Lateral extra-articular reconstruction length changes during weightbearing knee flexion and pivot shift: A simulation study

HAL Id: hal-02173085

https://hal.archives-ouvertes.fr/hal-02173085v2

Submitted on 27 Nov 2019

HAL

is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire

destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Lateral extra-articular reconstruction length changes during weightbearing knee flexion and pivot shift: A

simulation study

Yoann Blache, Biova Kouevidjin, Jacques de Guise, Raphaël Dumas, Adnan Saithna, Bertrand Sannery-Cottet, Mathieu Thaunat

To cite this version:

Yoann Blache, Biova Kouevidjin, Jacques de Guise, Raphaël Dumas, Adnan Saithna, et al.. Lateral

extra-articular reconstruction length changes during weightbearing knee flexion and pivot shift: A

simulation study. Orthopaedics and Traumatology - Surgery and Research, Elsevier, 2019, 105 (4),

pp. 661-667. �10.1016/j.otsr.2019.02.020�. �hal-02173085v2�

1

Length Change Behavior of Lateral Extra-Articular Reconstruction During

1

Weight-Bearing Knee Flexion and Simulated Pivot Shift

2 3

Blache Y

. (PhD), Kouevidjin B

. (MD), de Guise J

. (PhD), Dumas R

. (MD), Saithna A

4

(MD), Sonnery-Cottet B

. (MD), Thaunat M

(MD).

5 6

1

Univ Lyon, Université Claude Bernard Lyon 1 ; Laboratoire Interuniversitaire de Biologie 7

de la Motricité – EA 7424 8

2

Centre Orthopédique Santy- Group Ramsay-Générale de Santé - Hôpital privé Jean Mermoz, 9

Lyon, France 10

3

Laboratoire de recherche en imagerie et orthopédie (LIO), Centre de recherche du Centre 11

hospitalier de l'Université de Montréal (CRCHUM) , Montréal , Canada 12

4

Univ Lyon, Université Claude Bernard Lyon 1, IFSTTAR, LBMC UMR_T9406, F69622 13

Lyon, France 14

5

Southport & Ormskirk Hospitals, Lancashire, UK 15

6

Department of Clinical Engineering, University of Liverpool, Merseyside, UK 16

17

Corresponding author 18

Blache Yoann 19

Univ Lyon, Université Claude Bernard Lyon 1 20

Laboratoire Interuniversitaire de Biologie de la Motricité – EA 7424 21

27-29 bd du 11 novembre 1918, 69622 Villeurbanne Cedex 22

[email protected] 23

+33 4 72 23 48 28 24

25

BLACHE, Yoann, KOUEVIDJIN, Biova, DE GUISE, Jacques, DUMAS, Raphaël, SAITHNA, Adnan, SANNERY-COTTET, Bertrand, THAUNAT, Mathieu, 2019, Lateral extra-articular reconstruction length changes during weightbearing knee flexion and pivot shift: A simulation study, Orthopaedics & Traumatology: Surgery & Research, 105, 4, Elsevier, pp. 661-667, DOI: 10.1016/j.otsr.2019.02.020

2 Abstract

26 27 28

Background: Lateral extra-articular reconstruction (LER) length variation has been widely 29

investigated during knee flexion while its behavior during pivot shift scenario is still 30

unknown. Then, this study aimed to assess the length-change behavior of a LER during 31

physiological weight-bearing knee flexion in a normal knee and in a computer simulated pivot 32

shift scenario 33

Hypothesis: The hypothesis was that a femoral postero-proximal graft location allow a 34

desirable LER behavior both during weigth-bearing knee flexion and simulated pivot-shift.

35

Patients and Methods: A computer model was used to simulate weight-bearing knee flexion 36

and pivot-shift scenarios. LER length was calculated in both scenarios by the distance from 37

six femoral attachment sites (posterior and proximal to the lateral femoral epicondyle) and 38

two tibial locations (Gerdy’s tubercle [GT] and the anatomic attachment site of the 39

anterolateral ligament [ALL]).

40

Results: During physiological knee flexion, the LER lengthened up to 11mm, corresponding 41

to 22% of the resting length, in the early degrees of knee flexion and shortened from 40 to 60°

42

of knee flexion regardless of femoral or tibial attachment locations. However, the ALL tibial 43

position allowed complete LER shortening at 60° of flexion whereas an LER using a GT 44

tibial attachment remained more lengthened. For the movements corresponding to a pivot- 45

shift test, the LER lengthened between 10.6 and 15.9 mm for GT tibial attachment site and 46

between 5.9 and 11.6 mm for ALL attachment site.

47

Discussion: During physiological weight-bearing knee flexion, a proximo-posterior femoral 48

location was associated with a lengthened LER in the early degrees of flexion and a 49

shortening between 40 and 60° of flexion. In addition, LER lengthening occurred during a

50

3 simulated pivot-shift test supporting the concept that a postero-proximal femoral LER 51

attachment site presents a desirable behavior both during physiological knee flexion and 52

altered knee kinematics.

53 54

Keywords: Anterolateral ligament; anterior cruciate ligament; orthopedics; computer 55

modelling 56

57

58

4 1. Introduction

59

Return to sport after anterior cruciate ligament reconstruction is often not satisfying 60

since only 63% of patients resume their pre-injury level of sports participation [1] and even 61

then, this is associated with a high risk of re-injury [2-4]. One of the possible causes of these 62

suboptimal outcomes is a persistent lack of rotational and translational stability of the knee 63

even after a technically adequate anterior cruciate ligament reconstruction [5]. Consequently, 64

lateral extra-articular reconstruction (LER) has generated renewed interest [6-9] and is 65

increasingly performed in patients with high-grade pivot-shift, in order to better control the 66

coupled tibial internal rotation and anterior translation [10].

67

However, there has been considerable debate regarding the anatomy of the anterolateral 68

structures of the knee and therefore also the optimum positioning of femoral and tibial tunnels 69

for LER. These should be located such that the LER confers physiological knee kinematics, 70

specifically restoration of rotational control and avoidance of overconstraint [10]. The recent 71

literature, including in-vivo and in-vitro studies, has reported a range of different femoral 72

tunnel positions which includes directly proximal to the lateral epicondyle [9-11-12], postero- 73

proximal [13-14] or postero-distal [15]. However, there is increasing consensus that a postero- 74

proximal femoral attachment allows the most favorable behavior, with an LER that is 75

lengthened in knee extension and shortened in flexion [10-15]. With respect to the tibial graft 76

insertion, Wieser, Furnstahl, Carrillo, Fucentese and Vlachopoulos [12] reported that, during a 77

weight-bearing squat, the most isometric point was located at 37% of the postero-anterior 78

width of the tibial plateau.

79

Although all of these studies yield precious information for surgeons regarding the 80

optimum positioning of LER grafts during non-altered physiological knee flexion, none of 81

them has assessed the LER length changes that occur in the presence of the altered knee 82

kinematics seen in the knee with persistent internal rotation and anterior translation instability

83

5 following ACL reconstruction. To the best of our knowledge, only Imbert, Lutz, Daggett, 84

Niglis, Freychet, Dalmay and Sonnery-Cottet [14] investigated the effect of knee rotation on 85

LER length. Among the three femoral graft positions tested, the postero-proximal graft had 86

the greatest LER length variation when a 2 N.m internal rotation torque was applied with the 87

knee fixed at 90°, while no difference was observed between the graft positions at 20° of knee 88

flexion. However, this study was performed in-vitro and may not reflect LER length variation 89

during a physiological in-vivo weight-bearing task. Furthermore, LER length variations were 90

not measured during coupled knee internal rotation and anterior translation that corresponds 91

to the knee instability observed during a pivot-shift test.

92

The purpose of this study was to assess the length-change behavior of LER during a 93

physiological weight-bearing knee flexion task in a normal knee and in a computer simulated 94

pivot shift scenario. It was hypothesized that a proximo-posterior femoral attachment site 95

would allow graft lengthening in knee extension and shortening in flexion, and lengthening 96

during altered knee internal rotation and anterior translation.

97 98

2. Patients and Methods 99

2.1. Patients 100

The baseline model for the analysis was based on a healthy male participant (Age: 35 years, 101

Height: 1.75 m, Mass: 80 Kg) who volunteered to take part in the study. The patient declared 102

no history of major lower limb injury. The protocol was approved by the Institutional Review 103

Board of Hopital Privé Jean Mermoz (# 2017-02) in accordance with the ethical standards in 104

the 1964 Declaration of Helsinki, and the participant gave informed consent to participate in 105

the study.

106 107

2.2. Procedure and data collection

108

6 Firstly, a three-dimensional CT-scan of the participant right knee was performed (slice of 0.6 109

mm, Siemens Somaton, Erlangen, Germany). Secondly, in line with the methodology of 110

Clement, Dumas, Hagemeister and de Guise [16], the participant performed five quasi-static 111

squats with 0, 10, 20, 30 and 60° of knee flexion standardized with a positioning jig. Each 112

position was maintained inside the EOS

system (EOS Imaging Inc., Paris, France) for five 113

seconds in order to obtain biplane radiographic images.

114 115

2.3. Image processing and knee 3D kinematics 116

CT images were used to reconstruct the femur and tibia meshes using 3D Slicer 4.6.2

. 3D 117

meshes of the tibia and femur were also reconstructed from the five biplane radiographs [17- 118

18]. This reconstruction process gives an accuracy of the positions and orientations of the 119

femur and tibia of less than 0.3±0.3° and 0.3±0.2 mm respectively [18]. CT-scan meshes were 120

matched with the corresponded EOS meshes using an iterative closest point algorithm. In line 121

with ISB recommendations [19], 3D knee kinematics with six degrees of freedom were 122

obtained for the five quasi-static poses. Then, a cubic spline interpolation was used (Matlab 123

R2017b, The MathWorks Inc., Natick, MA) to compute the couplings between knee flexion 124

angle (from 0 to 60°) and the five other degrees of freedom, namely, abduction-adduction, 125

internal-external rotation, anterior-posterior translation, medio-lateral translation and cranio- 126

caudal translation.

127 128

2.4. LER attachment sites and length 129

A computer model (Opensim 3.3[20]) was used to assess LER length-changes with respect to 130

knee kinematics. The model was composed of the CT-scan meshes and the couplings between 131

the knee flexion angle and the five other degrees of freedom computed above from the EOS 132

quasi-static squat images. In order to avoid the ligaments penetrating through the bones

133

7 during the computer simulation, two wrapping objects were located on the femoral condyle 134

and tibia.

135

In accordance with the consensus from the anterolateral ligament expert group [10], femoral 136

graft positions located posteriorly and proximally to the lateral femoral epicondyle were 137

simulated. Specifically, six femoral graft insertions were located on two quarter-circles with a 138

radius of 5 and 10 mm, and centered on the femoral lateral epicondyle.

139

For the tibia, two locations were considered. The first intended to replicate procedures that 140

use a reflection of the iliotibial band, and was therefore located directly at its attachment to 141

Gerdy’s tubercle (GT). The second tibial graft located intended to replicate the most anterior 142

location of an anterolateral ligament (ALL) graft at 1 cm posterior to Gerdy’s tubercle [6]

143

(Figure 1). LER length-change behavior was determined by computing the distance between 144

the femoral and tibial insertions.

145 146

2.5. Analysis 147

Firstly, LER length changes were computed for both physiological weight-bearing knee 148

flexion with respect to the LER length computed for a resting position with the knee in full 149

extension (Eq. 1).

150

(Eq. 1)

with i, j, and k referring to a given knee flexion, internal rotation and anterior translation 151

respectively.

152

Secondly, LER length were computed for altered kinematics, namely, increased internal 153

rotation (up to 8°) and anterior translations (up to 12 mm) that can be observed in patients 154

[21]. In addition, average length variation of the LER was computed for two volumes 155

representative of potential knee kinematics during a simulated pivot-shift test. The first 156

volume corresponded to combined knee flexions between 10 and 30°, rotations between 1°

157

8 and 8°, and anterior translations between 1 mm and 6 mm. The second volume was computed 158

for the same knee flexions and rotations ranges but for anterior translations between 7 mm 159

and 12 mm. Finally, a specific combination of knee flexion, internal rotation and anterior 160

translation was also computed following the curve of the pivot-shift test reported by Amis, 161

Bull and Lie [5] (Figure 2).

162 163 164

3. Results 165

During the physiological, weight-bearing knee flexion task, knee flexion (from 0 to 60°) was 166

associated with a non-linear knee internal rotation ranging from 0 to 21.7°, and with an 167

anterior tibial translation ranging from 0 to 6.3 mm (Figure 2). In addition, regardless of the 168

femoral and tibial graft locations, the LER lengthened by up to 11mm (corresponding to 22%

169

of the surgery length) in the early degrees of knee flexion (between 0 and 40°) and shortened 170

from 40 to 60° of knee flexion (Figure 3). Furthermore, for all femoral attachment sites, the 171

posterior tibial graft (ALL position) led to less LER lengthening throughout knee flexion in 172

comparison to the GT tibial attachment site (- 7.72 ± 0.77 % on average).

173

In the case of simulated altered knee kinematics, an increased tibial internal rotation, 174

particularly after 20° of knee flexion led the LER to lengthen whatever the graft location. In 175

the same way, the LER lengthened with increased anterior tibial translation (Figure 4 and 176

supplementary file). For the specific volumes corresponding to a pivot-shift test (increased 177

internal rotation and anterior translation between -10 and -30° of knee flexion), regardless of 178

the femoral graft location, the LER lengthened between 20 and 34 % of the resting length 179

when a GT tibial attachment was used and between 11 and 26% when an ALL tibial 180

attachment was used (Table 1). In addition, regardless of the femoral graft location, the 181

anterior tibial position (GT) always lengthened more than the posterior one (ALL) for all

182

9 simulated knee kinematics. Finally, when replicating the specific knee kinematics recorded 183

for a pivot-shift test by Amis, Bull and Lie [5], the LER lengthened during the early phase of 184

the test (until 15° of knee flexion), with maximal values corresponding to the peaks of knee 185

anterior translation and internal rotation, regardless of the femoral graft location.

186 187

4. Discussion 188

This study aimed to assess the length-change behavior of LER during a physiological 189

weight-bearing knee flexion task (i.e. quasi-static squats) in a normal knee and in a computer 190

simulated pivot shift scenario. The main findings are that postero-proximal femoral 191

attachment sites result in LER lengthening in the early degrees of knee flexion and shortening 192

around 40-60° of knee flexion during physiological weight-bearing knee flexion and that 193

when a postero-proximal femoral attachment site is used, an LER becomes lengthened during 194

a simulated pivot shift test.

195

Although this study was based on a single participant, the kinematics recorded during 196

physiological weight-bearing knee flexions could be considered as representative of a healthy 197

male population. Normal kinematics during a weight-bearing squatting task are usually 198

characterized by a tibial internal rotation and anterior translation associated with knee flexion 199

[22]. In the current study the participant knee kinematics were consistent with these previous 200

reports since an internal rotation of between 0 and 20° and an anterior translation from 0 to 6 201

mm were demonstrated to be coupled with knee flexion between 0 and 60°.

202

The results of the current study are consistent with previous reports with respect to the 203

observation that proximo-posterior femoral attachment sites are associated with LER 204

lengthening in extension and shortening in physiological flexion [14-15]. In addition, 205

similarly to a previous study [15] we reported that an ALL tibial attachment location led to a 206

greater shortening close to 60° of knee flexion when compared to a GT tibial attachment.

207

10 Therefore, we may assume that an ALL graft location attachment site presents a more 208

desirable behavior than GT since it is less likely to overconstrain physiological internal 209

rotation in the flexed knee.

210

LER’s are performed in order to try to better control rotational and translational knee 211

stability [10]. The results of this study suggest that this function is achieved when a proximo- 212

posterior femoral graft location is used because the LER lengthened for both simulated 213

increased tibial internal rotation and for increased tibial anterior translation. These results are 214

in accordance with Imbert, Lutz, Daggett, Niglis, Freychet, Dalmay and Sonnery-Cottet [14]

215

who observed a lengthening of LER when an internal rotation torque was applied. The LER 216

also lengthened during a simulated pivot-shift test, namely, a combination of tibial internal 217

rotation and anterior translation. More precisely, the maximal lengthening was observed when 218

the maximal anterior translation and internal rotation occurred during the pivot-shift test when 219

simulated as described by Amis, Bull and Lie [5]. It therefore seems logical that LER may 220

help to restore normal knee kinematics especially when the knee is subjected to a combination 221

of tibial internal rotation and translation. This is consistent with the findings of Inderhaug, 222

Stephen, Williams and Amis [23], who demonstrated significant improvement in rotational 223

knee kinematics in cadaveric knees with LER. However, LER lengthening is only one 224

component of the complex inter-play of factors that ensure the control of knee stability. The 225

direction of LER vector should be assessed in future studies to address the assumption that it 226

is indeed the lengthening of the LER that improves knee stability. Nevertheless, the 227

lengthening behavior of LER observed in the current study reinforces the concept that LER 228

could be indicated after anterior-cruciate rupture when the patient presents a pivot-shift grade 229

2 or 3 [10] in order to better control knee stability.

230

Finally, in comparison to an ALL tibial LER graft position, the GT tibial attachment site 231

resulted in greater LER lengthening during simulated increased tibial internal rotation and

232

11 anterior translation. This raised concerns regarding the risk of overconstraint with a GT 233

attachment site particularly because this location was not associated with a complete 234

relaxation of the LER in flexion. Moreover, when examining the increased internal rotation, it 235

was observed that the LER was instantaneously lengthened, even in the absence of 236

pathological anterior translation. Therefore it is suggested that LERs based on a GT location 237

are associated with a greater potential risk of overconstraint [7] when compared to 238

reconstructions based on a more posterior tibial position such as that used for anatomic ALL 239

reconstruction.

240

There were several limitations of this study. The first was that only one participant was 241

enrolled in this study. Nevertheless, the analysis performed in our study explores a large range 242

of internal rotations and anterior translations and shall encompass the kinematic variability 243

that would have been observed in a larger cohort study. Similarly, it is assumed that the range 244

of femoral LER insertions encompasses the knee anatomy variability that would have been 245

observed in a larger cohort study. Finally, ours results obtained during the physiological knee 246

flexion are very similar than those of Kernkamp, Van de Velde, Hosseini, Tsai, Li, van Arkel 247

and Li [15] who performed their study on eighteen healthy knees. The second limitation was 248

that altered kinematics were simulated from ranges of motion observed in the literature, which 249

may differ from weight-bearing altered kinematics observed after anterior cruciate ligament 250

reconstruction. Once again, an assumption was made that the wide analysis encompassed the 251

variability encountered in post-operative knee kinematics. The third limitation was that only 252

two tibial insertions were tested, instead of a broader range. However, these were selected to 253

represent the two most popular tibial locations in contemporary surgical practice [6-7].

254 255

5. Conclusion

256

12 To conclude, computer modelling made it possible to analyze LER length variations for 257

numerous knee positions corresponding to weight-bearing knee flexion, and for the first time, 258

to typical pivot-shift kinematics. During physiological weight-bearing knee flexion, and 259

regardless of the tibial attachment site, a proximo-posterior femoral location was associated 260

with a lengthened LER in the early degrees of flexion and a shortening between 40 and 60° of 261

flexion. In addition, LER lengthening occurred during simulated pivot-shift scenarios 262

supporting the concept that a postero-proximal femoral LER attachment site presents a 263

desirable behavior both during physiological knee flexion and altered knee kinematics.

264 265

Acknowledgment:

266

We would like to thank Thierry Cresson from the « Laboratoire de recherche en imagerie et 267

orthopédie, École de technologie supérieure, Centre de recherche du Centre hospitalier de 268

l'Université de Montréal »,for his contribution concerning the mesh reconstructions from the 269

biplanar radiographies. In addition, this work was performed within the framework of the 270

LABEX PRIMES (ANR-11-LABX-0063) of Université de Lyon, within the program 271

"Investissements d'Avenir" (ANR-11-IDEX-0007) operated by the French National Research 272

Agency (ANR).

273 274

Funding 275

The study was supported by the group “Ramsey Générale de Santé”, Paris, France.

276 277

Author’s Contributions 278

Y.B. devised the study idea; treated and analysed the data; drafted of the manuscript; read and 279

approved the final manuscript. B.K, R.D. and J.deG. performed the experiment; treated the 280

data; read and approved the final manuscript. A.S., B.S. and M.T. analyzed the data; helped to 281

draft the manuscript; read and approved the final manuscript.

282

13 283

Disclosure of interest 284

Y.B, R.D., B.K. and J.deG. declare no conflict of Interest. M.T., B.S. and A.S. are consultant 285

for Arthrex.

286 287 288

References 289

1. Ardern CL, Webster KE, Taylor NF, et al. Return to sport following anterior cruciate 290

ligament reconstruction surgery: A systematic review and meta-analysis of the state of 291

play. Br J Sports Med 2011; 45:596-606.

292

2. Paterno MV, Rauh MJ, Schmitt LC, et al. Incidence of second acl injuries 2 years after 293

primary acl reconstruction and return to sport. Am J Sports Med 2014; 42:1567-73.

294

3. Salmon L, Russell V, Musgrove T, et al. Incidence and risk factors for graft rupture and 295

contralateral rupture after anterior cruciate ligament reconstruction. Arthroscopy 2005;

296

21:948-57.

297

4. Shelbourne KD, Gray T and Haro M. Incidence of subsequent injury to either knee within 5 298

years after anterior cruciate ligament reconstruction with patellar tendon autograft. Am J 299

Sports Med 2009; 37:246-51.

300

5. Amis AA, Bull AMJ and Lie DTT. Biomechanics of rotational instability and anatomic 301

anterior cruciate ligament reconstruction. Oper Tech Orthop 2005; 15:29-35.

302

6. Sonnery-Cottet B, Daggett M, Helito CP, et al. Combined anterior cruciate ligament and 303

anterolateral ligament reconstruction. Arthrosc Tech 2016; 5:e1253-e9.

304

7. Slette EL, Mikula JD, Schon JM, et al. Biomechanical results of lateral extra-articular 305

tenodesis procedures of the knee: A systematic review. Arthroscopy 2016; 32:2592-611.

306

8. Marcacci M, Zaffagnini S, Marcheggiani Muccioli GM, et al. Arthroscopic intra- and extra- 307

articular anterior cruciate ligament reconstruction with gracilis and semitendinosus 308

tendons: A review. Curr Rev Musculoskelet Med 2011; 4:73-7.

309

9. Krackow KA and Brooks RL. Optimization of knee ligament position for lateral 310

extraarticular reconstruction. Am J Sports Med 1983; 11:293-302.

311

14 10. Sonnery-Cottet B, Daggett M, Fayard JM, et al. Anterolateral ligament expert group 312

consensus paper on the management of internal rotation and instability of the anterior 313

cruciate ligament - deficient knee. J Orthop Traumatol 2017; 18:91-106.

314

11. Kittl C, Halewood C, Stephen JM, et al. Length change patterns in the lateral extra- 315

articular structures of the knee and related reconstructions. Am J Sports Med 2015;

316

43:354-62.

317

12. Wieser K, Furnstahl P, Carrillo F, et al. Assessment of the isometry of the anterolateral 318

ligament in a 3-dimensional weight-bearing computed tomography simulation.

319

Arthroscopy 2017; 33:1016-23.

320

13. Van de Velde SK, Kernkamp WA, Hosseini A, et al. In vivo length changes of the 321

anterolateral ligament and related extra-articular reconstructions. Am J Sports Med 2016;

322

44:2557-62.

323

14. Imbert P, Lutz C, Daggett M, et al. Isometric characteristics of the anterolateral ligament 324

of the knee: A cadaveric navigation study. Arthroscopy 2016; 32:2017-24.

325

15. Kernkamp WA, Van de Velde SK, Hosseini A, et al. In vivo anterolateral ligament length 326

change in the healthy knee during functional activities-a combined magnetic resonance 327

and dual fluoroscopic imaging analysis. Arthroscopy 2017; 33:133-9.

328

16. Clement J, Dumas R, Hagemeister N, et al. Soft tissue artifact compensation in knee 329

kinematics by multi-body optimization: Performance of subject-specific knee joint 330

models. J Biomech 2015; 48:3796-802.

331

17. Chaibi Y, Cresson T, Aubert B, et al. Fast 3d reconstruction of the lower limb using a 332

parametric model and statistical inferences and clinical measurements calculation from 333

biplanar x-rays. Comput Methods Biomech Biomed Engin 2012; 15:457-66.

334

18. Kanhonou M, Cresson T, Lavoie F, et al. A method to study 3d knee pseudo-kinematics 335

using low-dose stereoradiography during static squat. Comput Methods Biomech Biomed 336

Engin 2014; 17 Suppl 1:138-9.

337

19. Wu G, Siegler S, Allard P, et al. Isb recommendation on definitions of joint coordinate 338

system of various joints for the reporting of human joint motion--part i: Ankle, hip, and 339

spine. International society of biomechanics. J Biomech 2002; 35:543-8.

340

20. Delp SL, Anderson FC, Arnold AS, et al. Opensim: Open-source software to create and 341

analyze dynamic simulations of movement. IEEE Trans Biomed Eng 2007; 54:1940-50.

342

21. Bull AM, Earnshaw PH, Smith A, et al. Intraoperative measurement of knee kinematics in 343

reconstruction of the anterior cruciate ligament. J Bone Joint Surg Br 2002; 84:1075-81.

344

15 22. Yamaguchi S, Gamada K, Sasho T, et al. In vivo kinematics of anterior cruciate ligament 345

deficient knees during pivot and squat activities. Clin Biomech 2009; 24:71-6.

346

23. Inderhaug E, Stephen JM, Williams A, et al. Anterolateral tenodesis or anterolateral 347

ligament complex reconstruction: Effect of flexion angle at graft fixation when combined 348

with acl reconstruction. Am J Sports Med 2017; 363546517724422.

349

350

16 Table 1. Mean ± standard deviation of the lateral extra-articular reconstruction (LER) length 351

variation with respect to the LER length measured at full knee extension (expressed in 352

absolute and relative values) for the specific volumes corresponding to the pivot-shift 353

kinematics. Results are reported for the six simulated femoral graft (#1 to #6) and two 354

tibial graft (GT and ALL) locations.

355

LER femoral graft

Altered kinematics #1 #2 #3 #4 #5 #6

+1 to 8° and +1 to 6 mm

LER

Δ length (mm) 10.8±1.2 11.1±1.3 11.3±1.3 10.6±1.2 11.2±1.3 11.3±1.4 LER

Δ length (mm) 6.0±1.3 6.6±1.1 6.9±1.1 5.9±0.9 7.0±1.1 7.5±1.2 LER

Δ length (%) 22±2 23±3 25±3 20±2 21±2 23±3 LER

Δ length (%) 13±2 14±2 16±3 11±2 14±2 17±3 +1 to 8° and +7 to 12 mm

LER

Δ length (mm) 14.8±1.3 15.4±1.4 15.7±1.4 14.5±1.2 15.6±1.4 15.9±1.5 LER

Δ length (mm) 9.3±1.1 10.2±1.2 10.7±1.3 9.1±1.1 10.8±1.3 11.6±1.3 LER

Δ length (%) 30±3 32±3 34±3 27±2 29±3 33±3 LER

Δ length (%) 20±2 22±3 25±3 17±2 22±2 26±3 356

357

358

17 Figure captions

359 360

Figure 1. Lateral view of the right knee mesh with the femoral and tibial attachment sites of 361

the simulated lateral extra-articular reconstruction. Femur: Simulated femoral graft positions 362

at a radius of 5mm (blue dots) and 10mm (orange dots) from the lateral epicondyle (white 363

dot). Tibia: ALL (anterolateral ligament position), GT (Gerdy’s tubercle) 364

365

Figure 2 – At the top, changes in tibial internal rotation for the physiological weight-bearing 366

knee flexion task (solid blue line) and for the simulated exaggerated tibial internal rotation 367

(shaded blue zone). At the bottom, anterior tibial translation with respect to knee flexion for 368

the physiological weight-bearing knee flexion task (solid red line) and for a simulation of an 369

exaggerated tibial anterior translation (shaded red zone). The blue and red dashed lines 370

correspond to the tibial internal rotation and anterior translation measured during a simulated 371

pivot shift test according to Amis, Bull and Lie [5]. The black dashed arrow refers to the 372

direction of the movement during the pivot-shift test.

373

Figure 3- Relative (left) and absolute (right) lateral extra-articular reconstruction (LER) 374

length variation during the physiological weight-bearing knee flexion task for the six femoral 375

attachment sites (#1 to #6) and the two tibial graft locations (GT and ALL).

376 377

Figure 4 – Length variation of LER with regard to knee flexion and internal rotation at three 378

anterior tibial translations (0 mm, 6 mm and 12 mm) for three femoral attachment sites (#4, 379

#5 and #6) and two tibial locations (GT and ALL). The inferior trajectory surrounded by 380

black circles represent the coupling between knee flexion and internal rotation during weight- 381

bearing knee flexion task.

382

383

384

385