPa. TF internal . Valgus . . HyperextenSIon . Region Compressmn Torsron Bending Lateral Medial Lateral Medial Lateral Medial Lateral Medial Anterior - 0.5 - 4.5 5 6 1.5 - Central 4 5.5 4 2.5 4.5 4 6 2.5 Posterior 6 6 5 - - - 3.5 1.5 Figure 7.11 Percentage of specimens with cartilage microdamage in each region for TF compression (A), internal tibial torsion (B), hyperextension (C), and valgus bending (D). The Tresca stresses in the cartilage were highest along the articular surfaces and decreased towards the subchondral bone (Figure 7.12). The subchondral bone stress, on the other hand, was distributed evenly with depth. In the medial covered region, the maximum Tresca stress damage threshold was exceeded only under TF compression, while the lateral covered region was exceeded only for the applied valgus bending 167 contact pressures (Table 7.2). The damage criterion threshold was exceeded for TF compression for both the medial and lateral facets not covered by the menisci (uncovered), as well as for hyperextension in the medial uncovered region. The trabecular bone had the highest Tresca stresses at the interface with the subchondral bone and decreased with depth away from the surface. The damage threshold was exceeded in both facets during TF compression, the medial facet for hyperextension and the lateral facet for valgus bending. When comparing the Tresca stress pattern in the medial and lateral facets during TF compression there was slightly higher stress in the medial facet, but the lateral facet had a deeper extending pattern of stress that exceeded the damage criterion threshold. The regions of predicted subchondral bone failure matched the location and rates of experimental subchondral bone microcracks (Figure 7.13). The mean percentage of experimental specimens suffering damage in the six regions exceeding the failure criteria was 62%. This was significantly higher than 4% in the 18 undamaged regions (p<0.001). Table 7.2 Maximum Tresca stresses (MPa) in the subchondral bone. Highlights indicate that tissue region that exceeded the Tresca damage criterion threshold. Tresca . . TF Internal Hyperextensio Valgus CriterIon . . . Region (Mp a) Compressron Tors Ion n Bending Lateral Medial Lateral Medial Lateral Medial Lateral Medial Lateral Medial C°"°’°d 31 35 25 4o 15 20 10 30 32 5 Subchondral Bone Uncovered Subchondral Bone Trabecular Bone 15 15 16 18 12 12 14 16 18 4 34 30 50 20 30 20 35 15 10 168 Medial A Odnursumo Anterior 7 _ Posterior Figure 7.12 Coronal slices through the tibial plateau at the point of maximum Tresca stress (MPa) for TF compression (A), internal tibial torsion (B), hyperextension (C), and valgus bending (D). 169 Figure 7.13 Percentage of specimens with subchondral bone microdamage in each region for TF compression (A), intemai tibial torsion (B), hyperextension (C), and valgus bending (D). DISCUSSION The results of this study demonstrate a model of the knee for computing the stress in the cartilage and bone to predict regions of damage caused by different contact pressure distributions that occur during ACL failure. In the first analysis, the TF contact was simulated by applying the average axial tibial displacement from TF compression pre-failure tests. The pressure magnitudes and areas were measured on both plateaus and compared with corresponding values from the pressure film during experimentation. In the second analysis, a representative pressure distribution from each of the experimental loading conditions was applied to the tibial plateau. Tresca stress was used to predict regions of damage in the cartilage, subchondral bone and trabecular bone, using literature-based relationships between bone mineral density and shear strength. Then the 170 osteochondral microdamage documented in each type of experiment was compared to the regions of damage predicted by the computational model. The maximum articular cartilage shear stresses produced in the current study were between 4.5-6 MPa. The damage threshold was 5.5 MPa of Tresca stress. This criterion for damage was shown in a previous study of direct impact onto cartilage. The number came out of a 2-D FEM that correlated with the initiation of damage (Atkinson et al., 1998). Most other studies have computed much lower values of maximum shear stress in computational models of TF compression at much lower magnitudes of input force (Haut-Donahue et al., 2002; Li et al., 2001; Wilson et al., 2003). Osteochondral damage was documented experimentally in the tibial plateau and related with contact pressures from a variety of injury mechanisms in the current study. The pressure distributions produced Tresca stress patterns which predicted damage for three out of the four loading mechanisms. Specifically, damage occurred in the central and posterior regions of the lateral and medial facets of TF compression specimens. For the hyperextension specimens the damage was in the central and anterior regions of the lateral and medial facets. Under valgus loads, the damage was predicted in the central and posterior regions of the lateral facet. There was no predicted damage from internal tibial torsion. These regions correlated well with the locations of damage, both in the form of cartilage fissures and subchondral bone microcracks in the corresponding experiments. In the contact analysis, the contact pressures were slightly higher and the contact area was larger on the lateral facet than the medial facet. These distributions matched the pre-failure contact pressure results from Series 1 of the TF compression experiments (Table 3.2). This data was obtained with the ACL still intact and varus/valgus rotation l7l constrained, which was the most similar to the boundary condition for the current analysis. It was also interesting that the applied displacement of the tibia along its anatomical axis produced a contact force between the tibia and femur of 5 .4 kN. This also matched the results of pre-failure TF compression experiments (Table 2.3). A previous experimental study measured contact pressures due to TF compression with and without the meniscus intact (Ahmed and Burke, 1983). They showed that for a 1.3 kN force the maximum contact pressure was slightly increased when the meniscus was removed, but the contact area was similar to when the meniscus was intact. The biggest difference between these two conditions was in the location of contact. With the meniscus removed, the maximum pressure was located more towards the center of the facets, in the uncovered regions. A limitation of the current analysis was the lack of menisci, so it is likely that would have affected the location and magnitude of the computed contact pressures. In failure TF compression experiments, on the other hand, there were slightly higher contact pressures on the medial facet than the lateral facet. These values were used for the generalized pressure distribution analysis and produced higher Tresca stress in. the medial compartment cartilage. Similar stresses were produced in the subchondral and trabecular bone between the two facets, however. This lends support to the idea of medial bone bruises occurring from a “countercoup” effect suggested by Kaplan et al. (1999). In this scenario, the lateral compartment contusion occurs first, just prior to the ACL failure. Then for ACL injuries that occur with large TF compressive forces, a medial compartment contusion occurs during the compensatory varus bending that takes place as the knee forces are reducing. A limitation of the current study was the lack of temporal 172 contact pressure distributions to use in the knee model to investigate this effect further. On the other hand, it is clear that a bone bruise in the posterior medial compartment must occur with a large TF compressive force due to the lack of stress in this region from the other loading conditions used in the current study. The hyperextension and valgus bending experiments also produced regions with high Tresca stress in trabecular bone. These areas of damage compare well with the characteristic bone bruise patterns described in clinical studies for these ACL injury mechanisms (Hayes et al., 2000). Isolated internal tibial torsion, on the other hand, likely does not produce significant cartilage or subchondral bone damage, based on the current analysis. In clinical cases when a patient cites internal twisting, and bone bruises are documented, it might be suggested that there must also have been a large axial force in the tibia generated by the ground reaction forces. Most of the video analyses of these events show that these injuries typically occur with a large percentage of the ground reaction force being carried by the injured limb (Krosshaug et al., 2007; Olsen et al., 2004), and these values are usually several times body weight during running, cutting, or landing from a jump. A limitation of the current study was the use of linear elastic materials, especially for articular cartilage. Although, the instantaneous response has been shown to be equivalent to an incompressible elastic material (Armstrong et al., 1984; Ateshian et al., 1994), this approach does not allow for the decomposition of stress into its fluid and solid phases. Cartilage fissuring would be more closely related with the amount of stress carried by the solid phase. Additionally, for bone there are other failure criteria than Tresca stress that might be more appropriate, such as a maximum principal stress or 173 strain criteria. The Tresca, or the maximum shear stress criterion applied to all types of tissue simplified the interpretation of the results. This was deemed appropriate because the focus was placed on scaling the elastic modulus and damage threshold based on bone mineral density. Many studies have shown approximately linear relationships between bone mineral density and material properties of both cortical and trabecular bone (Ciarelli et al., 1991; Cuppone et al., 2004; Hvid and Hansen, 1985; Snyder and Schneider, 1991). However, little data exists for the subchondral plate, other than the differences between the covered and uncovered areas of the meniscus (Burr and Schaffler, 1998). An assumption was made that the material properties for subchondral bone would follow a linear relationship between trabecular and cortical bone, based on the bone mineral density. In conclusion, the current study was the first to develop a 3-D FEM of the knee to predict osteochondral microdamages generated during gross ligamentous injury. Three of the loading mechanisms induced damage in the model, with the exception being internal tibial torsion. The patterns of theoretical damage correlated well with the patterns of osteochondral microdamage documented in experiments on cadaveric specimens. 174 REFERENCES Ahmed A, Burke D. ln-vitro measurement of static pressure distribution in synovial joints—Part I: Tibial surface ofthe knee. J Biomech Eng. 1983;105:216—225. Armstrong C, Lai W, Mow V. An analysis of the unconfined compression of articular cartilage. J Biomech Eng. 1984;106:165-173. Armstrong C, Mow V, Wirth C. Biomechanics of impact induced microdamage to articular cartilage—a possible genesis for chondromalacia patella. AAOS Symp Sports Med. The Knee CV. Mosby Co, St. Louis. 7(F84. Ashman RB, Cowin SC, Van Bruskirk WC, Rice JC. A continous wave technique for the measurement of the elastic properties of bone. J Biomech. 1984;17:349-361. Askew M, Mow V. The biomechanical function of the collagen fibril ultrastructure of articular cartilage. J Biomech. 1978;100:105-115 Ateshian G, Lai W, Zhu W, Mow V. An asymptotic solution for the contact of two biphasic cartilage layers. J Biomech. 1994;37:1347-1360. Atkinson PJ, Haut RC. Subfracture insult to the human cadaver patellofemoral joint produces occult injury. J Orthop Res. 1995;13:936-944. Atkinson T, Haut R, Altiero N. Impact-induced fissuring of articular cartilage: an investigation of failure criteria. J Biomech Eng. 1998;120:181-187. Bendjaballah M, Shirazi-Adl A, Zukor D. Finite element analysis of human knee joint in varus-valgus. Clin Biomech. 1997;12:139-148. Bendjaballah M, Shirazi-Adl A, Zukor D. Biomechanical response of the passive human knee joint under anterior-posterior forces. Clin Biomech. 1998;13:625-633. Bentzen S, Hvid l, Jorgensen J. Mechanical strength of tibial trabecular bone evaluated by x-ray computed tomography. J Biomech. 1987;20:743-752. Bizot 1’, Meunier A, Christel P, Witvoet J. Experimental passive hyperextension injuries of the knee: Biomechanical aspects of their consequences. Revue de Chirurgie Orthopedique. 1995;81 :21 1-220. Burr D, Schaffler M. The involvement of subchondral mineralixed tissues in osteoarthrosis: Quantitative microscopic evidence. Micro Res Tech. 1998;37:343- 357. Ciarelli MJ, Goldstein SA, Kuhn JL, Cody DD, Brown MB. Evaluation of orthogonal mechanical properties and density of human trabecular bone from the major metaphyseal regions with materials testing and computed tomography. J Orthop Res. 1991;9:674-682. 175 Cuppone M, Seedhom BB, Berry E, Ostell AB. The longitudinal Young’s modulus of cortical bone in the midshaft of human femur and its correlation with CT scanning data. C alcif Tissue Int. 2004;74:302-309. Eberhardt A, Lewis J, Keer L. Normal contact of elastic spheres with two elastic layers as a model ofjoint articulation. J Biomech Eng. 1990;113:410-417. Goldstein S, Wilson D, Sonstegard D, Matthews L. The mechanical properties of human tibial trabecular bone as a function of metaphyseal location. J Biomech. 1983;16:965-969. Haut-Donahue T, Hull M, Rashid M, Jacobs C. A finite element model of the human knee joint for the study of tibio-femoral contact. J Biomech Eng. 2002;124:273- 280. Haut RC. Contact Pressures in the Patello-Femoral Joint During Impact Loading on the Human F lexed Knee. J Orthop Res. 1989;7z272-280. Haut RC, Atkinson PJ. Insult to the Human Cadaver Patello-Femoral Joint: Effects of Age on Fracture Tolerance and Occult Injury. Stapp Car Crash J. 1995;39:281-294. Hayes C, Brigido M, Jamadar D, Propeck T. Mechanism-based pattern approach to classification of complex injuries of the knee depicted at MR imaging. RadioGraphics. 2000;20:8121-8134. Hurwitz D, Sumner D, Andriacchi T, Sugar D. Dynamic knee loads during gait predict proximal tibial bone distribution. J Biomech. 1998;31:423-430. Hvid l, Hansen S. Trabecular bone strength patterns at the proximal tibial epiphysis. J Orthop Res. 1985;32464-472. Jilani A, Shirazi-Adl A, Bendjaballah M. Biomechanics of human tibio-femoral joint in axial rotation. Knee. 1997;4:203-213. Kaplan P, Gehl R, Dussault R, Anderson M, Diduch D. Bone contusions of the posterior lip of the medial tibial plateau (Countercoup injury) and associated internal derangements of the knee at MR imaging. Radiology. 1999;211:747-753. Knets I, Malmeister A. Deforrnability and strength of human compact bone tissue. Mechanics of Biological Solids, Bulgarian Academy of Sciences. 1977:133. Krosshaug T, Nakamae A, Boden B, Engebretsen L, Smith G, Slauterbeck J, Hewett T, Bahr R. Mechanisms of anterior cruciate ligament injury in basketball: Video analysis of 39 cases. Am J Sports Med. 2007;35:359-367. Kuhn JL,Goldstein SA, Ciarelli MJ, Matthews LS. The limitations of canine trabecular bone as a model for human: A biomechanical study. J Biomech. 1989;22:95-107. 176 Law, Li G, Lopez 0, Rubash H. Variability of a three-dimensional finite element model constructed using magnetic resonance images of a knee for joint contact stress analysis. J Biomech Eng. 2001;123:341—346. Muhlbauer R, Lukasz S, Faber S, Stammberger T, Eckstein F. Comparison of knee joint cartilage in triathletes and physically inactive volunteers based on magnetic resonance imaging and three-dimensional analysis. Am J Sports Med. 2000;28:541— 546. Olsen 0, Myklebust G, Engebretsen L, Bahr R. Injury mechanism for anterior cruciate ligament injuries in team handball: A systematic video analysis. Am J Sports Med. 2004;32:1002-1012. Shepard D, Seedhom B. The ‘lnstantaneous’ compressive modulus of human articular cartilage in joints of the lower limb. Rheumatology. 1999;38:124-132. Snyder S, Schneider E. Estimation of mechanical properties of cortical bone by computed tomography. J Orthop Res. 1991;9:422-431. Stone J, Beaupre G, Hayes W. Multiaxial strength characteristics of trabecular bone. J Biomech. 1983;16:743-752. Thambyham A, Goh J, Das De S. Contact stress in the knee joint in deep flexion. Med Eng Phys. 2005;27:329-335. Wilson W, van Rietbergen B, van Donkelaar C, Huiskes R. Pathways of load-induced cartilage damage causing cartilage degeneration in the knee after meniscectomy. J Biomech. 2003;36:845-851. 177 CHAPTER EIGHT DISCUSSION ISOLATED ACL INJURY The first hypothesis of this dissertation was; “The external tibial and valgus femoral rotations frequently identified after ACL injury are not representative of the relative displacements that cause isolated ACL failure to occur.” This hypothesis was based on biomechanical data showing that the ACL is not a primary restraint for either of these types of motion. Therefore, the objective of this study was to measure the pre- failure and failure characteristics of different loading mechanisms in order to determine the cause-and-effect relationship in the relative motion between the tibia and femur and the types of ligamentous injuries sustained. Tibiofemoral compression induced a large amount of anterior tibia subluxation in pre-failure tests (Post/Ant Motion in Table 8.1). Since the ACL is the primary restraint for anterior tibial subluxation, this motion was expected to produce tension in the ACL and eventually lead to failure. There was 50% less anterior tibial subluxation produced in the intemai tibial torsion and hyperextension bending than TF compression experiments. In the valgus bending experiments there was a small amount of anterior tibial subluxation that occurred during the compressive preload of combined experiments, but very little during the valgus bending. Internal tibial torsion and valgus bending experiments both had strongly coupled internal and valgus rotations. In the TF compression and hyperextension experiments there was very little internal tibial rotation or valgus bending of the knee. 178 After gross failure of the knee joint, the primary motions were significantly increased for each type of experiment (Table 8.1). In the TF compression experiments the magnitude of anterior tibial subluxation was more than doubled after ACL failure. In the internal tibial torsion and valgus bending experiments, intemal and valgus rotations were significantly increased after combined ACL and MCL failure, but little additional motion was observed in the other directions. There was a significant increase in the hyperextension bending angle after ACL and PCL failure, as would be expected for those experiments. Therefore these are likely the most important motions for correlating knee joint injury with the loading mechanisms from sports scenarios. Table 8.1 Joint motions prior to failure for each loading mechanism. #Significant difference between pre-failure and failure tests. Compigssion intel-I'graslilipial Hyperextension Valgus Post/Ant Motion (mm) 12 (5.2)# 6.3 (5.1) 6.9 (3.3) NA Med/Lat Motion (mm) 0.7 (3.9) 8.4 (3.3) 1.7 (1.4) NA ht/Ext Angle (deg) 3.9 (4.0)# 46 (16)# 0 38 (14)# ValgNarAngle (deg) 1.3 (3.2)# 11 (6.0)# —0.1 (3.3) 29 (7.2)# Flex/Ext Angle (deg) 30 30 -34 (11)# 30 An interesting result was documented for internal/external tibial rotation and valgus/varus bending during TF compression failure tests. These motions were observed in two separate series of experiments, due to the problems associated with applying large TF compressive forces when they were simultaneously unconstrained. In the pre-failure tests there were small intemai and valgus rotations, but in the failure tests there was either external tibial rotation or the valgus rotation was significantly increased, depending on the motion constrained. These results showed that the motions observed after ACL failure 179 were not representative of the relative displacements that produce tension in the ACL. Prior to failure there was more relative displacement in the lateral compartment, producing the small amount of internal tibial rotation. After AC L rupture, relative displacement in the medial compartment was released which produced a net external rotation of the tibia. In the case of valgus bending, the motion was limited in the pre- failure tests for all loading mechanisms except direct valgus bending. After failure in both the TF compression and intemai tibial torsion experiments, there was more than double the amount of valgus rotation as in the pre-failure test. Video analyses have identified the timing, body positioning and activities most frequent during sports ACL injuries. The injury timing is usually during the initial foot strike and the position is an extended, slightly valgus and externally rotated knee. The most frequent activities are plant-and-cut and landing maneuvers with most of the body weight distribution on the injured leg. Therefore, in both the plant-and-cut and landing maneuvers there are likely high levels of TF compression that could be of similar magnitude to loads in the current study. Similar injury mechanism characteristics have been noted in patient surveys following their ACL injury, but one area of conflict noted in these studies is the direction of tibial rotation. In many cases the patient recalls internal tibial rotation rather than external tibia rotation. This disagreement is explained based on the motion observed in the TF compression experiments. The exact time when the ACL injury occurs cannot be recorded from a video analysis and the body position can change rapidly after failure. A common problem encountered in biomechanical testing to failure is when the applied force is substantially higher than the failure tolerance of the tissues. In that case 180 the loading causes a total joint dislocation or complex fracture without a clear indication of when a particular tissue was damaged. The force and displacement data obtained for such an experiment is called right censored data, because failure occurred but it is not apparent what the exact values were at the time of failure. In the opposite scenario, a single experiment with an applied force that is too low could result in no tissue damage (left-censored data). In this case it is unclear how much more force or displacement could have been applied before the tissue failed. The alternative is to use repeated tests at increasing load levels, as was done in the current series of experiments. Frequently in these studies, failure is a force-limiting event and therefore both the peak force and tissue failure occur at the same moment (uncensored data). However, the joint displacements may continue to increase after failure due to the increased laxity available from the damaged structures (Kent and Funk, 2004). In the current study, the pre-failure tests were left-censored data points, but the forces were not very far off from the failure forces, so the peak displacement data in those tests may represent the maximum joint displacements that can be withstood without an injury. The failure test peak displacements were right- censored, but only for the injuries reported (i.e. the rest of the joint remained intact but did not prevent the large documented motions). In contrast, the forces dropped sharply in failure tests so the peak values likely correspond to the actual knee ligament injury tolerence (uncensored data). TIBIOFEMORAL COMPRESSION The second hypothesis of this dissertation was; “Tibiofemoral compression will produce anterior tibial subluxation and isolated ACL injuries, while other loading mechanisms will produce combination ligament injuries.” This hypothesis was identified 181 based on previous experiments showing that isolated ACL failure can be caused by TF compressive forces in the flexed knee. The objective of the current investigations was to provide evidence that TF compression is a primary component of the loading mechanism for non-contact, isolated AC L injuries in the extended knee. ACL injuries were produced under all four of the loading conditions, although these injuries were frequently combined with injuries to other ligamentous structures as well (Table 8.2). In the TF compression experiments, ACL injuries were produced in every one of the specimens and combined ligamentous injuries occurred in only two specimens. Internal tibial torsion also produced injuries of the ACL, but these were frequently combined with an injury to one or both of the collateral ligaments. Hyperextension of the knee frequently caused a complete anterior dislocation of the knee (injury to both the ACL and PCL). Finally, valgus bending produced ACL failure in three specimens and MCL failure in seven specimens. Therefore, in the current study it is apparent that the only loading mechanism that produced isolated failure of the ACL was TF compression. 182 Table 8.2 Failure forces and injuries sustained by the ligaments of the knee for each loading mechanism. #Significant difference between pre-failure and failure tests. TF Internal Tibial H t ' V ' Compression Torsion yperex ens on a 9'" Failure Load or Moment (kN oer) 6.0 (1 .9)# 37 (17)# 108 (46)# 106 (63) Anterior Cruciate Ligament 100% 100% 83% 38% Posterior Cruciate Ligament 15% 14% 83% 0% Medial Collateral Ligament 0% 43% 0% 88% , Lategf'frfigfile'a' 8% 29% 0% 13% Frequently lnjuried Knee Ligaments —ACL — LCL —PCL —MCL Complex injuries of the knee are fiequent in all levels of sports, and they usually occur as a result of multiple forces applied to the knee simultaneously. The isolated loading mechanisms applied to the knee joints in the current study was a way of simplifying this complex problem to more straightforward experimental conditions. The primary loading mechanisms important in cases of isolated ACL injury were identified as internal rotation of the tibia, valgus knee bending, hyperextension of the knee, and axial loading through the tibia. Varus bending, external tibial rotation, hyperflexion and anterior shear of the tibia were not tested in the current series of experiments. It is possible to have all of these loading mechanisms occur during noncontact sports injury scenarios, with the exception of anterior tibial shear, which only occurs during contact with another player or object. However, the only loading mechanisms that were tested were the ones expected to produce ACL failure, based on previous biomechanical studies. Most biomechanical evaluations of knee joint response for external loading mechanisms have been at low force levels, or to compare relative joint motions for intact 183 and ACL sectioned knees. One other study that produced isolated ACL injuries did so with a 4500 N quadriceps muscle force (DeMorat et al., 2004). At 20° of knee flexion the patellar tendon has a slight anterior directed force component which may produce anterior subluxation of the tibia during quadriceps muscle contraction. An additional loading mechanism that was not tested was distraction of the tibia and femur. Although this is a common experimental technique for measuring the stiffness and failure strength of the ACL in biomechanical studies, it was not considered relevant to the current study due to the complete lack of TF joint contact. In the real world it might be expected that more complicated combined loading mechanisms would also produce isolated ACL injury. Several possible loading combinations might be intemai tibial torsion with combined valgus bending and/or combined hyperextension moments, or any of the previously mentioned loading mechanisms combined with a significant axial tibial load component. Compressive loads in the knee joint during running and landing from a jump can be up to 15 times body weight. These TF joint contact forces are combinations of both the ground reaction and muscle contraction forces. Therefore, it is possible that the quadriceps muscle contraction in the DeMorat et al. (2004) experiments produced high TF compressive loads and anterior tibial subluxation occurred due to the posterior slope of the tibial plateau. The combined experimental results of the current study and DeMorat et al., (2004) provide substantial evidence for the importance of saggital plane motion in non-contact ACL injury mechanisms, which has been previously overlooked in the clinical literature. 184 POST-TRAUMATIC OSTEOARTHRITIS The final hypothesis of the current studies was; “The mechanism-based clinical classification of knee injuries and bone bruise pattems would correspond to characteristic distributions of high levels of contact pressure and osteochondral microdamage across the tibial plateau for each loading mechanism.” There is evidence that post-traumatic osteoarthritis is due, in part, to osteochondral microdamage that occurs at the time of the acute ligamentous injury. The final objective was to record the contact pressure distribution in the TF joint that occurred during each injury mechanism. The osteochondral microdamage patterns for each injury mechanism were also predicted in each tissue based on a 3-D computational model of the tibial plateau. Each loading mechanism produced a distinct contact pressure distribution in the TF joint (Figure 7.1). There were unanticipated similarities between the magnitudes of contact pressure developed in three of the loading mechanisms. The maximum pressures were greater than 30 MPa for at least one location on the tibial plateau for the TF compression, hyperextension and valgus bending experiments (Figure 8.1). For intemai tibial torsion experiments, the maximum pressure was approximately 25 MPa on the anterior-medial facet and even lower on the posterior-lateral facet. In the computational models, all but the intemai tibial torsion model produced maximum shear stresses in the articular cartilage, subchondral bone and trabecular bone which exceeded the thresholds for predicted tissue damage. The locations of these predicted tissue damages were variable depending on the applied pressure distribution, but cartilage damage was predicted in the articular cartilage of all loading mechanisms and subchondral and trabecular bone damage was predicted for all mechanisms except internal tibial torsion. 185 TF Compression at Anterior Central Posterior I None Figure 8.1 Percentage of specimens with articular cartilage (AC) and subchondral bone (SB) microdamage in each region for each loading mechanism (do not sum to 100%, due to multiple damaged regions per specimen). Circles represent regions of contact pressure and their line thickness represents the pressure magnitude. The pressure distributions and computational model predictions correlated well with the location of osteochondral damage, both in the form of cartilage fissures and subchondral bone microcracks. In the central and posterior regions of the lateral and medial facets of TF compression specimens, there was osteochondral damage in at least one of these regions, and sometimes multiple regions were damaged. For the hyperextension specimens the damage was limited to the central and anterior regions of the lateral and medial facets. In valgus, nearly all of the damage was in the central and posterior regions of the lateral facet, with very few specimens suffering damage in the central medial facet. Finally, intemai tibial torsion produced damage primarily in the 186 posterior-lateral and anterior-medial facets, but the percentage of specimens without any damage was elevated. The final point to be made in these comparisons was that the higher frequency of intemai tibial torsion specimens without any osteochondral damage is most likely due to the lower level of maximum joint contact pressures from this type of loading. This dose-dependant effect is also observed in the medial facet of the valgus bending specimens, where the maximum pressure was greater than 20 MPa in only three out of eight specimens. Bone bruise patterns in certain knee injuries, such as ACL rupture, have been linked to a characteristic “footprint” of the injury loading mechanism. These osseous injuries may be caused by a direct blow to the bone or by tensile forces that occur during an avulsion injury. During ACL injuries, however, bone bruises are fi'equently “kissing” contusions which are produced in the tibia and femur at the site where the opposing articular surfaces are in contact at the moment of injury. The bone bruise pattern is a static representation of the impact forces that occurred at the time of injury and provides clues to the associated soft-tissue injuries as well as the mechanism of rupture. Radiologists have used this information to construct a mechanism-based classification system to relate the patterns of bone bruising and ligament damages for complex, clinical cases of knee injury (Table 8.3). The ligamentous injury patterns described for each loading mechanism in Table 8.3 are similar to the injury results for the failure experiments in the current study (Table 8.2). Based on previous biomechanical studies at physiological force levels and a limited number of cadaver experiments at failure load level, a correlation between these loading mechanisms and ligamentous injuries was expected. However, the main purpose of the 187 current series of experiments was not just to cause gross failures of the knee joint, but to use the injury mechanisms as models for understanding patterns of osteochondral damage and evaluating the risk of developing post traumatic osteoarthritis. Contact pressure distributions in the TF joint have been measured (Thambyah et al., 2005) and computed by finite element analysis (Bendjaballah et al., 1995; Bendjaballah et al., 1997; Jilani et al., 1997) for a number of loading mechanisms, but not at failure level forces. Additionally, osteochondral damage has also been documented in the tibial plateau from a direct impact (Thambyah et al., 2008) or from TF compression in porcine knees (Yeow et al., 2008) and rabbit knees (Isaac et al., 2008), but this dissertation represents the first series of experiments to relate osteochondral damage with contact pressures for a variety of injury mechanisms. Table 8.3 Classification of loading mechanism, bone bruise patterns and gross ligament injuries (adapted from Hayes et al., 2000). Loading Mechanism Bone Bruise Location Ligament Injuries . Anterior central tibia, anterior femoral ACL, PCL, posterior Pure hyperextensron condyles capsule Pure valgus Lateral tibia, lateral femoral condyle ACL, MCL Valgus, external Posterior lateral tibia, lateral femoral rotation, axial tibia condyle, medial tibia, medial femoral ACL, MCL compression condyle Internal trbral Posterior lateral tibia, lateral femoral ACL, M CL rotation, valgs condyle Internal tibia rotation, Posterior lateral tibia, lateral femoral axial tibia condyle, medial tibia, medial femoral ACL, MCL, LCL compression condyle Axial tibia Posterior lateral tibia, lateral femoral . condyle, posterior medial tibia, medial ACL compressron femoral condyle 188 One clinical study tried to describe an ACL injury mechanism that produces medial compartment bone contusions from loading conditions that included a high TF compressive force (Kaplan et al., 1999). In this “countercoup” scenario, a lateral compartment contusion occurs first during anterior tibia subluxation on an internally rotated tibia. Then, after failure of the ACL the anterior tibia subluxation continues, but the tibia rotates externally relative to the femur. The high level of TF contact pressure remains during this secondary motion and produces a medial compartment contusion. Based on the motion and contact pressure results of the TF compression experiments this is a likely scenario, and it probably represents one of the most severe injury mechanisms in terms of the risk for developing post-traumatic osteoarthritis. In hyperextension the pre-failure contact pressure magnitudes were comparable to the failure pressures. For the other loading mechanisms the magnitude of pressure was slightly lower, but still exceeded 30 MPa. In these scenarios there may be the potential for osteochondral microdamage to occur without gross ligamentous knee injury. This occurs frequently in the clinical literature, as bone bruises are often identified on MRI without a corresponding ACL or other soft tissue injury (Atkinson et al., 2008). CONCLUSIONS Isolated ACL injuries occur from TF compression, but internal tibial torsion and valgus bending caused combined MCL injuries and hyperextension caused combined PCL injuries. TF compression produced anterior tibial subluxation leading up to isolated ACL injury. After failure, there were significant increases in external tibial rotation and valgus knee bending. Based on the current studies the magnitude of vertical ground reaction forces and muscle contraction forces producing TF compression should be 189 considered as an important mechanism of sports ACL injury scenarios. In addition, each loading mechanism produces a distinct contact pressure distribution which correlates well with the location of osteochondral microdamage. Importantly, TF compression, hyperextension and valgus bending mechanisms of joint injury produce regions of contact pressure exceeding 30 MPa. In the computational model, this level of contact pressure produces maximum shear stresses in the articular cartilage, subchondral bone and trabecular bone that exceed the threshold for predicted tissue damage. There is a long term risk of developing post-traumatic osteoarthritis from these failure levels of joint loading. In addition, however, certain pre-failure tests the maximum contact pressure were also severe, suggesting that even if ligamentous injury is absent, there may still be an enhanced risk of chronic joint degeneration. The data presented in this dissertation may be applicable to injury prediction/prevention and for clinicians’ treatment strategies for dealing with a variety of knee injuries and their potential for developing post- traumatic osteoarthritis in the longer term. 190 References Atkinson P, Cooper T, Anseth S, Walter N, Kargus R, Haut R. Association of knee bone bruise frequency with time postinjury and type of soft tissue injury. Orthop. 2008;31:440. Bendjaballah M, Shirazi-Adl A, Zukor D. Biomechanics of the human knee joint in compression: reconstruction, mesh generation and finite element analysis. Knee. 1995.;2:69-79. Bendjaballah M, Shirazi-Adl A, Zukor D. Finite element analysis of human knee joint in varus-valgus. Clin Biomech. 1997;12:139-148. DeMorat G, Weinhold P, Blackburn T, Chudik S, Garrett W. Aggressive quadriceps loading can induce noncontact anterior cruciate ligament injury. Am J Sports Med. 2004;32:477-483. Hayes C, Brigido M, Jamadar D, Propeck T. Mechanism-based pattern approach to classification of complex injuries of the knee depicted at MR imaging. RadioGraphics. 2000;20:8121-8134. Isaac DI, Meyer EG, Haut RC. Contact pressures and associated chondrocyte damage in the rabbit tibiofemoral joint under impact. J Biomech Eng. 2008;130:041018 1-5. Jilani A, Shirazi-Adl A, Bendjaballah M. Biomechanics of human tibio-femoral joint in axial rotation. Knee. 1997;4:203-213. Kaplan P, Gehl R, Dussault R, Anderson M, Diduch D. Bone contusions of the posterior lip of the medial tibial plateau (Countercoup injury) and associated internal derangements of the knee at MR imaging. Radiology. 1999;211:747-753. Kent R, Funk J. Data censoring and parametric distribution assignment in the development of injury risk functions from biomechanical data. SAE Paper. 2004-01- 0317. Meyer E, Villwock M, Haut R. Tibiofemoral contact pressures that induce cartilage damage and subchondral bone microcracks during valgus bending of the human knee. Clin Biomech. (In Review) Meyer E and Haut R. Anterior cruciate ligament injury induced by internal tibial torsion of tibiofemoral compression. J Biomech. 10.1016/ j.jbiomech.2008.09.023 Meyer E, Baumer T, Slade J, Smith W, Haut R. Tibio-femoral contact pressures and osteochondral microtrauma during ACL rupture due to excessive compressive loading and internal torque of the human knee. Am J Sports Med. 2008;36:1966- 1977. 191 Thambyam A, Goh J, Das De S. Contact stress in the knee joint in deep fiexion. Med Eng Phys. 2005;27:329-335. Thambyah A, Shim V, Chong L, Lee V. Impact-induced osteochondral fracture in the tibial plateau. J Biomech. 2008;41:1236-1242. Yeow C, C heong C, Ng K, Lee P, Goh J. Anterior cruciate ligament failure and cartilage damage during knee joint compression: A preliminary study based on the porcine model. Am J Sports Med. 2008;36:934-942. 192 RESEARCH PUBLICATIONS Peer Reviewed Manuscripts 1. Meyer EG, Villwock MR, Haut RC. Tibiofemoral contact pressures that induce cartilage damage and subchondral bone microcracks during valgus bending of the human knee. Clin Biomech (In Review) 2. Isaac DI, Meyer EG, Haut RC. Development of a traumatic anterior cruciate ligament 00 rupture model for the study of post-traumatic osteoarthritis. Clin Biomech. (In Preparation) . Meyer EG, Baumer TG, Haut RC. Tibiofemoral contact pressures and osteochondral microtrauma during ACL rupture due to hyperextension of the human knee. (in Preparation) 4. Villwock MR, Meyer EG, Powell JW, Haut RC. External Rotation Ankle injuries: 10. 11. 12. Investigating Ligamentous Ruptures. Am J Sports Med. (In Review) Meyer EG and Haut RC. Anterior cruciate ligament injury induced by internal tibial torsion of tibiofemoral compression. J Biomech. 10.1016/ j.jbiomech.2008.09.023 Villwock MR, Meyer EG, Powell JW, Fouty AJ, Haut RC. The effects of various infills and fiber structures on generating rotational traction on an artificial surface. J Sports Eng Tech. (In Publication) ' Ting D, Cabassu J, Guillou RP, Sinnott MT, Meyer EG, Haut RC, Dejardin LM. In vitro evaluation of the effect of fracture configuration on the mechanical properties of standard and novel interlocking nail systems in bending. Vet Surg. (In Review) Villwock M, Meyer E, Powell J, Fouty A, Haut R. Football playing surface components and shoe designs may affect lower extremity injury risk potential: An in situ torsional resistance assessment. Am J Sports Med. 2008 (In Publication) Meyer E, Baumer T, Slade J, Smith W, Haut R. Tibio—femoral contact pressures and osteochondral microtrauma during ACL rupture due to excessive compressive loading and intemai torque of the human knee. Am J Sports Med. 2008;36:1966- 1977. lsaac DI, Meyer EG, Haut RC. Contact pressures and associated chondrocyte damage in the rabbit tibiofemoral joint under impact. J Biomech Eng. 2008;130: 0410181-5. Guillou RP, Frank JD, Sinnott MT, Meyer EG, Haut RC, Dejardin LM. Biomechanical evaluations of medial plating for pantarsal arthrodesis-Comparative in vitro study in dogs. Am J Vet Res. 2008;69(11):1406—1412. Hansen M, Long Le, Wertheimer S, Meyer E, Haut R. Syndesmosis Fixation: Analysis of Shear Stress via Axial Load on 3.5mm and 4.5mm Quadricortical Syndesmotic Screws. J Foot Ankle Surg. 2006;45(2):65-69. 193 13. 14 15. 16. 17. Meyer EG, Haut RC. Excessive Compression of the Human Tibio-Femoral Joint Causes ACL Rupture before Bone Fracture. J Biomech. 2005; 38(11):2311-2316. .Von Pfeil D, Dejardin L, Meyer E, Weerts R, Decamp C, Haut R. Biomechanical Comparison of an Interlocking Nail and a Plate-Rod Combination; An in vitro analysis in a canine fracture model. Am J Vet Res. 2005; 66(9): 1536-1543. Kirsch J, Dejardin L, Meyer E, Decamp C, Haut R. Effect of lntramedullary Pin-plate Combination on the Mechanical Properties of Pantarsal Arthrodesis; A Comparative in Vitro Analysis in Dogs. Am J Vet Res. 2005; 66(1): 125-131. Meyer EG, Sinnott MT, Jayaraman GS, Smith WE, Haut RC. The Effect of Biaxial Impact on the 90 Flexed Human Knee Joint Stability and Injury Tolerance. Stapp Car Crash Conf J. 2004; 48: 53-70. Meyer EG, Haut RC. The Effect of Knee Impact Angle on Tolerance to Rigid Impacts. Stapp Car Crash Conf. J. 2003; 74: 1-20. Peer Reviewed Abstracts 1. Meyer EG, Baumer TG, Haut RC. Knee joint relative motion during ACL rupture by intemai tibial torsion of tibiofemoral compression. North Am. Congress on Biomechanics (NACOB), Ann Arbor, Michigan, 2008. . Meyer EG, Baumer TG, Haut RC. Tibiofemoral contact pressures and osteochondral microtrauma from ACL rupture via hyperextension and joint compression. NACOB, Ann Arbor, MI, 2008. . Meyer EG, Baumer TG, Haut RC. Tibiofemoral contact pressures and osteochondral microtrauma during ACL rupture due to excessive compressive loading and internal tibial torsion. NACOB, Ann Arbor, MI, 2008. . Villwock M, Meyer E, Powell J, Fouty A, Haut R. Football playing surface components may affect lower extremity injury risk. NACOB, Ann Arbor, MI, 2008. . Villwock M, Meyer E, Powell J, Fouty A, Haut R. Football shoe designs may affect lower extremity injury risk. NACOB, Ann Arbor, MI, 2008. . Isaac, DI, Meyer, EG, Kepich, E, Haut, RC. Acute chondrocyte damage and chronic changes in the rabbit tibiofemoral joint after impact. 54th Ann. Mtging of the Orthopaedic Research Society, San Francisco, California, 2008. . Meyer, E, Haut, R. Anterior cruciate rupture due to excessive internal torque of the human tibia. Ann. Mtging Am Society of Biomechanics, Palo Alto, CA, 2007. .Guillou, R, Frank, J, Dejardin, L, Olivier, B, Meyer, E, Haut, R. Biomechanical evaluation of medial plating for pantarsal arthrodesis- A comparative in vitro study in dogs. Am College Vet Surg, 17th Ann Symp, Chicago, Illinois, 2007. 194 9. Guillou. RP, Cabassu, JB, Sinnott, MT, Ting, D, Meyer, EG, Haut, RC, Dejardin, LM. 10. 11. 12. 13. 14. 15. 16. 17. 18. Effect of bending direction on the mechanical behavior of interlocking nail systems. 17th Ann. Phi Zeta Research Day, College of Vet Med, East Lansing, MI, 2007. Guillou, RP, Frank, JD, Sinnott, MT, Meyer, EG, Haut, RC, Dejardin, LM. Biomechanical evaluation of medial plating for pantarsal arthrodesis - A comparative in vitro study in dogs. 17th Ann. Phi Zeta Research Day, College of Vet Med, East Lansing, MI 2007. Guillou, RP, Frank, JD, Sinnott, MT, Meyer, EG, Haut, RC, Dejardin, LM. Biomechanical evaluation of medial plating for pantarsal arthrodesis—A comparative in vitro study in dogs. Ann Mtg Am College Vet Surg, 2006. Guillou, RP, Cabassu, JB, Ting, D, Sinnott, MT, Meyer, EG, Haut, RC, Dejardin, LM. Effect of bending direction on the mechanical behavior of interlocking nail systems. Ann Mtg Amer. Coll. Vet. Surg., 2006. Meyer E, Handzo D, Haut R. Chronic Changes in the Rabbit Tibiofemoral Joint Following a Single Blunt Impact. Orthop Res Soc Meeting, Chicago, IL, 2006. Bonawandt KA, Dejardin LM, Meyer EG, Haut RC. In Vitro Mechanical Evaluation of a New 3.5/2.7mm Hybrid Plate For Pantarsal Arthrodesis in Dogs. Am College Vet Surg Meeting, 2005. Dejardin LM, Bonawandt K, Meyer EM, Lansdowne JL, Haut RC: in vitro mechanical evaluation of new 3.5/2.7mm hybrid plates for tarsal arthrodesis in dogs. 14th Annual European College of Veterinary Surgeons Scientific Meeting, Lyon, France, July 6-9, 2005, pp 379. Von Pfeil D, Dejardin L, Meyer E, Weerts R, Decamp C, Haut R. Biomechanical Comparison of an Interlocking Nail and a Plate-Rod Combination; An in vitro analysis in a canine fracture model. Vet Orthop Soc Meeting, 2004. Haut R, Jayaraman V, Sevensma E, Meyer E. Anterior Cruciate Ligament Rupture Due to Compression of the Human Tibiofemoral Joint. Orthop Res Soc, 2003. Kirsch J, Dejardin L, Meyer E, Decamp C, Haut R. Mechanical evaluation of canine pantarsal arthrodesis. Am College Vet Surg, 2003. 195 w N U E T A T S N A m H m M .- .3 -