Effects of a 10-week toe-out gait modification intervention in people with medial knee osteoarthritis: a pilot, feasibility study

Accepted Manuscript Effects of a 10-week toe-out gait modification intervention in people with medial knee osteoarthritis: A pilot, feasibility study Michael A. Hunt , PT, PhD Judit Takacs , MSc PII:

S1063-4584(14)01042-5

DOI:

10.1016/j.joca.2014.04.007

Reference:

YJOCA 3119

To appear in:

Osteoarthritis and Cartilage

Received Date: 16 August 2013 Revised Date:

9 March 2014

Accepted Date: 9 April 2014

Please cite this article as: Hunt MA, Takacs J, Effects of a 10-week toe-out gait modification intervention in people with medial knee osteoarthritis: A pilot, feasibility study, Osteoarthritis and Cartilage (2014), doi: 10.1016/j.joca.2014.04.007. 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

Effects of a 10-week toe-out gait modification intervention in people with medial knee

RI PT

osteoarthritis: A pilot, feasibility study

Michael A. Hunt PT, PhD1

Department of Physical Therapy, University of British Columbia: Vancouver, BC, Canada

MAH: [email protected] JT: [email protected]

M AN U

1

SC

Judit Takacs MSc1

EP

TE D

Corresponding Author: Dr. Michael A. Hunt Department of Physical Therapy – University of British Columbia 212-2177 Wesbrook Mall Vancouver, BC Canada V6T 1Z3 Tel: (604) 827-4721 Fax: (604) 822-1870 E-mail:[email protected]

AC C

Running Title: Toe-out gait modification for knee OA Word Count: 4448

ACCEPTED MANUSCRIPT

Abstract Objective: To examine the feasibility of a ten week gait modification program in people with medial tibiofemoral knee osteoarthritis (OA), and to assess changes in clinical and biomechanical outcomes.

SC

RI PT

Design: Fifteen people with medial knee OA completed ten weeks of gait modification focusing on increasing toe-out angle during stance 10 degrees compared to their self-selected angle measured at baseline. In addition to adherence and performance difficulty outcomes, knee joint symptoms (Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale and total score, numerical rating scale (NRS) of pain, and knee joint loading during gait (late stance peak knee adduction moment (KAM)) were assessed.

M AN U

Results: Participants were able to perform the toe-out gait modification program with minimal to moderate difficulty, and exhibited significant increases in self-selected toe-out angle during walking (p < 0.001). Joint discomfort was reported by five participants (33%) in the hip or knee joints, though none lasted longer than two weeks. Participants reported statistically significant reductions in WOMAC pain (p = 0.02), NRS pain (p < 0.001), WOMAC total score (p = 0.02), and late stance KAM (p = 0.04). Conclusions: These preliminary findings suggest that toe-out gait modification is feasible in people with medial compartment knee OA. Preliminary changes in clinical and biomechanical outcomes provide the impetus for conducting larger scale studies of gait modification in people with knee OA to confirm these findings.

AC C

EP

TE D

Key words: knee; osteoarthritis; knee adduction moment; gait modification

ACCEPTED MANUSCRIPT

Introduction

2

Osteoarthritis (OA) is one of the most common musculoskeletal impairments and is currently

3

one of the leading causes of chronic physical disability in adults 1. The knee is the weight-

4

bearing joint most commonly affected with OA. Given that quality of life decreases and the

5

economic burden increases with more severe symptoms, it is imperative to implement treatment

6

approaches that effectively slow disease progression. Excessive loading and alterations in load

7

distribution within the joint are believed to play major roles in the breakdown of articular

8

cartilage indicative of knee OA progression 2-5. Accordingly, there has been increased interest in

9

the knee OA literature to develop treatments that target abnormal loading patterns within the

SC

M AN U

10

RI PT

1

knee joint.

11

The knee adduction moment (KAM) is generally regarded as an important proxy for medial

13

tibiofemoral compartment loading during walking 3, 6-7. Further, the overall peak KAM has been

14

shown to be associated with clinical outcomes unique to medial compartment knee OA including

15

lower limb malalignment 8-9, medial compartment disease severity 10-11, and tibial bone mineral

16

density ratios 12. Importantly, Miyazaki et al 13 showed that an approximate 25% increase in

17

overall peak magnitude of the KAM at baseline was associated with 6.6 times the risk of

18

radiographic medial compartment disease progression over six years while Bennell et al 14,

19

showed that the total area under the KAM-time curve (KAM impulse) at baseline was predictive

20

of loss of cartilage volume over 12 months using magnetic resonance imaging. Taken together,

21

these findings point to an important role for quantification of the KAM in knee OA research and

22

many studies aiming to examine the effectiveness of load-reducing treatments for knee OA have

23

reported the KAM.

AC C

EP

TE D

12

1

ACCEPTED MANUSCRIPT

24 One such treatment is gait modification. A recent systematic review has shown consistent

26

changes in KAM magnitudes following single-session gait modifications such as altering toe-out

27

angle 15. These findings are consistent with previous cross-sectional studies that have shown

28

significant inverse correlations between late stance KAM magnitudes and toe-out (r > -0.26)

29

magnitudes during walking 16. Previous studies have examined the effects of changing toe-out

30

(foot progression) angle on late stance KAM magnitudes in young, healthy individuals without

31

knee OA 17-19 or in those with radiographic evidence of knee OA 19-20. Though the methodologies

32

differed, consistent findings of reductions in the late stance KAM magnitude suggest a beneficial

33

effect on medial compartment load with toe-out gait modification. Though the clinical relevance

34

of the late stance peak KAM is still unclear (most early studies examined the early stance peak

35

KAM), increases in toe-out angle have been shown to have a protective role against knee OA

36

progression.

SC

M AN U

TE D

37

RI PT

25

Chang et al 21 studied 56 people with radiographic knee OA over the course of 18 months and

39

assessed gait biomechanics as well as radiographic features of knee OA. As a group, each 5

40

degree increase in baseline toe-out angle had an associated odds ratio of 0.60 (95% CI: 0.37,

41

0.98) for medial tibiofemoral progression. These findings are consistent with the known effects

42

of toe-out gait modification on knee joint load parameters – specifically the late stance peak

43

KAM – and suggest that toe-out gait modification may have longer-term biomechanical and

44

clinical benefits for people with knee OA.

AC C

EP

38

45

2

ACCEPTED MANUSCRIPT

Despite the potential treatment benefits of toe-out gait modification, there are no published

47

studies of the effects of longer-term (i.e. more than one session) toe-out modification in people

48

with knee OA. However, Shull et al recently reported significant improvements in biomechanical

49

and clinical outcomes following six weeks of toe-in gait retraining guided by haptic feedback 22.

50

Their study provided initial confirmation of the potential benefits of gait retraining for people

51

with knee OA. However, despite their findings, much is still unknown about gait retraining as a

52

treatment strategy for those with knee OA. First, they chose toe-in rather than toe-out for

53

modification in their study. Thus, the effects of a toe-out only modification remain unknown.

54

Further, the effects of any load-reducing gait modification on joints other than the knee have

55

received little interest. Specifically, given that modification of a given gait characteristic will

56

involve changes at more than one joint or body segment (increasing the toe-out angle, for

57

example, will involve external rotations at the ankle, knee, and hip), positive benefits at the knee

58

could be associated with concurrent negative consequences at other joints. Finally, the feasibility

59

of delivering a longer-term toe-out gait modification program to people with knee OA is

60

unknown.

SC

M AN U

TE D

EP

61

RI PT

46

Therefore, the primary objective of this exploratory, pilot study was to evaluate the efficacy,

63

safety, and adherence to a 10-week toe-out gait modification program in a group of individuals

64

with knee OA. Our secondary objective was to obtain pilot data pertaining to the effect of a

65

longer-term toe-out gait modification intervention on clinical and biomechanical outcomes

66

relevant to knee OA. We hypothesized that individuals would adhere to the program and perform

67

the toe-out gait modification with minimal to moderate difficulty. We also hypothesized that

AC C

62

3

ACCEPTED MANUSCRIPT

68

changes in knee joint loading and pain could be achieved with minimal reports of pain at

69

adjacent lower limb joints.

70 Methods

RI PT

71 72

Participants

74

Community-dwelling individuals with medial tibiofemoral OA were recruited from our

75

laboratory participant database and via advertisements in print media. Inclusion criteria included:

76

(i) definitive medial tibiofemoral osteophytes; (ii) joint space narrowing greater in the medial

77

tibiofemoral compartment compared to the lateral compartment; (iii) predominance of pain over

78

the medial aspect of the knee; (iii) history of knee pain longer than six months; and (iv) average

79

knee pain of at least 3 out of 10 (using an 11-point numerical rating scale (NRS) with terminal

80

descriptors of 0 = “no pain” and 10 = “worst pain imaginable”) over the one month prior to

81

initial screening. Exclusion criteria included: (i) knee surgery or intra-articular pain relief

82

injection within six months; (ii) current or past (within six weeks) oral corticosteroid use; (iii)

83

history of knee joint replacement or tibial osteotomy; (iv) any other condition affecting lower

84

limb function; and (v) participation in a new structured exercise program (defined as at least 45

85

minutes of continuous, planned exercise three or more days per week) within the past three

86

months or planning to commence exercise or other treatment for knee OA in the next three

87

months. The study was approved by the Institution’s Clinical Research Ethics Board and all

88

participants provided written informed consent.

AC C

EP

TE D

M AN U

SC

73

89 90

Study Design

4

ACCEPTED MANUSCRIPT

Interested participants were initially screened for inclusion and exclusion criteria over the

92

telephone and eligible individuals were referred for radiographic evaluation. Standing, semi-

93

flexed, postero-anterior radiographs were obtained and graded for disease severity using the

94

Kellgren and Lawrence (KL) OA classification system 23. Individuals who met the radiographic

95

criteria listed above were invited to the laboratory for a baseline (Week 0) testing session where

96

self-report, objective physical function, and biomechanical data were obtained. Participants

97

returned to the laboratory the following week (Week 1) to begin the gait modification program.

98

This program lasted ten weeks and involved six training visits with the study therapist at Weeks

99

1, 2, 3, 5, 7, and 9 of the intervention. Follow-up testing occurred upon completion of the

M AN U

SC

RI PT

91

intervention (Week 11). Ten weeks was chosen as an appropriate intervention length based on

101

previous exercise studies in the knee OA population (typically lasting approximately 8-12

102

weeks), and to provide sufficient time to establish motor re-learning of the new gait

103

modification.

104

TE D

100

Gait modification intervention

106

Participants were instructed to increase the toe-out angle of their study limb (most painful in the

107

case of bilateral involvement) by 10 degrees over and above the self-selected amount measured

108

at baseline. Ten degrees was chosen based on self-reported difficulty in obtaining 15-20 degrees

109

of toe-out increase by those with knee OA (unpublished data), and previous research indicating

110

an approximate 40% reduction in risk of radiographic disease progression over 18 months for

111

every five degree increase in baseline toe-out angle 21. Toe-out modification was facilitated

112

through the use of real-time biofeedback of performance 24-26 at each treatment visit. Participants

113

walked on a treadmill in their own walking shoes at a speed similar to their over ground walking

AC C

EP

105

5

ACCEPTED MANUSCRIPT

speed, and were provided with tracking data pertaining to their toe-out angle (transverse plane

115

angle of line connecting toe and heel with respect to line of forward progression of the body) in

116

real-time. Twenty-two retro-reflective markers were affixed to the participant according to a

117

modified Helen Hayes marker set 27, and their movements were captured using ten high-speed

118

digital cameras (Motion Analysis Corp., Santa Rosa, CA) sampling at 120 Hz. An initial

119

standing static trial was collected using additional markers placed over the medial malleoli and

120

femoral epicondyles to determine segment orientations. During the treadmill walking, toe-out

121

angle was calculated in real-time using the Biofeedtrak option within Cortex software (Motion

122

Analysis Corp., Santa Rosa, CA) and displayed graphically in front of the participant as a

123

continuous line streamed live via an overhead projector.

M AN U

SC

RI PT

114

124

Prior to each training session, participants stood on a protractor device at their target toe-out

126

angle (Figure 1) to ensure that the anatomical toe-out angle matched the calculated toe-out angle

127

obtained from the motion analysis system. While remaining on the protractor device, a vertical

128

target zone was then placed on-screen that corresponded to the target toe-out angle. During each

129

30-minute training session, participants were instructed to match the line corresponding to their

130

actual toe-out angle with the vertical target zone (Figure 2) during stance. To promote motor

131

learning, a faded feedback paradigm was used 25 with the amount of real-time biofeedback

132

displayed decreasing from Weeks 1 and 2 (30 minutes) throughout the intervention (Week 3 = 25

133

minutes; Week 5 = 20 minutes; Week 7 = 15 minutes; Week 9 = 10 minutes).

EP

AC C

134

TE D

125

135

Fidelity of the training sessions was assessed by calculating the difference (toe-out error)

136

between the intended target angle and the actual toe-out angle produced during 10 consecutive

6

ACCEPTED MANUSCRIPT

stance phases at each of four time points during each training session (1 minute, 10 minutes, 20

138

minutes, and 30 minutes) and averaged within the session. Perceived difficulty in achieving the

139

target toe-out angle was also assessed at each training session using an 11-point NRS (0 = “no

140

difficulty”, 10 = “unable to perform”), knee joint pain was also assessed using an 11-point NRS

141

(0 = “no pain” and 10 = “worst pain imaginable”), while perceived physical exertion was

142

measured during training using the 15-point Borg rating scale of perceived exertion 28.

143

Participants were instructed to maintain the increased toe-out angle during walking outside the

144

laboratory.

SC

RI PT

137

M AN U

145

Adherence to the training program was assessed as the total number of supervised walking

147

sessions attended for each participant. Compliance with the gait modification program was

148

assessed weekly via self-report as the perceived confidence in maintenance of the toe-out

149

modification throughout the week outside the laboratory using an 11-point NRS (0 = “not

150

confident at all”, 10 = “very confident”). Adverse events were assessed using open-ended

151

questions included in a log book maintained by the participants. Finally, physical activity levels

152

were monitored throughout the intervention using a wearable pedometer device (Fitbit One,

153

Fitbit Inc., San Francisco, CA) that uploaded weekly step count data to a centralized database.

EP

AC C

154

TE D

146

155

Pain and physical function assessment

156

At baseline and follow-up, participants completed the Western Ontario and McMaster

157

Universities Arthritis Index (WOMAC) Likert version. Average knee pain over the previous

158

week was assessed using an 11-point NRS with terminal descriptors of 0 = “no pain” and 10 =

159

“worst pain imaginable”. Overall perceived change at follow-up compared to baseline was

7

ACCEPTED MANUSCRIPT

assessed using a 5-point Likert scale (1 = “much worse”, 2 = “slightly worse”, 3 = “no change”,

161

4 = “slightly better”, and 5 = “much better”) 29. Finally, participants completed the timed stair

162

climb test where they were instructed to ascend 12 stairs “as quickly as possible” and the fastest

163

time from two attempts was recorded 30-31.

164

RI PT

160

Gait biomechanics assessment

166

Gait data were recorded for barefoot over ground (i.e. non-treadmill) walking trials at a self-

167

selected speed using the same marker set and camera system as described above. As in the

168

training sessions, an initial standing static trial with the additional four medial markers was

169

collected. Kinematic data from the cameras were synchronized with two force platforms (OR6-6,

170

Advanced Mechanical Technologies Inc.) visually concealed in the floor of a ten metre walkway

171

and sampling data at 1200 Hz. Walking speed was monitored by using photoelectric timing

172

gates. In instances where the self-selected walking speed at follow-up differed by more than +/-

173

5% from baseline, additional trials constrained to this range were conducted and analyzed.

174

Participants were provided no instructions pertaining to walking mechanics during any of the

175

gait assessments. In particular, participants were told to walk “naturally” and no mention of toe-

176

out angle was made during these assessments.

M AN U

TE D

EP

AC C

177

SC

165

178

Inverse dynamics techniques and commercially available software (Orthotrak, Motion Analysis

179

Corp.) were used to calculate gait variables including: KAM (early stance peak, late stance peak,

180

impulse), peak external knee flexion moment (KFM) during stance, and mean toe-out angle

181

(angle between a line connecting the heel and second metatarsal markers with the forward

182

progression of the body) during foot-flat. Trials in which a definitive late stance peak KAM was

8

ACCEPTED MANUSCRIPT

183

not identifiable were excluded from analysis of that variable only. The mean value from five

184

trials was calculated for each variable.

185 Statistical analysis

187

Differences in outcomes pertaining to the feasibility or completion of the intervention were

188

assessed using either paired t-tests (weekly step counts) or repeated measures analysis of

189

variance (perceived difficulty, exertion, toe-out error, compliance). All biomechanical and

190

clinical outcomes were assessed for differences between baseline and follow-up using paired t-

191

tests. The main outcomes included pain (WOMAC and NRS) and overall symptoms (total

192

WOMAC), as well as self-selected toe-out angle, late stance peak KAM, and KAM impulse

193

calculated from raw time (i.e. non-time normalized) values. Other relevant outcomes included

194

the timed stair climb and other biomechanical variables that may increase with toe-out gait

195

modification and influence total knee joint load (peak KFM and early stance peak KAM). All

196

statistical analyses were conducted using the Statistical Package for the Social Sciences (v. 21,

197

SPSS Inc.).

EP

198

TE D

M AN U

SC

RI PT

186

Results

200

Between October 2012 and February 2013, 43 individuals underwent telephone screening for the

201

study (Figure 3). Twenty-two of these individuals underwent radiographic screening, and 17

202

underwent baseline testing. One individual became unable to attend any training sessions and

203

dropped out prior to their first week of training, leaving 16 individuals (7 males; age = 64.8 +/-

204

10.4 years; BMI = 29.9 +/- 6.8 kg/m2) who began gait modification training. Four participants

205

exhibited mild radiographic signs of OA (KL grade 2), nine participants exhibited moderate OA

AC C

199

9

ACCEPTED MANUSCRIPT

206

(KL 3), and three participants exhibited severe knee OA (KL 4). One participant was unable to

207

continue with training after Week 5 due to an acute injury to the contralateral knee unrelated to

208

the intervention.

RI PT

209

Attendance at the training sessions was high (85 of 90 sessions attended; 94%) and participants

211

were able to achieve toe-out angles similar to the targeted 10 degree increase compared to

212

baseline (overall mean toe-out error averaged across all 85 training sessions = 2.6 +/- 2.1 degrees

213

below the targeted ten degree increase) with moderate perceived difficulty and exertion (Table

214

1). Further, self-confidence in the participants’ perceived ability to maintain the increased toe-out

215

angle outside the laboratory was moderate to high (overall mean throughout the intervention =

216

6.9 +/- 2.0) and improved significantly over the duration of the intervention (Week 1 = 5.8 +/-

217

2.5, Week 10 = 7.9 +/- 2.3; p < 0.01).

TE D

218

M AN U

SC

210

Five participants reported an adverse event during training, as measured using open-ended

220

questions asked throughout each training session. These adverse events were all joint discomfort

221

experienced during the first two sessions, and none of which lasted more than two sessions. This

222

included discomfort in the contralateral hip or knee (n=1 each) as well as the ipsilateral ankle,

223

knee, or hip (n=1 each). Outside of the laboratory, there were five reports of increased joint

224

discomfort at the ipsilateral knee (n=2) as well as the ipsilateral ankle or hip (n=1 each) or

225

contralateral knee (n=1). No discomfort lasted longer than two weeks. Finally, weekly step

226

counts were consistent across the intervention, with no significant difference between Week 1

227

(46 393 +/- 16 144 steps) and Week 10 (41 735 +/- 23 578 steps) (p = 0.59).

AC C

EP

219

228

10

ACCEPTED MANUSCRIPT

As a group, participants exhibited significant increases in self-selected toe-out angle and

230

significant differences in late stance KAM and self-reported pain following the intervention

231

compared to baseline (Table 2). Specifically, significant decreases in the WOMAC pain subscale

232

(p = 0.02), total WOMAC score (p = 0.02), and average knee pain as measured using the NRS (p

233

< 0.001) were observed, as were significant decreases in the late stance peak KAM (p = 0.04).

234

Changes at follow-up equated to reductions in excess of 28% for pain (28.4% reduction for

235

WOMAC pain, 44.4% reduction in NRS pain), 28.5% for overall symptoms (WOMAC total

236

score) and 10.5% for the late stance peak KAM. Thirteen of the 15 participants (87 %) reported

237

that their overall symptoms were “slightly better” or “much better” following the intervention

238

compared to baseline, with the same number reporting these responses for change in pain alone,

239

while 11 of 15 (73%) reported that their physical function was “slightly better” or “much better”.

240

No gait variable other than the late stance KAM or toe-out angle was significantly altered (p >

241

0.05), and participants completed the timed stair climb test in a similar time at both testing

242

sessions (p = 0.45). Of the 75 total trials analyzed for each testing session, the number of trials

243

without a definitive late stance peak KAM was two and four for the baseline and follow-up

244

assessments, respectively.

EP

245

TE D

M AN U

SC

RI PT

229

Discussion

247

Results from the current study provide evidence as to the feasibility and initial indications of

248

potential benefits of toe-out gait modification training in individuals with medial knee OA. This

249

study showed that toe-out gait modification performed over ten weeks can be accurately

250

achieved with minimal difficulty and does not produce any adverse effects at any lower limb

251

joint as reported by the participants in the short-term. Further, these preliminary results suggest

AC C

246

11

ACCEPTED MANUSCRIPT

that prolonged gait modification focusing on increasing toe-out angle does not have a deleterious

253

effect on symptoms and joint loading parameters relevant to medial knee OA. Since gait

254

modification treatment is part of standard clinical practice for a variety of musculoskeletal

255

pathologies and is commonly delivered with minimal resources and expense (for example, using

256

a mirror), it represents an important treatment strategy with potential implications on the growing

257

personal and economic burdens of knee OA. That said, methods to provide toe-out gait retraining

258

outside the laboratory setting to people with knee OA must be identified.

SC

RI PT

252

259

This study showed that the uptake of gait modification can be successfully achieved over the

261

short-term (10 weeks) by individuals with knee OA with minimal difficulty, and the changes in

262

self-selected walking parameters can be maintained in the absence of biofeedback of

263

performance. Though participants expectedly reported moderate amounts of difficulty at the first

264

training session (mean of 5 out of 10), this improved significantly over the course of the

265

intervention (mean of 2.9 out of 10 by the sixth training session in Week 9). Changes in self-

266

selected toe-out angle observed at the follow-up assessment were similar to those experienced

267

during the training. Specifically, participants exhibited an average of slightly more than eight

268

degrees of toe-out increase during training (mean error of 2.6 degrees less than the target 10

269

degree increase), which translated to an improvement of approximately seven degrees in the

270

amount of self-selected toe-out during over ground walking. These findings suggest that

271

successful motor learning was achieved, and that practice with some form of biofeedback was

272

likely beneficial to overall changes in gait. Importantly, this study also showed that the number

273

of adverse events associated with toe-out gait modification was minimal and similar to those

274

expected when commencing a new exercise program. These included muscle and joint

AC C

EP

TE D

M AN U

260

12

ACCEPTED MANUSCRIPT

275

discomfort that were experienced for no more than 7-10 days. Future research with longer

276

follow-up periods is needed to better identify any negative consequences of toe-out gait

277

modification in people with knee OA.

RI PT

278

Though limited by the small sample size for this pilot study, these findings appear to support the

280

only other published report of repeated gait modification training for those with knee OA 22.

281

Shull et al found that six weeks of toe-in gait modification delivered via haptic feedback based

282

on tibial rotation produced significant changes in pain and function, as well as the overall peak

283

KAM, immediately and one-month post-intervention in 12 people with knee OA. They found a

284

20% reduction in the overall peak KAM, which is substantially larger than the 10% reduction

285

observed in the late stance KAM in the present study, despite the shorter intervention length than

286

the current study. This suggests that modest changes in knee OA symptoms may be achieved

287

with small decreases in either the first or second peak KAM. However, our finding of a 1.9/10

288

reduction in self-reported pain as measured using a numerical rating scale is still smaller than the

289

minimal clinically important improvement (2.0/10) for knee OA 32. Therefore, the overall

290

clinical significance of these findings is unknown and any firm conclusions regarding symptom

291

improvement cannot be made.

M AN U

TE D

EP

AC C

292

SC

279

293

Importantly, the small sample size in this pilot study precluded the identification of any

294

definitive trend in the biomechanical and clinical data due to heterogeneity of results. For

295

example, the hypothesized relationship between mechanical loading (i.e. KAM) changes with

296

toe-out and a reduction in knee joint pain was unable to be directly confirmed. Indeed, there was

297

variation in biomechanical and clinical outcomes in this study, despite the finding that all but two

13

ACCEPTED MANUSCRIPT

participants reported an increase in toe-out angle over the intervention. Only seven participants

299

exhibited an observable change in late stance KAM and self-reported reductions in knee pain.

300

Though we observed a small, inverse relationship (r = -0.15) between the change in toe-out angle

301

and change in late stance KAM, this relationship was not statistically significant. Given the

302

complexity between gait characteristics and KAM values (for example, other factors such as

303

lateral trunk lean have been shown to influence KAM values) and the myriad factors known to

304

influence self-reported pain, further research with larger sample sizes and longer intervention

305

times is necessary to best understand the biomechanical and clinical benefits of toe-out gait

306

modification.

M AN U

SC

RI PT

298

307

Previous studies have shown that decreasing the toe-out angle (or toeing-in) predominantly

309

improves the early stance KAM with minimal effect on the late stance KAM, while increasing

310

the toe-out angle has the opposite effect 18-19, 33. This has been shown to occur due to a

311

lateralization of the centre of pressure (COP), and subsequent GRF vector origin, that is more

312

prevalent when the COP is situated anteriorly during late stance compared to under the heel

313

during early stance 34. Though much of the known clinical validation work on the KAM has

314

focused on the overall peak value or the KAM impulse 10, 13-14, 35-36, Chang et al showed a

315

decreased risk of OA progression with larger toe-out angles and lower late stance KAM values 21

316

suggesting important clinical implications related to toe-out angle and late stance knee joint load.

317

Further work to determine the clinical relevance of a reduced late stance KAM is warranted.

318

However, despite the significant decrease in the late stance KAM, we only observed a small, yet

319

not statistically significant reduction in the KAM impulse. Given that the peak KAM is typically

320

smaller in late stance compared to early stance, it is possible that changes in KAM impulse may

AC C

EP

TE D

308

14

ACCEPTED MANUSCRIPT

be linked more closely to the overall peak KAM. Conversely, the lack of significant difference in

322

KAM impulse values may have been limited by the sample size or the magnitude of observed

323

change in self-selected toe-out angle. Finally, as the magnitude of reduction in the KAM that

324

equates to a clinically-significant amount is unknown, as with the clinical outcomes in this study,

325

conclusions regarding structural improvements must be made with caution.

RI PT

321

326

Though not statistically significant (p = 0.67), participants exhibited on average, a 10% increase

328

in the peak knee flexion moment. This finding is consistent with previous studies suggesting that

329

decreases in frontal plane moments with toe-out may be associated with increases in sagittal

330

plane moments 34, thus negating any benefits on total knee joint loading. This has been

331

confirmed in a single-subject case study examining a bi-planar gait modification showing that a

332

reduction in the overall peak KAM magnitude did not necessarily result in a reduction in medial

333

tibiofemoral compartment joint loading as measured directly with an instrumented prosthesis 37.

334

Though results from a single gait analysis session must be interpreted with extreme caution,

335

these findings do point to limitations of relying on a single peak value. Importantly, the benefits

336

of increasing toe-out on late stance KAM values as well as patient-reported outcomes observed

337

in this study must be weighed against the potential consequence to sagittal plane biomechanics

338

and overall knee joint load.

M AN U

TE D

EP

AC C

339

SC

327

340

Despite the favourable outcomes found in this study, there are a number of limitations that must

341

be taken into account and should be addressed in future studies. First, the exploratory nature of

342

this pilot study was limited by a small sample size. However, these initial findings do support

343

future studies with larger sample sizes and/or control groups. Indeed, another limitation of the

15

ACCEPTED MANUSCRIPT

current study was the lack of a control group for purposes of comparison. Though reductions in

345

KAM would not necessarily be expected from a placebo effect or from a walking program alone,

346

aerobic exercise has been shown previously to improve symptoms 31. Thus, it is unknown

347

whether the changes in symptoms observed in the present study can be attributed solely to the

348

gait modification. However, a lack of increase in weekly step counts over the duration of the

349

intervention suggests that study participants were not more physically active. Further, our

350

changes in pain (effect sizes of 0.7-1.0) are greater than those suggested by Shull et al to be

351

expected due to the placebo effect (effect size of approximately 0.5) 22. Thus, when combined

352

with similar findings from Shull et al 22, this study provides a rationale for conducting a more

353

resource-intensive randomized controlled trial to best examine the impact of gait modification on

354

those with knee OA. Finally, though participants were asked to subjectively report any adverse

355

effects at other lower limb joints, analysis of joint kinetics at other lower limb joints was not

356

undertaken. As a result, the true consequences of increasing toe-out throughout the lower limb

357

kinetic chain remain unclear.

SC

M AN U

TE D

358

RI PT

344

Though this pilot, feasibility study shows promising findings, it is clear that more research is

360

needed. For example, it is possible that ten weeks and/or an intended change in toe-out angle of

361

ten degrees may be insufficient to optimize biomechanical and clinical outcomes. As discussed

362

above, we observed much variability in the biomechanical and clinical outcomes. Longer

363

performance of the gait modification or larger increases in toe-out angle may produce the

364

intended combination of biomechanical and clinical improvement desired in most participants.

365

Future studies in this area should also identify ways to better generalize the results and determine

366

effectiveness. For example, inclusion of a retention testing session after the intervention would

AC C

EP

359

16

ACCEPTED MANUSCRIPT

permit the examination of any carry-over effects of the intervention, in particular without the use

368

of real-time biofeedback. Future research should also identify more feasible methods of

369

providing the gait modification training. Though beneficial to best standardize the change in toe-

370

out angle, real-time biofeedback in association with motion capture systems is not widely

371

available clinically. However, as pointed out by Shull et al 22, though it is likely that the use of

372

some form of biofeedback can assist in the training of gait modification, it is not necessarily

373

imperative to achieve changes in gait biomechanics as evidenced by the absence of biofeedback

374

for the vast majority of the time over the course of the ten week intervention. Identification of

375

accessible methods of delivering gait modification outside the laboratory setting would greatly

376

enhance the clinical applicability of this treatment strategy. As gait modification is already a

377

commonly used component of rehabilitation for a number of different pathologies (e.g. stroke,

378

brain injury, patellofemoral pain syndrome) 38-40, the capacity for gait modification delivery to

379

those with knee OA does exist.

TE D

M AN U

SC

RI PT

367

380

In conclusion, we found that ten weeks of targeted toe-out gait modification could be safely

382

achieved by older individuals with knee OA and that some statistically significant changes in

383

biomechanics and symptoms may be realized.. Though these results are based on an exploratory

384

study with a relatively small sample size, if findings can be improved in future controlled trials

385

and delivered using less resource-intensive modalities (e.g. mirror, raw video), gait modification

386

would represent a new treatment approach that would not only have clinical and biomechanical

387

implications, but could have a significant impact on the growing economic burden of knee OA.

AC C

EP

381

388 389

Acknowledgements

17

ACCEPTED MANUSCRIPT

390

Support was provided to Judit Takacs by the Canadian Institutes of Health Research and to

391

Michael A. Hunt by the Michael Smith Foundation for Health Research.

392 Author Contributions

394

M. Hunt was involved in study design, data analysis; drafting and approval of the manuscript.

395

J. Takacs was involved in study design, data collection, drafting and approval of the manuscript.

RI PT

393

397

SC

396

J. Takacs and M. Hunt take responsibility for the integrity of the work as a whole

399

Conflict of Interest

400

None.

M AN U

398

401

405 406 407 408 409 410

EP

404

AC C

403

TE D

402

411 412

18

ACCEPTED MANUSCRIPT

References

414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456

1.

Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 disease and injuries 1990-2010: a systematic analysis for the Global Burden of Diseases Study 2010. The Lancet 2012;380:2163-96.

2.

Andriacchi T. Dynamics of knee malalignment. Orthopedic Clinics of North America 1994;25:395-403.

3.

Andriacchi T, Mundermann A, Smith R, Alexander E, Dyrby C, Koo S. A framework for the in vivo pathomechanics of osteoarthritis at the knee. Annals of Biomedical Engineering 2004;32:447-57.

4.

Radin E, Paul I. Response of joints to impact loading. I. In vitro wear. Arthritis and Rheumatism 1971;14:356-62.

5.

Radin EL, Parker HG, Pugh JW, Steinberg RS, Paul IL, Rose RM. Response of joints to impact loading III. Relationship between trabecular microfractures and cartilage degeneration. Journal of Biomechanics 1973;6:51-57.

6.

Andriacchi T, Lang P, Alexander E, Hurwitz D. Methods for evaluating the progression of osteoarthritis. Journal of Rehabilitation Research and Development 2000;37:163-70.

7.

Andriacchi T, Mundermann A. The role of ambulatory mechanics in the initiation and progression of knee osteoarthritis. Current Opinion in Rheumatology 2006;18:514-8.

8.

Hurwitz D, Ryals A, Case J, Block J, Andriacchi T. The knee adduction moment during gait in subjects with knee osteoarthritis is more closely correlated with static alignment than radiographic disease severity, toe out angle and pain. Journal of Orthopaedic Research 2002;20:101-08.

9.

Specogna A, Birmingham T, Hunt M, Jones I, Jenkyn T, Fowler P, et al. Radiographic measures of knee alignment in patients with varus gonarthrosis: effect of weightbearing status and associations with dynamic joint load. American Journal of Sports Medicine 2007;35:65-70.

11.

SC

M AN U

TE D

EP

AC C

10.

RI PT

413

Sharma L, Hurwitz D, Thonar E, Sum J, Lenz M, Dunlop D, et al. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis and Rheumatism 1998;41:1233-40.

Baliunas A, Hurwitz D, Ryals A, Karrar A, Case J, Block J, et al. Increased knee joint loads during walking are present in subjects with knee osteoarthritis. Osteoarthritis and Cartilage 2002;10:573-9.

19

ACCEPTED MANUSCRIPT

Jackson B, Teichtahl A, Morris M, Wluka A, Davis S, Cicuttini F. The effect of the knee adduction moment on tibial cartilage volume and bone size in healthy women. Rheumatology 2004;43:311-14.

13.

Miyazaki T, Wada M, Kawahara H, Sato M, Baba H, Shimada S. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Annals of the Rheumatic Diseases 2002;61:617-22.

14.

Bennell KL, Bowles KA, Wang Y, Cicuttini F, Davies-Tuck M, Hinman RS. Higher dynamic medial knee load predicts greater cartilage loss over 12 months in medial knee osteoarthritis. Annals of the Rheumatic Diseases 2011;70:1770-74.

15.

Simic M, Hinman RS, Wrigley TV, Bennell KL, Hunt MA. Gait modification strategies for altering medial knee joint load: A systematic review. Arthritis Care and Research 2010;63:405-26.

16.

Hunt MA, Birmingham TB, Bryant D, Jones I, Giffin JR, Jenkyn TR, et al. Lateral trunk lean explains variation in dynamic knee joint load in patients with medial compartment knee osteoarthritis. Osteoarthritis Cartilage 2008;16:591-9.

17.

Lin CJ, Lai KA, Chou YL, Ho CS. The effect of changing the foot progression angle on the knee adduction moment in normal teenagers. Gait & Posture 2001;14:85-91.

18.

Lynn S, Kajaks T, Costigan P. The effect of internal and external foot rotation on the adduction moment and lateral-medial shear force at the knee during gait. Journal of Science and Medicine in Sport 2008;11:444-51.

19.

Lynn SK, Costigan PA. Effect of foot rotation on knee kinetics and hamstring activation in older adults with and without signs of knee osteoarthritis. Clinical Biomechanics 2008;23:779-86.

20.

Guo M, Axe MJ, Manal K. The influence of foot progression angle on the knee adduction moment during walking and stair climbing in pain free individuals with knee osteoarthritis. Gait & Posture 2007;26:436-41.

21.

22.

23.

EP

TE D

M AN U

SC

RI PT

12.

AC C

457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502

Chang A, Hurwitz D, Dunlop D, Song J, Cahue S, Hayes K, et al. The relationship between toe-out angle during gait and progression of medial tibiofemoral osteoarthritis. Annals of the Rheumatic Diseases 2007;66:1271-75. Shull PB, Silder A, Shultz R, Dragoo JL, Besier TF, Delp SL, et al. Six-week gait retraining program reduces knee adduction moment, reduces pain, and improves function for individuals with medial compartment knee osteoarthritis. Journal of Orthopedic Research 2013;31:1020-25. Kellgren JH, Lawrence JS. Radiological assessment of osteoarthrosis. Annals of the Rheumatic Diseases 1957;16:494-501. 20

ACCEPTED MANUSCRIPT

Hunt MA, Simic M, Hinman RS, Bennell KL, Wrigley TV. Feasibility of a gait retraining strategy for reducing knee joint loading: Increased trunk lean guided by real-time biofeedback. Journal of Biomechanics 2011;44:943-47.

25.

Barrios J, Crossley K, Davis I. Gait retraining to reduce the knee adduction moment through real-time visual feedback of dynamic knee alignment. Journal of Biomechanics 2010;43:2208-13.

26.

Simic M, Hunt MA, Bennell KL, Hinman RS, Wrigley TV. Trunk lean gait modification and knee joint load in people with medial knee osteoarthritis: The effect of varying trunk lean angles. Arthritis Care and Research 2012;64:1545-53.

27.

Kadaba MP, Ramakrishnan HK, Wootten ME, Gainey J, Gorton G, Cochran GV. Repeatability of kinematic, kinetic, and electromyographic data in normal adult gait. Journal of Orthopaedic Research 1989;7:849-60.

28.

Borg G. Psychophysical bases of perceived exertion. Medicine and Science in Sports and Exercise 1982;14:377-81.

29.

Crossley KM, Bennell KL, Cowan SM, Green S. Analysis of outcome measures for persons with patellofemoral pain: which are reliable and valid? Archives of Physical Medicine and Rehabilitation 2004;85:815-22.

30.

Fisher NM, Gresham GE, Abrams M, Hicks J, Horrigan D, Pendergast DR. Quantitative effects of physical therapy on muscular and functional performance in subjects with osteoarthritis of the knees. Archives of Physical Medicine and Rehabilitation 1993;74:840-47.

31.

Ettinger WH, Burns R, Messier SP, Applegate W, Rejeski J, Morgan T, et al. A randomized trial comparing aerobic exercise and resistance exercise with a health education program in older adults with knee osteoarthritis. JAMA 1997;277:25-31.

32.

Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement. Annals of the Rheumatic Diseases 2005;64:29-33.

33.

34.

EP

TE D

M AN U

SC

RI PT

24.

AC C

503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548

Shull PB, Shultz R, Silder A, Dragoo JL, Besier TF, Cutkosky MR, et al. Toe-in gait reduces the first peak knee adduction moment in patients with medial compartment knee osteoarthritis. Journal of Biomechanics 2013;46:122-28. Jenkyn TR, Hunt MA, Jones IC, Giffin JR, Birmingham TB. Toe-out gait in patients with knee osteoarthritis partially transforms external knee adduction moment into flexion moment during early stance phase of gait: a tri-planar kinetic mechanism. Journal of Biomechanics 2008;41:276-83. 21

ACCEPTED MANUSCRIPT

Thorp L, Sumner D, Block J, Moisio K, Shott S, Wimmer M. Knee joint loading differs in individuals with mild compared with moderate medial knee osteoarthritis. Arthritis and Rheumatism 2006;54:3842-49.

36.

Prodromos C, Andriacchi T, Galante J. A relationship between gait and clinicial chanages following high tibial osteotomy. Journal of Bone and Joint Surgery 1985;67A:1188-94.

37.

Walter JP, D'Lima DD, Colwell CW, Fregly BJ. Decreased knee adduction moment does not guarantee decreased medial contact force during gait. Journal of Orthopedic Research 2010;28:1348-54.

38.

Tate JC, Milner CE. Real-time kinematic, temporospatial, and kinetic biofeedback during gait retraining in patients: a systematic review. Physical Therapy 2010;90:1123-34.

39.

Noehren B, Scholz J, Davis I. The effect of real-time gait retraining on hip kinematics, pain and function in subjects with patellofemoral pain syndrome. British Journal of Sports Medicine 2011;45:691-96.

40.

Crowell HP, Davis IS. Gait retraining to reduce lower extremity loading in runners. Clinical Biomechanics 2011;26:78-83.

EP

TE D

M AN U

SC

RI PT

35.

AC C

549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593

22

ACCEPTED MANUSCRIPT

RI PT

SC

M AN U TE D EP

601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632

Table 1. Mean (sd) participant training data over the six training sessions. Toe-out error corresponds to the average difference between the measured toe-out angle and target angle across four time points at each session. Difficulty refers to participants’ self-reported difficulty in achieving the target toe-out angle during training (0 = “no difficulty”, 10 = “unable to perform”). Pain was assessed using an 11 point numerical rating scale (0 = “no pain”, 10 = “worst pain imaginable”) at four time points during each training session and the mean for each session was calculated. Rating of perceived exertion (RPE) was assessed using the 15-point Borg scale. Training Session Week 1 Week 2 Week 3 Week 5 Week 7 Week 9 2.9 (2.3) 3.2 (2.1) 2.2 (1.7) 2.6 (2.1) 2.9 (2.1) 2.0 (1.8) Toe-out error (º) 5.0 (2.4) 4.2 (2.5) 3.9 (2.4) 3.7 (2.2) 3.4 (2.0) 2.9 (1.9) Difficulty (0-10) 3.1 (2.5) 2.3 (2.3) 2.5 (2.2) 2.4 (1.9) 1.6 (1.8) 2.0 (1.5) Pain (0-10) 11.4 (2.0) 12.1 (2.0) 11.4 (1.9) 11.2 (1.2) 10.6 (1.7) 10.6 (2.1) RPE (6-20)

AC C

594 595 596 597 598 599 600

23

ACCEPTED MANUSCRIPT

636 637 638 639 640

Table 2. Group mean (sd) biomechanical, objective physical function, and self-report data during the intervention. Mean differences (95% confidence intervals (CI)) are also included to compare Baseline and Follow-up values.

Timed stair climb (seconds)

5.58 (3.10)

5.13 (1.44)

WOMAC pain subscale (0-20) WOMAC total score (0-96) NRS pain (0-10)

7.4 (3.4) 36.9 (14.8) 4.5 (1.7)

5.3 (2.9) * 26.4 (13.5) * 2.6 (1.8) *

646 647 648 649

-2.1 (-3.9, -0.4) -10.5 (-18.9, -2.1) -1.9 (-2.8, -1.0)

0.45

0.02 0.02 <0.001

TE D

645

-0.45 (-1.67, 0.77)

EP

644

0.90 <0.001 0.12 0.04 0.20 0.67

AC C

643

p-value

BW = body weight; ht = height; sec = second, KAM = knee adduction moment; KFM = knee flexion moment; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index; NRS = numerical rating scale. * denotes significant difference (p < 0.05) compared to Baseline (Week 0) values

641 642

Mean Difference (95% CI) 0.00 (-0.06, 0.05) 6.66 (4.08, 9.23) -0.26 (-0.60, 0.07) -0.30 (-0.57, -0.02) -0.08 (-0.20, 0.04) 0.13 (-0.52, 0.79)

RI PT

Follow-up (Week 11) 1.14 (0.16) 11.41 (6.46) * 3.19 (0.72) 2.57 (0.84) * 1.24 (0.34) 1.51 (1.29)

SC

Walking speed (m/s) Toe-out angle during foot flat (º) Early stance peak KAM (%BW*ht) Late stance peak KAM (%BW*ht) KAM impulse (%BW*ht*sec) Peak KFM (%BW*ht)

Baseline (Week 0) 1.14 (0.18) 4.75 (6.59) 3.45 (0.82) 2.87 (0.92) 1.33 (0.29) 1.38 (1.36)

M AN U

633 634 635

650 651 652

24

653 654 655

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 1. Protractor device used to calibrate actual toe-out angle with calculated toe-out angle based on motion analysis marker data.

656 657

25

664 665 666 667

Figure 2. During training, participants were instructed to match their calculated toe-out angle (solid line) with the target angle corresponding to a 10 degree increase in toe-out angle (thick vertical line). Specifically, participants were instructed to focus on the straight sections of the real-time toe-out angle – which corresponded to stance phase toe-out angle – rather than the curved portions – which corresponded to swing phase toe-out angle.

EP

659 660 661 662 663

AC C

658

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

668 669 670 26

ACCEPTED MANUSCRIPT

671 Assessed for eligibility by phone (n=43)

672

Other pain/ pathology (n=6) Knee pain <3/10 (n=5)

673

Too far to travel/unable to do multiple sessions (n=3) Assessed for eligibility by xray (n=22)

676

678

Excluded (n=5)

679

SC

Previous hip replacement (n=1)

677

No longer interested (n=2)

Baseline testing (n = 17)

682

M AN U

Did not meet OA inclusion criteria (n=1)

680 681

Unable to walk continuously for 30 minutes (n=2)

RI PT

Pain predominantly in patellofemoral joint (n=4)

674 675

Excluded (n=21)

Acute injury prior to baseline testing (n=2)

Lost to follow-up (n=2)

Unable to attend sessions (n=1)

TE D

683

Acute injury (n=1)

Analyzed (n=15)

686 687 688 689 690

Figure 3. Flow chart of eligible and recruited participants.

EP

685

AC C

684

691 692 693 27

ACCEPTED MANUSCRIPT

694 695 696

RI PT

697 698 699

SC

700 701

M AN U

702 703 704 705

709 710 711 712 713 714 715 716 717 718 719 720 721

EP

708

AC C

707

TE D

706

Figure 4. Individual participant data for self-selected toe-out angle (top panel), late stance peak KAM (middle panel), and knee joint pain as measured using an 11-point numerical rating scale (bottom panel). White bars show data from the baseline assessment, while black bars pertain to follow-up data for each participant. Data for toe-out angle and late stance peak KAM pertain to the mean (95% CI) values from the five walking trials at each testing session. Negative toe-out angles denote self-selected toe-in. 28