Original Article

Pre-Clinical Assessment of Total Knee Replacement Anterior-

Posterior Constraint

Halewood, C.a*, Athwal, K.K.a*, Amis, A.A.a,b

aDepartment of Mechanical Engineering, Imperial College London, South Kensington

Campus, London SW7 2AZ, UK

and b Musculoskeletal Surgery group, Department of Surgery and Cancer, Imperial

College London School of Medicine, Charing Cross Hospital, London, W6 8RF, UK

* Credited as co-first authors, as equal contribution to the manuscript

Corresponding author:

Dr Kiron Athwal,

Department of Mechanical Engineering,

Imperial College London,

South Kensington Campus,

London SW7 2AZ, UK

Tel: +44 (0)20 7594 1986

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

k.athwal12@imperial.ac.uk

Keywords: TKR, TKA, Constraint, Laxity

Word count (Intro-Discussion): 3420

2

21

22

23

24

Abstract

Pre-clinical, bench-top assessment of Total Knee Replacements (TKR) can provide

information about the inherent constraint provided by a TKR, which does not depend on

the condition of the patient undergoing the arthroplasty. However little guidance is given by

the ASTM standard on test configurations such as medial-lateral (M:L) loading distribution,

flexion angle or restriction of secondary motions. Using a purpose built rig for a materials

testing machine, four TKRs currently in widespread clinical use, including medial-pivot and

symmetrical condyle types, were tested for anterior-posterior translational constraint.

Compressive joint loads from 710 to 2000 N, and a range of medial-lateral (M:L) load

distributions, from 70:30% to 30:70% M:L, were applied at different flexion angles with

secondary motions unconstrained. It was found that TKA constraint was significantly less

at 60 and 90° flexion than at 0°, whilst increasing the compressive joint load increased the

force required to translate the tibia to limits of AP constraint at all flexion angles tested.

Additionally when M:L load distribution was shifted medially, a coupled internal rotation

was observed with anterior translation and external rotation with posterior translation. This

paper includes some recommendations for future development of pre-clinical testing

methods.

(193w)

3

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

1. Introduction

According to the UK National Joint Registry (NJR, 2016), 59 different designs of Total

Knee Replacement (TKR) were implanted into patients in 2015. Post-operative data may

be available for established replacements but there is a lack of information about how a

new device is likely to perform clinically (Liow and Murray, 1997). The ASTM standard

tests F1223 (ASTM-F1223, 2014) measure the inherent constraint of the TKR prosthesis

itself, that which is independent of the patient’s physiological condition or the surgical

implantation process. The ASTM standard describes test guidelines for determining

constraint in anterior-posterior (AP) drawer, medial-lateral shear, internal-external and

varus-valgus rotations, and in distraction. This information may help the surgeon in

choosing the most appropriate TKR for each patient, depending on factors such as the

intrinsic stability of the native knee which is affected by the condition of the soft tissues

surrounding it (Kakarlapudi and Bickerstaff, 2000). The ASTM-F1223 (2014) standard

aims to: “provide a database of product functionality capabilities …... that is hoped will aid

the physician in making a more informed total knee replacement (TKR) selection”. In the

European Union, this testing is mandatory for all new TKRs before they are marketed and

used clinically if CE marking is required by the manufacturer (European Parliament, 2007).

Haider and Walker (2005) used the test methods outlined in the 2005 version of the

standard to assess the constraint of three designs of TKR. Moran et al. (2008) assessed

one TKR device experimentally in order to validate a computer simulation of the ASTM test

methods. These studies looked at TKR AP constraint, but did not consider the effect on

constraint of the medial: lateral (M:L) tibiofemoral loading distribution, which varies

depending on subject and activity (Mündermann et al., 2008; Varadarajan et al., 2008;

Zhao et al., 2007). The ASTM standard itself does not include guidance on this M:L

4

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

loading distribution, therefore experiments attempting to replicate this standard would likely

assume a 50:50 M:L axial load.

Haider and Walker (2005) explored whether keeping secondary motions restricted during

translation tests, as suggested by the ASTM standard, led to anomalous results. They

concluded that, other than flexion angle and the degree of freedom (DoF) being measured,

all the other motions should be left unrestricted in order to obtain reliable results. Heim et

al. (1996 and 2001) looked at AP constraint of mobile bearing and posterior stabilised

TKRs but restricted all the DoF of motion other than the one being measured. That

restriction could be expected to lead to unrealistic edge-loading conditions when there is

displacement between the components of an asymmetrical TKR, for example. In a

symmetrical TKR with equal M:L loading, we would expect minimal coupled rotation with

AP drawer. However a shift of the resultant force either medially or laterally would create

friction-induced constraint on the more-loaded compartment and more displacement on

the less-loaded compartment, which overall manifests as a coupled internal-external

rotation.

The objective of this study was to assess how AP displacement outcomes from the ASTM-

F1223 standard for measuring AP constraint in TKRs were affected by unrestricting

coupled rotations, and varying the M:L loading distribution, axial load and flexion angle. It

was expected that altering the load distribution medially would cause a coupled internal

rotation and external rotation of the tibia when displaced anteriorly and posteriorly

respectively, with the opposite occurring with a more lateral load distribution. When

considering different flexion angles, it was hypothesised that most AP constraint would be

shown at full extension, with an increase in laxity exhibited with flexion. Furthermore,

increasing the compressive joint load was hypothesised to increase the displacing AP

force required to reach the translation limits of TKR.

5

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

2. Materials and Methods

The tests described for this study follow the general standard as set out in ASTM-F1223

(2014). Any changes to this standard have been highlighted in the methods.

2.1 Test rig set-up

A single-axis, screw-driven Instron model 5565 materials testing machine was employed

for the constraint tests. A test rig was designed and constructed, which could

accommodate the femoral and tibial components of a TKR (Figure 1). The femoral

component was mounted using polymethylmethacrylate (PMMA) bone cement onto an

aluminium alloy cross-bar shaped to match the component’s internal geometry, similar to

the shape of the distal femur as prepared during surgery. The flexion angle could be

adjusted by rotating and then fixing the cross-bar into position. A pivoting frame was used

so that the femoral component was free to rotate in varus-valgus, about an anterior-

posterior axis at the level of the flexion axis, not far from the joint line. The pivot point could

be adjusted medially-laterally, in order to vary the load distribution between the medial and

lateral compartments of the TKR, across the range 30:70% to 70:30% M:L. The pivot

frame was mounted on linear bearings, which allowed it to translate proximally-distally. A

calibrated pneumatic cylinder forced the femoral mounting distally, against the tibial

component, thus providing the compressive joint force. Being a pneumatic cylinder, it did

not prevent secondary proximal translations occurring when the TKR was tested.

The tibial components were mounted into the end of a freely-rotating shaft, which allowed

internal-external rotation. The posterior slopes of the tibial components were set at 0° in

this study. This assembly was mounted onto a linear bearing which allowed free medial-

lateral translation. The whole tibial assembly was then mounted on another linear bearing,

which allowed anterior-posterior translation. This was attached directly to the load cell on

6

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

the moving cross-head of the Instron, which provided the AP motion and measured both

force (N) and translation (mm).

Thus, the Instron machine imposed AP translation of the TKR at a chosen angle of flexion

and M:L load distribution, while all other degrees-of-freedom were unrestricted. This varies

from ASTM-F1223, which restricts movement other than AP translation and does not

suggest M:L variation. ASTM-F2083 (2012) recommends testing implants at 0°, 15°, 90°

and maximum flexion; in this study 0°, 30°, 60° and 90° flexion were chosen to fully

explore extension to deep flexion at equal increments and a ‘true maximal flexion’ was not

be tested because of the difficulty of relating it to the clinical situation.

2.2 Implants and tests:

Four TKRs were tested. AP constraint tests were conducted on two MatOrtho TKRs

(MatOrtho, Leatherhead, UK): the Medial Rotation Knee (MRK) which had been in clinical

use for over twenty years; and a newer design, the Saiph Knee. Both of these devices

were medial-sphere, highly congruent posterior cruciate ligament (PCL)-sacrificing type

TKRs, with asymmetrical condylar geometry. AP constraint tests, at a range of

compressive loads and ML loading distributions, were also conducted using the

conventionally designed, PCL-retaining Stryker Triathlon (Stryker (UK) Ltd, Newbury, UK)

and Smith & Nephew Legion (Smith & Nephew, Memphis, TN, USA), which both had

symmetrical condylar geometries. The schedule of tests conducted is shown in Table 1.

2.3 Test method:

ASTM-F1223 determines the neutral position by either ‘applying a compressive force of

100 N and allowing the implant to settle or by measuring the vertical position of the

movable component with respect to the stationary and using the low point of the

component as the neutral point’. Pilot testing deemed this not to be a repeatable method

with which to find a neutral position with differing geometries, therefore a different method

7

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

at a higher load was proposed. The femoral component was fixed at the desired flexion

angle and required M:L load distribution, then AP drawer was imposed by the Instron. To

first approximate the neutral AP position, a 350 N axial compressive load was applied and

the AP position was adjusted until the femoral components sat in the lowest compressive

point on the concave tibial surface. Small AP translations of ±3 mm were applied to the

tibia and the neutral AP position was adjusted until the hysteresis loop of the force versus

displacement graph was symmetrical above and below the zero load axis (Figure 2).

8

143

144

145

146

147

148

149

Once the AP position was adjusted, a 710 N compressive axial load was then applied and

the tibial component was anteriorly translated at a speed of 0.1 mm/sec until, as the ASTM

stipulates, ‘dislocation of the components is imminent… or if a dangerous or unrealistic

situation is about to occur’. In this experiment, the anterior limit was defined as the point at

which the force-displacement graph started to plateau (Figure 2); this suitable limit was

chosen by the authors to avoid permanent deformation of the edge of the UHMWPE

bearing, which would have affected the results of future tests using the bearing. This

displacement limit was recorded and the process was then repeated in the posterior

direction using the same procedure. The TKR was returned to the neutral position,

lubricated with water, reloaded to 710 N and cycled between the translation limits found

previously at a speed of 1 mm/sec (ASTM-F1223 states not to exceed 10 mm/sec). Three

“pre-conditioning” cycles were completed and data were collected on the fourth cycle. For

both the Triathlon and Legion implants, the axial load was increased to 2000 N and four

further cycles were performed at the same AP limits as the 710 N test (this higher load is

not included in ASTM-F1223).

2.4 Statistical Analysis:

For the Legion TKR, four nominally identical samples were tested. The following statistical

analyses were performed for the Legion TKR in SPSS 24 (IBM SPSS Statistics, version

22, Armonk, NY):

1. One-way repeated-measures analysis of variance (RM-ANOVA) at each M:L load

distribution to compare AP translations across different flexion angles.

2. Two-way RM-ANOVA to compare AP force at the translation limits across different axial

loads (710 N or 2000 N) and across different flexion angles.

9

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

Post-hoc paired t-tests with Bonferroni correction were applied when differences were

found, with significance level set at p < 0.05.

3. Results

The AP constraint characteristics varied considerably between the four different TKR

designs. The AP force-displacement graphs for the tests at full extension (here defined as

0° flexion) for 50:50 M:L loading distribution are shown in Figure 3. Constraint was

calculated as displacing force (N) per mm of translation per Newton of compressive load

(Haider and Walker, 2005), which was made after taking into account the frictional forces

present in the rig and implant articulation, as evident in the hysteresis curves.

The results were consistent with the geometry of the tibial bearings and the congruence of

the devices. The relatively incongruent Triathlon and Legion, which had shallow tibial

bearing concavities, were less constrained in AP translation than the MRK and Saiph, with

a shallow sloped force-displacement curve.

At full extension and 50:50 M:L loading distribution, total AP laxity ranged from 8.5 mm

with the MRK to 15 mm with the Triathlon (Table 2). At 90° flexion, the MRK and Saiph

again showed similar constraint characteristics to each other, while the Triathlon and

Legion allowed larger ranges of AP laxity, 22 mm and 24 ± 1 mm (average ± 95%

confidence interval) respectively. Constraint at full extension varied from 0.01 mm-1 with

the Triathlon to a maximum of 0.19 mm-1 with the MRK design in anterior tibial drawer. In

posterior drawer, the two medial-sphere type TKRs showed similar high levels of

constraint, while the Triathlon and Legion exhibited much less (Table 2).

3.1 Varying flexion angle

The total AP laxity of the Legion implant (at 50:50 M:L distribution and 710 N axial load)

ranged from 13.6 ± 1.0 mm at full extension to 24.3 ± 0.8 mm at 90° (Figure 4). The load-

10

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

displacement graphs at 30° and 60° were similar, and at 90° the most posterior translation

was demonstrated. An overall observation was that the articulations of the TKRs were

usually most-congruent near full extension, and progressively disengaged with flexion, and

this was reflected by reduced constraint as the knee flexed. When the M:L load distribution

was 60:40, anterior and posterior translations at 60° and 90° were significantly larger than

at 0° (all with p<0.05). At 50:50, increase in AP translation at 60° and 90° were also found,

with an additional significance found in posterior translation at 30° (p=0.008).

3.2 Varying Axial load

Two compressive loads were tested: 710 N and 2000 N. At all flexion angles tested with

the Legion TKR, increasing the axial load resulted in a higher AP force reached at the

displacement limits (all with p<0.05). For example at 0° flexion the displacing force

required for 6.7 ± 1.4 mm anterior translation of the Legion implant increased from 248 ±

66 N to 472 ± 87 N when axially loaded to 710 N and 2000 N respectively (p<0.01, Figure

5). For 6.9 ± 2.4 mm posterior drawer, the displacing forces increased from 320 ± 83 N to

635 ± 137 N (p<0.01). Similarly with the Triathlon implant, the displacing force for given AP

translation at 0° and 90° increased significantly (p<0.05) with axial load.

3.3 Varying M:L load distribution

Changing the M:L loading distribution from 50:50 to 60:40 increased the total AP laxity for

both the MRK and Saiph devices, particularly at 90° flexion. This increase in total laxity

was due to increased tibial rotation.

The medial-sphere type TKRs exhibited more coupled tibial IE rotation in AP translation at

the 50:50 M:L loading distribution than the symmetrical Triathlon and Legion, which

displayed almost 0° rotation for translations in both directions at full extension and 90°

flexion. Total tibial rotation at both 0° and 90° flexion for both the Saiph and MRK was 11°

at the 50:50 M:L loading condition. When shifted towards the medial side to give a 60:40

11

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

loading condition, an increased amount of tibial rotation was observed with both the MRK

and the Saiph at 90° flexion, with total rotation increasing to 15° and 19° respectively. This

secondary tibial rotation in the MRK resulted from greater movement in the less

constrained lateral compartment. At full extension, rotation only increased for the Saiph

TKR, this time to 18°. It should also be noted that the direction of the majority of the

rotation varied between the two devices. For the Saiph, most of the coupled rotation

occurred internally during anterior tibial drawer but with the MRK, it was external during

posterior tibial drawer.

At different M:L loading distributions from 30:70 to 70:30, both the Triathlon and Legion

implants exhibited differences in both magnitude and direction in associated tibial rotation

(Figure 6 and 7). As the loading was shifted further towards the medial condyle, there was

an increase in internal rotation coupled with anterior translation and external rotation

coupled with posterior translation. Conversely, with increasing lateral loading distribution

there was an increase in external rotation with anterior translation and internal rotation with

posterior translation.

3.4 Reproducibility of results

Four samples of the Legion knees were tested by the same investigator in order to assess

intra-rater repeatability of the results between nominally identical devices. At 60:40 M:L

loading the 95% confidence intervals (CI) from the total AP translation results at 0° and

90° flexion were 0.9 mm and 0.9 mm respectively; at 50:50 M:L loading this was 1.6 mm

and 1.3 mm respectively. The constraint, in both anterior and posterior translation, had a

95% CI of ± 0.01 mm-1.

12

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

4. Discussion

Four TKR designs were tested for AP constraint at different flexion angles, M:L load

distributions and compressive loads. Increasing the axial compressive load on the TKRs

increased the friction between femoral and tibial components, and thus increased AP

displacing forces as hypothesised. Maximal AP constraint occurred at full extension, with

increasing translations at larger flexion angles. The coupled tibial IE rotations, in response

to AP translations, were found to be sensitive to the M:L load distribution in the

symmetrical TKR designs: as the load was shifted medially there was a coupled internal

rotation and external rotation with anterior and posterior translations respectively, which

reversed with increasing lateral load distribution. With increasing load shift in either medial

or lateral directions, the magnitude of coupled rotation increased. Mündermann et al.

(2008) have shown that the resultant joint force may oscillate between the medial and

lateral compartments during daily activities. Taken together with our results, it implies that

secondary rotational instabilities may occur.

The focus of the study was not to compare AP constraints of TKRs from different

manufacturers in order to evaluate the best performance, but rather to assess the

sensitivity of the ASTM standard to different variations and unrestricted coupled rotations.

However, by testing different TKRs designed to produce distinct kinematic behaviour (such

as asymmetrical designs), it was demonstrated that these tests are an intrinsic measure of

the device without the influences of soft tissues. The ASTM standard testing is mandatory

for all new TKRs in order for the device to be CE marked; therefore every manufacturer

has these constraint data. Unfortunately there is almost no data publicly available in peer-

reviewed journal articles. Therefore this is a lost opportunity to learn about the effect of the

constraint characteristics on patient outcome.

13

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

These results can be used by a surgeon alongside patient specifics, such as the condition

of the PCL, in selecting the correct device. If these results are considered in isolation and

compared to data from normal knees (Amis, 1989; Shino et al., 1987) the Saiph was

closest to physiological, the MRK was more constrained and the Triathlon was less

constrained, which is in agreement with the designs of the devices and the way in which

they are intended to be used clinically.

Constraint results such as these may also be viewed with potential implant loosening in

mind; they suggest that a more constrained implant has a greater ability to transfer shear

forces to the underlying bone, and this may have implications for the efficacy of the fixation

of the device. It is difficult to scrutinise this theory any further from large datasets such as

joint registries, because although they contain failure mechanisms for all primary TKRs

that need revising, and revision rates for individual TKR designs, they do not combine the

two.

These bench-top tests could provide research benefits outside of pre-clinical use. The

tests can be used to validate computer modelling approaches (Rullkoetter et al., 2017),

although fidelity of the soft-tissue and knee geometry is still debated. The current rig

design can test implanted cadaveric specimens under the same loading regimes; thus the

data presented here may provide comparative information on how much stability is

provided by the inherent implant geometries as opposed to soft-tissues. The data may also

be used to establish protocols for robotic testing systems that could apply more complex

and clinically relevant 6 DoF movements and loads.

Bench-top tests are relatively quick and cheap to perform and producing easy to interpret

data. There was inherent variation in implant cementing and alignment and determining

the neutral position and subluxation points of the TKRs; despite this the repeat tests with

Legion implants demonstrated good intra-rater reproducibility within the same implant

14

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

design. However there are limitations with this approach to pre-clinical testing. The test

simulates a very specific motion of the tibial component relative to the femur which,

although representing a clinical laxity assessment (Daniel et al., 1985), is unlikely to be

representative in everyday movements. The constraint test described here should continue

to be used as a “common sense” check for new designs of implant, but care must be taken

when making functional inferences from constraint testing results. A limitation of the set-up

was that water was used for lubrication, as specified by the ASTM: it may not represent

the frictional behaviour of synovial fluid, but is convenient and reproducible.

4.1 Recommendations:

The authors recommend the following on the basis of the findings in this study:

1) The test protocol defined by ASTM F1223 should stipulate the variation of the M:L

loading distribution. It has been shown to vary between patients and activities

(Mündermann et al., 2008; Varadarajan et al., 2008; Zhao et al., 2007) and the AP

constraint and associated secondary motions have been shown to be very sensitive to

this distribution, particularly with a less-congruent device such as the Triathlon.

2) The secondary motions observed during AP constraint tests should be measured and

recorded as part of the ASTM standard, as these can provide more information about

the device’s stability characteristics.

3) The tests could be extended to include a higher axial load such as 2000 N

(Mündermann et al., 2008), approximately three times body weight, in order to

investigate coupled rotations and M:L distribution effects whilst under normal walking

gait loads.

4) The tests could be extended to include in-vitro cadaveric assessment to establish

whether or not the soft tissues interact with the TKR in the way that is intended.

15

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

Acknowledgements

This work was funded by MatOrtho Ltd. Part of the study design was in collaboration with

MatOrtho, but the sponsors had no part in the work or the preparation and submission of

this paper. The Instron machine was provided by the Arthritis Research UK charity. The

authors would like to thank Peter Barnado, Tamsin Bromley, Jennifer Lenz and Matthew

Taylor for designing and manufacturing the rig.

16

317

318

319

320

321

322

Tables

Table 1 Test Schedule (cross indicates a completed test)

Implant Flexion Angles (°)

Axial Load (N) Loading (Medial:Lateral)

70:30 60:40 50:5

040:60 30:70

MRK0 710 X X

90 710 X X

Saiph0 710 X X

90 710 X X

Triathlon

0710 X X X X X

2000 X X X X X

90710 X X X X X

2000 X X X X X

Legion

0710 X X X X X

2000 X X X X X

30710 X X X X X

2000 X X X X X

60710 X X X X X

2000 X X X X X

90710 X X X X X

2000 X X X X X

Table 2 Laxity (mm) and constraint (mm-1) for 50:50 medial: lateral load distribution.

For the Legion, the values given are the average ± 95% confidence interval (n = 4).

Flexion Angle (°) Measurement

Implant Type

MRK Saiph Triathlon Legion

0 Total AP laxity (mm) 8.5 10.1 15.0 13.6 ± 1.6

17

323

324

325

326

327

Anterior constraint (mm-1) 0.19 0.07 0.01 0.05 ± 0.01

Posterior constraint (mm-1) 0.13 0.12 0.05 0.07 ± 0.01

90 Total AP laxity (mm) 10.5 8.7 22.0 24.3 ± 1.3

18

Figure legends

Figure 1 The anterior-posterior translation testing rig. Inset: a close up image of the

implants in the testing rig (at 30° flexion).

19

328

329

330

331

332

Figure 2 Graphs showing the constraint of the Legion implant (90° flexion, 50:50 medial:

lateral load distribution, 710 N axial load). a) An approximate neutral position was first

found, and a ± 3mm cyclic draw showed that the hysteresis loop was not symmetrical

above and below the zero load axis. b) The neutral position of the tibia was moved 0.5 mm

anteriorly, and a new ± 3mm draw showed that the hysteresis loop was now symmetrical.

c) The tibia was translated anteriorly and posteriorly until the force-displacement graph

peaked and started to drop. d) The implant was cyclically translated between these two

limits four times.

20

333

334

335

336

337

338

339

340

341

342

Figure 3 Constraint of different implants (0° flexion, 50:50 medial: lateral load distribution,

710 N axial load).

21

343

344

345

346

Figure 4 Constraint of the Legion implant at different angles of flexion (50:50 medial:

lateral load distribution, 710N axial load).

22

347

348

349

350

Figure 5 Constraint of the Legion implant under 710N and 2000N axial loads (50:50

medial: lateral load distribution, 0° and 90° flexion angles).

23

351

352

353

354

Figure 6 Coupled tibial rotation during anterior-posterior translation of the Triathlon implant

at different medial: lateral load distributions (710 N axial load, 90° flexion).

24

355

356

357

358

Figure 7 Peak coupled tibial rotations of the Legion implant during anterior (left) and

posterior (right) translation at different medial: lateral load distributions and flexion angles

(710 N axial load).

25

359

360

361

362

363

References

Amis, A. 1989. Anterior cruciate ligament replacement. Knee stability and the effects of

implants. Journal of Bone & Joint Surgery - British Volume, 71-B, 819-824.

ASTM-F1223, 2014. Standard Test Method for Determination of Total Knee Replacement

Constraint. ASTM International, West Conshohocken, PA.

ASTM-F2083, 2012. Standard Specification for Knee Replacement Prosthesis. ASTM

International, West Conshocken, PA.

Daniel, D.M., Malcom, L.L., Losse, G., Stone, M.L., Sachs, R., Burks, R. 1985.

Instrumented measurement of anterior laxity of the knee. Journal of Bone and Joint

Surgery - American Volume, 67A, 720-726.

European Parliament 2007. European Parliament And The Council Of The European

Union 2007/47/EC, "Directive 2007/47/EC of the European Parliament and of the Council".

Official Journal of the European Union, http://ec.europa.eu/consumers/sectors/medical-

devices/files/revision_docs/2007-2047-en_en.pdf.

Haider, H., Walker, P.S. 2005. Measurements of constraint of total knee replacement.

Journal of Biomechanics, 38, 341-348.

Heim, C., Postak, P., Greenwald, A., 1996. Stability characteristics of posterior stabilized

total knee systems, Scientific Exhibit at the 63rd Annual Meeting of the American Academy

of Orthopedic Surgeons.

Heim, C.S., Postak, P.D., Plaxton, N.A., Greenwald, A.S. 2001. Classification of mobile-

bearing knee designs: Mobility and constraint. Journal of Bone and Joint Surgery-

American Volume, 83A, 32-37.

26

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

Kakarlapudi, T.K., Bickerstaff, D.R. 2000. Knee instability: isolated and complex. British

Journal of Sports Medicine, 34, 395-400.

Liow, R.Y., Murray, D.W. 1997. Which primary total knee replacement? A review of

currently available TKR in the United Kingdom. Annals of The Royal College of Surgeons

of England, 79, 335-340.

Moran, M.F., Bhimji, S., Racanelli, J., Piazza, S.J. 2008. Computational assessment of

constraint in total knee replacement. Journal of Biomechanics, 41, 2013-2020.

Mündermann, A., Dyrby, C.O., D'Lima, D.D., Colwell, C.W., Andriacchi, T.P. 2008. In vivo

knee loading characteristics during activities of daily living as measured by an

instrumented total knee replacement. Journal of Orthopaedic Research, 26, 1167-1172.

NJR, 2016. 13th Annual Report 2016, National Joint Registry for England and Wales,

http://www.njrcentre.org.uk/, date last accessed 03 August 2017.

Rullkoetter, P.J., Fitzpatrick, C.K., Clary, C.W. 2017. How Can We Use Computational

Modeling to Improve Total Knee Arthroplasty? Modeling Stability and Mobility in the

Implanted Knee. JAAOS - Journal of the American Academy of Orthopaedic Surgeons, 25,

S33-S39.

Shino, K., Inoue, M., Horibe, S., Nakamura, H., Ono, K. 1987. Measurement of anterior

instability of the knee - a new apparatus for clinical-testing. Journal of Bone and Joint

Surgery-British Volume, 69, 608-613.

Varadarajan, K.M., Moynihan, A.L., D’Lima, D., Colwell, C.W., Li, G. 2008. In vivo contact

kinematics and contact forces of the knee after total knee arthroplasty during dynamic

weight-bearing activities. Journal of Biomechanics, 41, 2159-2168.

27

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

Zhao, D., Banks, S.A., D'Lima, D.D., Colwell, C.W., Fregly, B.J. 2007. In vivo medial and

lateral tibial loads during dynamic and high flexion activities. Journal of Orthopaedic

Research, 25, 593-602.

28

408

409

410

411

412