The Impact of Imaging Modality on the Measurement of Coronal Plane Alignment After Total Knee Arthroplasty

Accepted Manuscript The Impact of Imaging Modality on the Measurement of Coronal Plane Alignment Following Total Knee Arthroplasty Denis Nam, MD, MSc, Sravya Vajapey, BS, Ryan M. Nunley, MD, Robert L. Barrack, MD PII:

S0883-5403(16)00214-X

DOI:

10.1016/j.arth.2016.02.063

Reference:

YARTH 55021

To appear in:

The Journal of Arthroplasty

Received Date: 23 December 2015 Revised Date:

16 February 2016

Accepted Date: 27 February 2016

Please cite this article as: Nam D, Vajapey S, Nunley RM, Barrack RL, The Impact of Imaging Modality on the Measurement of Coronal Plane Alignment Following Total Knee Arthroplasty, The Journal of Arthroplasty (2016), doi: 10.1016/j.arth.2016.02.063. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

The Impact of Imaging Modality on the Measurement of Coronal Plane Alignment Following Total Knee Arthroplasty

RI PT

Running Title: Effect of Imaging Modality on Alignment after TKA

Denis Nam, MD, MSc; Sravya Vajapey, BS; Ryan M. Nunley, MD; Robert L. Barrack, MD

AC C

EP

TE D

M AN U

SC

Washington University School of Medicine, Department of Orthopedic Surgery, St. Louis, MO

Please address all correspondence to: Denis Nam, MD, MSc Washington University School of Medicine Department of Orthopedic Surgery 660 S. Euclid Ave., Campus 8233 St. Louis, MO 63110

ACCEPTED MANUSCRIPT

RI PT

The Impact of Imaging Modality on the Measurement of Coronal Plane Alignment Following Total Knee Arthroplasty

EP

TE D

M AN U

SC

Running Title: Effect of Imaging Modality on Alignment after TKA

AC C

1 2 3 4 5 6 7 8 9 10 11

1

ACCEPTED MANUSCRIPT

Abstract

13

Background

14

The optimal coronal alignment following total knee arthroplasty (TKA) has become an area of

15

increased debate. Sources of variability amongst investigations include the radiographic

16

technique used for both preoperative surgical planning and postoperative alignment assessments.

17

This study’s purpose was to assess the impact of the imaging modality used on the measurement

18

of coronal plane alignment following TKA.

19

Methods

20

A consecutive series of patients undergoing TKA using the same cruciate-retaining prosthesis

21

were included for analysis. Postoperatively, all patients received both a rotationally controlled,

22

scout computed tomography (CT) scan and a hip-knee-ankle image using the EOS Imaging

23

system (EOS Inc., Paris, France). Two, independent observers measured the hip-knee-ankle

24

(HKA) angle, and femoral and tibial component alignment from each image.

25

Results

26

After classifying overall and component alignment as neutral, varus, or valgus, 40.6% (65/160)

27

of knees had a discordant alignment classification for HKA, 28.1% (45/160) for femoral

28

component alignment, and 26.9% (43/160) for tibial component alignment between their CT and

29

EOS images. 24.4% (39/160) of patients had a HKA difference of > 3° between the two images,

30

while 18.8% (30/160) and 20.0% (32/160) of patients had a femoral and tibial component

31

alignment difference of > 2°, respectively.

32

Conclusion

AC C

EP

TE D

M AN U

SC

RI PT

12

2

ACCEPTED MANUSCRIPT

Significant differences are present when comparing two measurement techniques of mechanical

34

alignment following TKA. The impact of imaging modality on postoperative assessments must

35

be accounted for and be consistent when comparing the results of different investigations.

36

Key Words: alignment; total knee arthroplasty; radiographs; biplanar radiography; computed

37

tomography

RI PT

33

38

SC

39 40

M AN U

41 42 43 44

48 49 50 51 52

EP

47

AC C

46

TE D

45

53

54

3

ACCEPTED MANUSCRIPT

55

Introduction

56

Traditionally, surgeons have targeted a “neutral” mechanical alignment with tibial and

57

femoral components aligned perpendicular to the coronal mechanical axis following total knee

58

arthroplasty (TKA)

59

neutral and component malpositioning has been associated with increased polyethylene contact

60

stresses and early aseptic loosening

61

neutral, postoperative mechanical axis alignment in TKA for both implant survivorship and

62

clinical outcomes

63

the fifteen-year implant survival rate in patients within and outside of a postoperative mechanical

64

axis of 0˚ + 3˚ 7. Furthermore, Vanlommel et al., in a retrospective review of 143 TKAs with a

65

pre-operative varus alignment, noted patients left in mild varus (3˚ to 6˚) to have superior Knee

66

Society Scores and Western Ontario and McMaster Universities Arthritis Indices compared with

67

knees corrected to neutral or left in severe varus (>6˚)

RI PT

3-6

. However, recent data has challenged the importance of a

. Parratte et al., in a review of 398 TKAs, demonstrated no improvement in

TE D

M AN U

7-9

. Failure to restore the overall mechanical alignment to within 3˚ of

SC

1-3

10

.

Thus, the optimal target for alignment following TKA remains an area of ongoing debate.

68 69

However, despite this controversy, modifications in surgical technique such as custom cutting

70

guides (CCGs)

71

with the aim of improving a surgeon’s accuracy in achieving their surgical target. De Steiger et

72

al. in a review of the Australian National Joint Registry did note the use of computer navigation

73

to reduce the overall rate of revision in patients less than 65 years of age 16, but other

74

investigations have failed to demonstrate similar improvements in clinical outcomes and

75

surivorship 14,17. In 2012, over 82,000 TKAs were performed worldwide with the use of CCGs

76

18

77

component alignment versus conventional methods 19. However, the potential benefit of

and computer-assisted navigation devices

14,15

continue to be developed

AC C

EP

11-13

, with one proposed benefit being an improvement in postoperative hip-knee-ankle (HKA) and

4

ACCEPTED MANUSCRIPT

improved alignment accuracy with CCGs has not been proven 20-22. One proposed source of

79

inaccuracy with the use of CCGs is the use of supine images for preoperative planning 23. While

80

3-dimensional imaging is crucial to fabricate cutting guides that fit securely to a patient’s native

81

anatomy, failure to incorporate the impact of weight-bearing both in preoperative planning and

82

postoperative assessments may be a potential source of inaccuracy when targeting a specific hip-

83

knee-ankle alignment. Furthermore, as the majority of studies evaluate postoperative alignment

84

using standing, 2-dimensional images, it is possible the discrepancy seen between the

85

preoperative plan and alignment achieved postoperatively is related in part to the imaging

86

modality used to evaluate postoperative alignment.

M AN U

SC

RI PT

78

Therefore, the optimal postoperative alignment following TKA and accuracy of surgical

88

techniques remain areas of continued investigation, yet comparisons across different studies are

89

difficult due to a lack of consistency in assessment methods. Postoperative, mechanical

90

alignment has typically been reported using measurement of the HKA from either standing AP

91

hip-to-ankle imaging or rotationally controlled scout computed tomography (CT) scans

92

However, to our knowledge, whether the outcomes from studies using these two, distinct

93

measurement methods can reliably be compared remains unclear. This study’s purpose was to

94

assess the impact of the imaging modality used on the measurement of coronal plane alignment

95

following TKA thru comparisons of supine, scout CT scans versus standing hip-knee-ankle

96

imaging in the same patient following TKA. Our hypothesis is that the imaging modality used

97

will have a significant impact on the measurement of a patient’s coronal plane alignment, thus

98

limiting the ability to compare data obtained using these two radiographic methods.

7,21

.

AC C

EP

TE D

87

99 100

5

ACCEPTED MANUSCRIPT

101

AC C

EP

TE D

M AN U

SC

RI PT

102

6

ACCEPTED MANUSCRIPT

103 104

Materials and Methods This was a retrospective review of a consecutive series of patients undergoing primary total knee arthroplasty with the same, cruciate-retaining TKA implant (Vanguard, Biomet Inc.,

106

Warsaw, IN, USA) by one of three attending surgeons. All TKAs were performed with the use

107

of a conventional, intramedullary femoral alignment guide and extramedullary tibial alignment

108

guide. Computer navigation and CCGs were not used in this investigation. Before initiation of

109

this study, institutional review board approval was obtained. All patients received both a supine,

110

rotationally controlled, scout computed tomography (CT) scan and a standing, hip-knee-ankle

111

image using the EOS® X-Ray Imaging Acquisition System (EOS Inc., Paris, France) following

112

their TKA, and all implants were well-fixed without signs of radiographic loosening. Patients

113

were excluded if they had prior traumatic fractures to the ipsilateral femur, knee, or tibia. Also,

114

patients with neuromuscular disorders, congenital anomalies, or ambulatory and/or standing

115

difficulties were excluded. Demographic information including age, gender, height, weight, and

116

body mass index (BMI) were recorded.

SC

M AN U

TE D

117

RI PT

105

The CT protocol included image acquisition of the hip, knee, and ankle in the supine position with extremities rotated into a neutral position in order to standardize measurements as

119

previously described by Nunley et al. 24. The TKA implant system used in this study has a

120

femoral component with two oval holes in the posterior flange to allow for potential placement

121

of posterior augments. The limb was rotated until the posterior augment holes were centered

122

relative to the anterior flange of the femoral component so that the extremities were in a similar

123

degree of rotation in order to minimize its impact on alignment measurements. The EOS

124

protocol included image acquisition of the hip, knee, and ankle in the standing position. Care

125

was taken to ensure the patellae were facing forward during image acquisition to control for

AC C

EP

118

7

ACCEPTED MANUSCRIPT

126

rotation. Only the 2-dimensional image obtained using the EOS® system was analyzed as it

127

corresponds to a standing, full-length, hip-knee-ankle radiograph obtained using conventional

128

radiography 25,26. For both the supine, coronal scout CT and weight-bearing hip-to-ankle images,

RI PT

129

radiographic measurements were performed using previously described methodology 3,7,14.

131

Measurements included the hip-knee-ankle angle (HKA), femorotibial angle (FTA), mechanical

132

lateral distal femoral angle (mLDFA), mechanical medial proximal tibial angle (mMPTA),

133

medial neck-shaft angle (MNSA), tibial bow, tibial length, femoral bow, and femoral length.

134

The mechanical femoral axis was defined as the line connecting the center of the femoral head,

135

as determined by a best-fit circle, and the midpoint of the widest dimension of the distal femur.

136

The mechanical tibial axis was set as the line connecting the center of the tibial spines to the

137

center of the talus. The anatomical femoral axis was defined as the line connecting the midpoint

138

of the endosteal cortices of the femoral isthmus to the midpoint of the femur 10 cm proximal to

139

the joint line. Similarly, the anatomical tibial axis was determined as the line connecting the

140

midpoint of the midshaft of the tibia to the midpoint of the tibia 10 cm distal to the joint line

141

The hip-knee-ankle angle was determined as the angle between the mechanical axes of the femur

142

and tibia 9. The femorotibial axis was the angle between the anatomic axes of the femur and

143

tibia. For convention, the HKA and FTA values were expressed as a deviation from 180° with a

144

negative value for varus and positive value for valgus alignment (Figure 1). The mMPTA was

145

determined as the medial angle between the mechanical tibial axis and the joint line of the

146

proximal tibial component and the mLDFA was defined as the lateral angle between the

147

mechanical femoral axis and the joint line of the distal femur component (Figure 2) 28. For

148

convention, varus/valgus component measurements were recorded as the difference between the

27

.

AC C

EP

TE D

M AN U

SC

130

8

ACCEPTED MANUSCRIPT

measurement angle and 90°, with negative values representing a varus alignment (i.e. -0.5°

150

represents a component in 0.5° of varus relative to the mechanical axis). Methods for measuring

151

the medial neck-shaft angle, tibial bow, tibial length, femoral bow, and femoral length have been

152

previously described and are presented in Table 1 29,30. Two, independent observers measured all

153

radiographs and the results were assessed for inter-observer reliability.

RI PT

149

For each image, a HKA <-3° was considered varus, neutral between -3° and 3°, and

155

valgus >3°. Additionally, when assessing alignment of the tibial and femoral component, a value

156

of <-2° was considered varus, neutral between -2° and 2°, and valgus >2°. These values were

157

chosen to maintain consistency with the definitions of varus and valgus established by clinical

158

outcomes and survivorship studies evaluating coronal alignment following TKA 3,7,14.

159

Statistical Analysis:

M AN U

160

SC

154

All data was collected and analyzed using Microsoft Excel software (Microsoft Corporation, Redmond, WA, USA). Independent sample t-tests were used to compare mean

162

radiographic alignments between the two imaging methods, while Chi-square tests were used to

163

compare the percentage of patients with a neutral HKA or component alignment. A p-value

164

<0.05 was considered statistically significant. The percentage of patients with discordant

165

classifications (i.e. neutral on the CT image, but varus on the hip-knee-ankle image) for HKA,

166

tibial, or femoral component alignment between the two imaging methods was determined.

167

Correlation coefficients were calculated to determine interobserver variability between

168

measurers. The association between patient demographics and morphologic anatomy with

169

differences between the two imaging methods was also determined - morphologic measurements

170

(i.e. medial neck-shaft angle, tibial length, tibial bow, etc.) were taken from the standing, EOS

171

images. Correlation coefficients were graded using a previously described semi-quantitative

AC C

EP

TE D

161

9

ACCEPTED MANUSCRIPT

172

criteria: excellent for 0.9 < r < 1.0, good for 0.7 < r < 0.89, fair/moderate for 0.5 < r < 0.69, low

173

for 0.25 < r < 0.49, and poor for 0.0 < r < 0.24

31

.

AC C

EP

TE D

M AN U

SC

RI PT

174

10

ACCEPTED MANUSCRIPT

175 176

Results One hundred sixty patients (73 left, 87 right) who received both a supine, rotationally controlled, scout CT scan and a standing, EOS, hip-knee-ankle image were available for

178

analysis. Patients had a mean age of 66.4 + 9.6 years, height of 168.3 + 11.7 cm, weight of 91.8

179

+ 19.4 kg, and BMI of 32.5 + 6.4 kg/m2. Interobserver reliability was excellent between the two,

180

independent observers for all radiographic measurements performed (r=0.90-0.96).

The mean HKA alignment measured on CT images was 1.3° + 2.7° versus -0.05° + 3.3°

SC

181

RI PT

177

on EOS images (p<0.001). The mean difference for HKA alignment for each patient between the

183

two imaging methods was 1.4° + 2.5°, with 24.4% of patients having a HKA difference of

184

greater than 3 degrees when comparing their supine, CT versus standing, EOS images. The

185

correlation for HKA alignment between the two imaging methods was fair/moderate (r=0.67).

186

When classifying the HKA alignment as varus, valgus, or neutral, 67.5% (108/60) of patients had

187

a neutral, HKA alignment on their CT image versus 60.0% (96/160) on the EOS image (p=0.2).

188

A significant proportion of patients had discordant classifications for HKA alignment as 40.6%

189

(65/160) of subjects had a discordant classification on their supine, CT image versus their

190

standing, EOS image (e.g. a neutral HKA alignment on the CT image, but varus HKA alignment

191

on the EOS image) (Table 2).

TE D

EP

The mean femoral component alignment measured on CT images was -0.3° + 2.3° versus

AC C

192

M AN U

182

193

-0.8° + 2.5° on EOS images (p=0.06). The mean difference for femoral component alignment

194

for each patient between the two imaging methods was 0.4° + 1.5°, with 5.0% of patients having

195

a femoral component difference greater than 3 degrees, and 18.8% of patients having a

196

difference greater than 2 degrees when comparing their supine, CT versus standing, EOS images.

197

The correlation for femoral component alignment between the two imaging methods was good

11

ACCEPTED MANUSCRIPT

(r=0.79). When classifying the femoral component alignment as varus, valgus, or neutral, 60.0%

199

(96/160) of patients had a neutral femoral component alignment on their CT image versus 62.5%

200

(100/160) on their EOS image (p=0.6). 28.1% (45/160) of subjects had a discordant

201

classification on their supine, CT image versus their standing, EOS image (Figure 3).

202

RI PT

198

The mean tibial component alignment measured on CT images was 1.2° + 2.0° versus 0.6° + 1.7° on EOS images (p=0.004). The mean difference for tibial component alignment for

204

each patient between the two imaging methods was 0.6° + 1.6°, with 5.6% of patients having a

205

tibial component difference greater than 3 degrees, and 20.0% of patients having a difference

206

greater than 2 degrees when comparing their supine, CT versus standing, EOS images. The

207

correlation of tibial component alignment between the two imaging methods was fair/moderate

208

(r=0.65). When classifying the tibial component alignment as varus, valgus, or neutral, 63.8%

209

(102/160) of patients had a neutral tibial component alignment on their CT image versus 73.1%

210

(117/160) on their EOS image (p=0.07). 26.9% (43/160) of subjects had a discordant

211

classification on their supine, CT image versus their standing, EOS image (Figure 4).

TE D

M AN U

SC

203

Alignment differences for the HKA, mLDFA, mMPTA between the two imaging

213

modalities were not associated with patient height, weight, body mass index, medial neck-shaft

214

angle, tibial bow, tibial length, femoral bow, or femoral length (r= -0.14 to 0.15).

216 217

AC C

215

EP

212

218 219 220

12

ACCEPTED MANUSCRIPT

221

Discussion The optimal postoperative mechanical alignment following total knee arthroplasty

223

continues to be an area of investigation and debate, as traditional targets of a neutral, mechanical

224

axis and joint line perpendicular to the coronal, mechanical axis have been questioned 7-10. As

225

the number of studies assessing the impact of postoperative alignment on clinical outcomes

226

continues to increase, it becomes increasingly difficult to formulate a consensus as varying forms

227

of measuring alignment are reported. Similarly, in the assessment of the accuracy of new

228

surgical techniques such as custom cutting guides, determining the true accuracy of these

229

techniques remains difficult due to various methods of evaluation. The purpose of this study was

230

to assess the impact of the imaging modality used on the measurement of coronal plane

231

alignment following TKA. We hypothesized that the imaging modality used would have a

232

significant impact on the measurement of a patient’s overall mechanical and component

233

alignment. This study demonstrates that significant differences are present when comparing the

234

use of supine versus standing images in the measurement of coronal alignment. Thus, this

235

difference must be accounted for during postoperative assessments of total knee arthroplasty.

SC

M AN U

TE D

This study does have several limitations that must be considered prior to interpretation of

EP

236

RI PT

222

our results. As with all radiographic assessments of coronal plane alignment, it is known that

238

rotational attitudes can affect standard measurements of limb alignment 32,33; thus, variability in

239

lower extremity rotation could contribute to our findings. However, a detailed, uniform protocol

240

was followed for each measurement technique that attempted to minimize this variability.

241

Furthermore, the possibility that rotation contributes to variability between supine and weight-

242

bearing images only strengthens the argument for a standardized method of postoperative

243

radiographic assessment. Second, this study is unable to elucidate which method is most

AC C

237

13

ACCEPTED MANUSCRIPT

accurate in measuring coronal alignment, as currently there is no gold standard for comparison.

245

However, this was not the purpose of this investigation, as this study demonstrates a proof-of-

246

principle that the imaging modality used can drastically affect the measurement of coronal

247

alignment. As standing images incorporate the impact of weight-bearing, soft tissue balance, and

248

potentially represent a more functional position, we propose that this method may be more useful

249

rather than supine imaging.

To our knowledge, few studies have investigated the impact of the specific imaging

SC

250

RI PT

244

modality on lower limb coronal alignment. Winter et al. analyzed 45 knees who received both

252

pre-operative standing full-length radiographs and supine magnetic resonance imaging in

253

patients undergoing patient specific TKA with the use of CCGs. They found the mean difference

254

in coronal alignment between the two techniques to be 2° with supine images under-estimating

255

the degree of deformity in 69% of cases 34. In addition, Clarke et al. assessed coronal alignment

256

in 30 asymptomatic controls and 31 patients following TKA using a non-invasive infrared

257

position capture system when going from the supine to standing position. They found a trend

258

towards increased varus when assuming a standing position with a mean difference of

259

approximately 1.8° 35. Our investigation demonstrates similar findings with a mean difference in

260

HKA alignment between supine, CT and standing, EOS images to be approximately 1.4°.

261

However, our study contributes to this pre-existing literature as we included a significantly

262

greater number of patients and also evaluated the potential impact of weight-bearing on the

263

measurement of overall alignment following TKA.

TE D

EP

AC C

264

M AN U

251

The results of this study can be interpreted in several ways. First, this study clearly

265

demonstrates the imaging method used to have a significant impact on the measurement and

266

classification (neutral, varus, valgus) of alignment following TKA. Of note, as suggested by

14

ACCEPTED MANUSCRIPT

Parratte et al. 7, describing alignment as a categorical variable based on achievement of an axis

268

within 3° of neutral may be of little utility, and assessment of survivorship based on alignment as

269

a continuous variable is likely of greater significance. However, our results also demonstrate a

270

significant percentage of patients to have a difference in HKA alignment of greater than 3°, and

271

femoral and tibial component alignment differences of greater than 2° when comparing their

272

supine, CT versus standing, EOS images. Thus, the differences seen in imaging modalities

273

extend beyond simple, categorical descriptions of varus, valgus, and neutral. Thus, we question

274

whether prior investigations measuring alignment can accurately be compared which limits the

275

ability to develop a consensus of optimal alignment following TKA. Furthermore, it limits the

276

ability to consistently assess the accuracy of surgical techniques such as custom cutting guides.

277

For example, studies evaluating the accuracy of CCGs have used both standing, weight-bearing

278

images 12 and supine, CT imaging

279

Thus, whether the outcomes of these studies can accurately be compared remains a concern.

280

Second, regarding CCGs, a potential source of inaccuracy is that weight-bearing is not accounted

281

for during preoperative planning. As 3-dimensional imaging in the supine position is most

282

commonly used to fabricate CCGs, weight-bearing and functional alignment is not taken into

283

consideration during preoperative planning. While weight-bearing should not significantly

284

impact tibial and femoral component alignment in the coronal plane if rotation is adequately

285

controlled, the contributions of ligamentous laxity and stability are not being accounted for

286

during preoperative planning. Due to variations in the soft tissue envelope, ligamentous

287

integrity, and intraoperative technique, is will be difficult to predict a patient’s postoperative

288

alignment without some form of preoperative, standing image. Furthermore, the discrepancy

289

seen in the preoperative plan and alignment achieved postoperatively may be related to the use of

for alignment measurements with conflicting results.

AC C

EP

TE D

21,22

M AN U

SC

RI PT

267

15

ACCEPTED MANUSCRIPT

290

different imaging modalities for preoperative planning versus during the measurement of

291

postoperative alignment.

292

In conclusion, this study confirms our hypothesis that the imaging modality used will have a significant impact on the measurement of coronal plane alignment and demonstrates the

294

necessity for a standardized method of postoperative assessment following total knee

295

arthroplasty. When classifying alignment as neutral, varus, or valgus, 40.6% of patients for the

296

HKA, 28.1% for the femoral component, and 26.9% for the tibial component alignment had

297

discordant classifications when comparing a supine, CT image to a standing, EOS image. The

298

impact of the imaging modality used for both preoperative planning and postoperative

299

assessments must be accounted for and be consistent when comparing the results of different

300

investigations.

M AN U

SC

RI PT

293

301

EP

304

AC C

303

TE D

302

16

ACCEPTED MANUSCRIPT

REFERENCES

306

1. Longstaff LM, Sloan K, Stamp N, Scaddan M, Beaver R. Good alignment after total knee

307

arthroplasty leads to faster rehabilitation and better function. J Arthroplasty. 2009;24(4):570-

308

578.

309

2. Ritter MA, Faris PM, Keating EM, Meding JB. Postoperative alignment of total knee

310

replacement. its effect on survival. Clin Orthop Relat Res. 1994;(299)(299):153-156.

311

3. Ritter MA, Davis KE, Meding JB, Pierson JL, Berend ME, Malinzak RA. The effect of

312

alignment and BMI on failure of total knee replacement. J Bone Joint Surg Am.

313

2011;93(17):1588-1596.

314

4. Jeffery RS, Morris RW, Denham RA. Coronal alignment after total knee replacement. J Bone

315

Joint Surg Br. 1991;73(5):709-714.

316

5. D'Lima DD, Chen PC, Colwell CW,Jr. Polyethylene contact stresses, articular congruity, and

317

knee alignment. Clin Orthop Relat Res. 2001;(392)(392):232-238.

318

6. Berend ME, Ritter MA, Meding JB, et al. Tibial component failure mechanisms in total knee

319

arthroplasty. Clin Orthop Relat Res. 2004;(428)(428):26-34.

320

7. Parratte S, Pagnano MW, Trousdale RT, Berry DJ. Effect of postoperative mechanical axis

321

alignment on the fifteen-year survival of modern, cemented total knee replacements. J Bone

322

Joint Surg Am. 2010;92(12):2143-2149.

AC C

EP

TE D

M AN U

SC

RI PT

305

17

ACCEPTED MANUSCRIPT

8. Howell SM, Howell SJ, Kuznik KT, Cohen J, Hull ML. Does a kinematically aligned total

324

knee arthroplasty restore function without failure regardless of alignment category? Clin Orthop

325

Relat Res. 2013;471(3):1000-1007.

326

9. Bellemans J, Colyn W, Vandenneucker H, Victor J. The chitranjan ranawat award: Is neutral

327

mechanical alignment normal for all patients? the concept of constitutional varus. Clin Orthop

328

Relat Res. 2012;470(1):45-53.

329

10. Vanlommel L, Vanlommel J, Claes S, Bellemans J. Slight undercorrection following total

330

knee arthroplasty results in superior clinical outcomes in varus knees. Knee Surg Sports

331

Traumatol Arthrosc. 2013.

332

11. Nam D, Park A, Stambough JB, Johnson SR, Nunley RM, Barrack RL. The mark coventry

333

award: Custom cutting guides do not improve total knee arthroplasty clinical outcomes at 2 years

334

followup. Clin Orthop Relat Res. 2015.

335

12. Ng VY, Declaire JH, Berend KR, Gulick BC, Lombardi AV,Jr. Improved accuracy of

336

alignment with patient-specific positioning guides compared with manual instrumentation in

337

TKA. Clin Orthop Relat Res. 2011.

338

13. Lombardi AV,Jr, Berend KR, Adams JB. Patient-specific approach in total knee arthroplasty.

339

Orthopedics. 2008;31(9):927-930.

340

14. Mason JB, Fehring TK, Estok R, Banel D, Fahrbach K. Meta-analysis of alignment outcomes

341

in computer-assisted total knee arthroplasty surgery. J Arthroplasty. 2007;22(8):1097-1106.

AC C

EP

TE D

M AN U

SC

RI PT

323

18

ACCEPTED MANUSCRIPT

15. Nam D, Cody EA, Nguyen JT, Figgie MP, Mayman DJ. Extramedullary guides versus

343

portable, accelerometer-based navigation for tibial alignment in total knee arthroplasty: A

344

randomized, controlled trial: Winner of the 2013 HAP PAUL award. J Arthroplasty.

345

2014;29(2):288-294.

346

16. de Steiger RN, Liu YL, Graves SE. Computer navigation for total knee arthroplasty reduces

347

revision rate for patients less than sixty-five years of age. J Bone Joint Surg Am. 2015;97(8):635-

348

642.

349

17. Kim YH, Park JW, Kim JS. Computer-navigated versus conventional total knee arthroplasty

350

a prospective randomized trial. J Bone Joint Surg Am. 2012;94(22):2017-2024.

351

18. Thienpont E, Bellemans J, Delport H, et al. Patient-specific instruments: Industry's

352

innovation with a surgeon's interest. Knee Surg Sports Traumatol Arthrosc. 2013;21(10):2227-

353

2233.

354

19. Nam D, McArthur BA, Cross MB, Pearle AD, Mayman DJ, Haas SB. Patient-specific

355

instrumentation in total knee arthroplasty: A review. J Knee Surg. 2012;25(3):213-219.

356

20. Slover J, Rubash H, Malchau H, Bosco J. Cost-effectiveness analysis of custom total knee

357

cutting blocks. J Arthroplasty. 2011.

358

21. Woolson ST, Harris AH, Wagner DW, Giori NJ. Component alignment during total knee

359

arthroplasty with use of standard or custom instrumentation: A randomized clinical trial using

360

computed tomography for postoperative alignment measurement. J Bone Joint Surg Am.

361

2014;96(5):366-372.

AC C

EP

TE D

M AN U

SC

RI PT

342

19

ACCEPTED MANUSCRIPT

22. Barrack RL, Ruh EL, Williams BM, Ford AD, Foreman K, Nunley RM. Patient specific

363

cutting blocks are currently of no proven value. J Bone Joint Surg Br. 2012;94(11 Suppl A):95-

364

99.

365

23. Nam D, Williams B, Hirsh J, Johnson SR, Nunley RM, Barrack RL. Planned bone resections

366

using an MRI-based custom cutting guide system versus 3-dimensional, weight-bearing images

367

in total knee arthroplasty. J Arthroplasty. 2015;30(4):567-572.

368

24. Nunley RM, Ellison BS, Ruh EL, et al. Are patient-specific cutting blocks cost-effective for

369

total knee arthroplasty? Clin Orthop Relat Res. 2012;470(3):889-894.

370

25. Thelen P, Delin C, Folinais D, Radier C. Evaluation of a new low-dose biplanar system to

371

assess lower-limb alignment in 3D: A phantom study. Skeletal Radiol. 2012;41(10):1287-1293.

372

26. Than P, Szuper K, Somoskeoy S, Warta V, Illes T. Geometrical values of the normal and

373

arthritic hip and knee detected with the EOS imaging system. Int Orthop. 2012;36(6):1291-1297.

374

27. Petersen TL, Engh GA. Radiographic assessment of knee alignment after total knee

375

arthroplasty. J Arthroplasty. 1988;3(1):67-72.

376

28. Nam D, Maher PA, Robles A, McLawhorn AS, Mayman DJ. Variability in the relationship

377

between the distal femoral mechanical and anatomical axes in patients undergoing primary total

378

knee arthroplasty. J Arthroplasty. 2013.

379

29. Chaibi Y, Cresson T, Aubert B, et al. Fast 3D reconstruction of the lower limb using a

380

parametric model and statistical inferences and clinical measurements calculation from biplanar

381

X-rays. Comput Methods Biomech Biomed Engin. 2012;15(5):457-466.

AC C

EP

TE D

M AN U

SC

RI PT

362

20

ACCEPTED MANUSCRIPT

30. Paley D, Herzenberg JE. Principles of deformity correction. 1st ed. New York: Springer-

383

Verlag; 2003.

384

31. Munro BH. Correlation. In: Statistical methods for healthcare research. 3rd ed. Lippincott-

385

Raven; 1997:224-245.

386

32. Radtke K, Becher C, Noll Y, Ostermeier S. Effect of limb rotation on radiographic alignment

387

in total knee arthroplasties. Arch Orthop Trauma Surg. 2010;130(4):451-457.

388

33. Jiang CC, Insall JN. Effect of rotation on the axial alignment of the femur. pitfalls in the use

389

of femoral intramedullary guides in total knee arthroplasty. Clin Orthop Relat Res.

390

1989;(248)(248):50-56.

391

34. Winter A, Ferguson K, Syme B, McMillan J, Holt G. Pre-operative analysis of lower limb

392

coronal alignment - a comparison of supine MRI versus standing full-length alignment

393

radiographs. Knee. 2014;21(6):1084-1087.

394

35. Clarke JV, Deakin AH, Picard F, Riches PE. The effect of weight-bearing on tibiofemoral

395

alignment in asymptomatic, osteoarthritic, and prosthetic knees. J Bone Joint Surg Br. 2012;94-

396

B(Suppl XLIV):57.

SC

M AN U

TE D

EP

AC C

397

RI PT

382

21

ACCEPTED MANUSCRIPT

Radiographic Measurement

Definition

Medial Neck-Shaft Angle (°°)

RI PT

The angle formed by a line drawn through the center of the femoral shaft, and a line from the center of the femoral head through the center of the femoral neck

The distance from the center of the tibial plateau to the center of the tibial plafond

Tibial Bow (mm)

The perpendicular distance between the anatomic axis of the tibia and the center of the talus

Femoral Length (mm)

The distance from the center of the femoral head to the distal femoral intercondylar notch

Femoral Bow (mm)

The difference between the hip lateral offset and the femoral bow offset.

M AN U

SC

Tibial Length (mm)

TE D

The hip lateral offset was measured as the perpendicular distance (mm) between the center of the femoral head, and a line drawn through the center of the femoral shaft originating 10 cm below the tip of the lesser trochanter.

AC C

EP

The femoral bow offset was the perpendicular distance (mm) from the center of the femoral head, and the anatomic axis of the femur drawn from the center of the distal third of the femur above the knee (10cm above the most distal aspect of the femur).

Table 1: Table summarizing the methodology of measuring the medial neck-shaft angle, tibial length, tibial bow, femoral length, and femoral bow.

ACCEPTED MANUSCRIPT

Supine, CT Image

Standing, EOS Image

108 (67.5%)

96 (60.0%)

Varus

11 (6.9%)

36 (22.5%)

Valgus

41 (25.6%)

28 (17.5%)

Hip-Knee-Ankle Alignment Neutral

RI PT

0.2

% Discordant

40.6%

Femoral Component Alignment Neutral

0.6

Valgus

24 (15.0%)

47 (29.4%) 13 (8.1%)

M AN U

40 (25.0%)

100 (62.5%)

SC

96 (60.0%)

Varus

% Discordant Tibial Component Alignment Neutral

117 (73.1%)

Varus

7 (4.4%)

10 (6.3%)

Valgus

51 (31.9%)

33 (20.6%)

TE D

102 (63.8%)

% Discordant

p-value/ discordance

28.1%

.07

26.9%

AC C

EP

Table 2: Table summarizing classifications of hip-knee-ankle, femoral, and tibial component alignment on the CT and EOS images. Values are presented as the absolute number followed by the percentage in parentheses.

ACCEPTED MANUSCRIPT

FIGURE LEGEND: Figure 1: Postoperative, standing EOS radiograph demonstrating measurement of the hip-knee-

RI PT

ankle alignment. This was measured to be 177.2° or 2.8° of valgus. Figure 2: Postoperative radiographs demonstrating measurement of tibial and femoral

component alignment relative to each, respective mechanical axis. The tibial component was

SC

measured to be in 1.3° of valgus (A) and the femoral component in 1.8° of valgus (B).

Figure 3: Postoperative radiographs in the same patient demonstrating a femoral component

measured on a standing, EOS image (B).

M AN U

alignment of 3.2° of valgus when measured on a supine, CT image (A) and 0.9° of varus when

Figure 4: Postoperative radiographs in the same patient demonstrating a tibial component alignment of 3.2° of valgus when measured on a supine, CT image (A) and 0.4° of varus when

AC C

EP

TE D

measured on a standing, EOS image (B).

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

EP

B.

AC C

A.

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

EP

B.

AC C

A.

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

EP

B.

AC C

A.

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT