- Email: [email protected]
Predictors and Functional Implications of Change in Leg Length After Total Knee Arthroplasty
Accepted Manuscript Predictors and Functional Implications of Change in Leg Length Following Total Knee Arthroplasty Dr Jason Chinnappa, MBBS, MS (Ortho), Dr Darren B. Chen, MBBS (Hons) FRACS FAOrthA, Professor Ian A. Harris, MBBS, FRACS(Ortho), FAHMS, MMed(Clin Epi), PhD, Dr Samuel J. MacDessi, MBBS (Hons) FRACS FAOrthA PII:
S0883-5403(17)30320-0
DOI:
10.1016/j.arth.2017.04.007
Reference:
YARTH 55809
To appear in:
The Journal of Arthroplasty
Received Date: 9 February 2017 Revised Date:
1 April 2017
Accepted Date: 5 April 2017
Please cite this article as: Chinnappa J, Chen DB, Harris IA, MacDessi SJ, Predictors and Functional Implications of Change in Leg Length Following Total Knee Arthroplasty, The Journal of Arthroplasty (2017), doi: 10.1016/j.arth.2017.04.007. 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
PREDICTORS AND FUNCTIONAL IMPLICATIONS OF CHANGE IN LEG LENGTH FOLLOWING TOTAL KNEE ARTHROPLASTY Dr Jason Chinnappa MBBS, MS (Ortho)1 Dr Darren B Chen MBBS (Hons) FRACS FAOrthA1,2
Dr Samuel J MacDessi MBBS (Hons) FRACS FAOrthA 1,2
RI PT
Professor Ian A Harris MBBS, FRACS(Ortho), FAHMS, MMed(Clin Epi), PhD3
1. Department of Orthopaedic Surgery, The Canterbury Hospital, Canterbury,
SC
NSW, Australia
2. Sydney Knee Specialists, St George Private Hospital, Sydney, NSW, Australia
School, UNSW, Australia
M AN U
3. Ingham Institute of Applied Medical Research, South Western Sydney Clinical
Please address all correspondence to
Dr Samuel J MacDessi
TE D
Sydney Knee Specialists
Suite 8, 19 Kensington Street
Kogarah NSW 2217 Australia
EP
Phone: +61 2 8307 0333 Fax: +61 2 8307 0334
AC C
Email: [email protected]
ACCEPTED MANUSCRIPT 1
PREDICTORS AND FUNCTIONAL IMPLICATIONS OF CHANGE IN LEG
2
LENGTH FOLLOWING TOTAL KNEE ARTHROPLASTY
3 ABSTRACT
5
Background: Leg lengthening occurs in 83% of primary TKA. The effects of leg
6
length discrepancy (LLD) on THA patients are well established. However, patient
7
function and satisfaction associated with LLD following primary TKA has not
8
been analyzed. This study aimed to quantify the magnitude of limb lengthening,
9
identify radiographic and perceived LLD, and correlate these with predictive
RI PT
4
factors and functional outcomes in a series of TKA patients.
11
Methods: Patients undergoing primary TKA who met inclusion criteria were
12
prospectively enrolled in this study. Leg length measurements were measured
13
on standardised pre and post-operative long leg radiographs. Patients completed
14
pre-operative and six month post-operative KSS and KOOS functional scores as
15
well as a post-operative satisfaction and customised leg length specific functional
16
questionnaire.
17
Results: 91 patients undergoing TKA surgeries were included. Mean overall
18
lengthening was 3.5mm (range -31.0 to 21.4mm, SD 8.4) with 77% of limbs
19
lengthened. 89% of patients had no LLD (defined as 10mm or greater) following
20
TKA. Post-operative radiographic LLD was associated with increased pre-
21
operative LLD (p<0.001). Perceived post-operative LLD was associated with
22
female gender (p=0.02), decreased satisfaction (18% vs. 84%, p<0.001) and
23
poorer functional score changes. Perceived LLD was not associated with
24
radiographic LLD.
25
Conclusions: Radiographic lengthened LLD is uncommon following primary TKA
26
(11%) and does not correlate with perceived LLD. Patients with perceived LLD
27
have decreased satisfaction and functional score improvements following TKA
28
surgery.
29
Level of Evidence: 2b
30
Keywords
31
Total Knee Replacement; Leg Length Inequality; Patient Reported Outcomes;
32
Patient Satisfaction
AC C
EP
TE D
M AN U
SC
10
1
ACCEPTED MANUSCRIPT INTRODUCTION
34
Leg length inequality is a well-described complication of hip arthroplasty (THA)
35
surgery. Patients with leg length discrepancy (LLD) after THA have been found to
36
have increased rates of gait disturbance, hip pain [1], low back pain and nerve
37
injury [2]. Furthermore, patients with leg length discrepancy have lower
38
functional hip scores [3,4], more dissatisfaction [5] and higher rates of litigation
39
than patients with equal leg lengths [6].
40
RI PT
33
Leg length discrepancy and its effects following total knee arthroplasty (TKA)
42
have not been extensively studied. A report by Lang et al found that leg
43
lengthening was common after TKA, occurring in 83% of patients [7]. It was
44
noted that the average lengthening following unilateral TKA was 6.3mm, with a
45
range from 11mm shorter to 24mm longer. However, their study did not
46
investigate the correlation of these findings with patient-reported outcomes.
47
Another recent study by Mufty et al found that 76% of patients underwent leg
48
lengthening following revision TKA [8]. Average lengthening was 5.3mm in their
49
series, resulting in a 3.8mm longer operative side (range from 24mm shorter to
50
35mm longer). They also found positive correlation between leg lengthening and
51
clinical outcomes whilst patients with perceived LLD did not have any
52
differences in clinical or functional outcomes. Kim et al have retrospectively
53
investigated both the quantitative changes in leg length as well as the functional
54
implications of this in primary TKA, however their series was specifically limited
55
to cases of varus knee osteoarthritis [9].
M AN U
TE D
EP
AC C
56
SC
41
57
The aim of our study was to investigate the magnitude of limb lengthening and
58
frequency of perceived limb length discrepancy after primary TKA and correlate
59
this with functional outcomes and predictive factors for these variables.
60 61
METHODS
62
Patients and setting
63
All patients who were undergoing elective primary TKA by two knee
64
arthroplasty surgeons (SJM, DBC) at a single institution were prospectively
65
selected for inclusion in the study. Patients with pre-existing extra-articular
2
ACCEPTED MANUSCRIPT deformity due to fracture or previous surgery were excluded from the study.
67
Patients with known leg length inequality due to other causes (radiographic leg
68
lengthening post THA >5mm and longstanding leg length inequality > 5 years
69
requiring orthotics) were excluded. Patients undergoing bilateral TKA (staged
70
within the study period or simultaneous) were excluded from the study. Finally
71
patients who did not undergo radiographs according to the protocol for the
72
study were also excluded from analysis.
73
RI PT
66
Surgical Procedure
75
All surgeries were performed using a medial parapatellar approach, with the aim
76
to restore a neutral mechanical axis followed by appropriate ligament balancing.
77
One of two cemented posterior-stabilised prostheses were used in all cases
78
(Smith and Nephew Legion, Memphis, TN, USA or Amplitude SCORE, Valence,
79
France). Bone resections were based on intramedullary jigs or computer
80
navigated alignment. All patients underwent the same post-operative
81
mobilisation and physiotherapy protocol.
82
M AN U
SC
74
Study variables and measurement
84
Patient demographic characteristics including age, sex, BMI, knee pathology and
85
operative side were recorded at time of initial pre-operative consultation. Pre
86
and post-operative flexion contractures were measured by the treating surgeon
87
using a goniometer and recorded when present.
EP
88
TE D
83
Radiographic Measurements
90
All patients underwent pre-operative and post-operative full length standing
91
radiographs at a single radiology centre. Pre-operative radiographs were
92
performed within six months of surgery, and post-operative radiographs were
93
performed two days post operation.
AC C
89
94 95
Leg length differences were measured from the top of the femoral head to the
96
center of the tibial plafond (Figure 1). The difference in leg lengths between the
97
operated and non-operated limb was recorded in millimeters (mm) with
3
ACCEPTED MANUSCRIPT 98
shortening of the operated side being recorded as a negative value and
99
lengthening as a positive value.
100 Pre-operative and post-operative hip-knee angle (HKA) was measured as the
102
angle between a line from the center of the femoral head to the center of the
103
distal femur and a line between the center of the proximal tibia and the center of
104
the tibial plafond. Varus angles were given negative values and valgus angles
105
were given positive values. All measurements were performed using the center’s
106
image processor (Medway, Healthcare Imaging Services, NSW, Australia) by one
107
of the authors (JC). The non-operative side leg length was used to calculate
108
magnification changes between pre and post-operative radiographs and
109
accordingly adjust post-operative leg length measurements. 10 patient pre-
110
operative leg length radiographic measurements were repeated to check the
111
intra-observer reliability of these values. The Intraclass Correlation Coefficient
112
(ICC) was 1.00 (p <0.001), indicating excellent reliability of these measurements.
113
M AN U
SC
RI PT
101
LLD and Perceived LLD
115
LLD for the purpose of comparison was defined as a leg length difference of
116
greater than or equal to 10mm and sub-analyzed as lengthened and non-
117
lengthened groups. These values were chosen as 10mm has been shown to have
118
biokinetic implications on gait [10] and been used in a prior study looking at hip
119
arthroplasty and leg length discrepancy [5].
EP
TE D
114
120
The frequency of perceived lengthened LLD at six months post-surgery was
122
obtained from the custom questionnaire (Appendix A- Question 3). Perceived
123
and actual LLD’s were correlated.
124 125
AC C
121
Predictive factors investigated for correlation with actual or perceived LLD were:
126
-
Patient baseline characteristics: age, sex, body mass index (BMI)
127
-
Pre-operative coronal deformity
128
-
Degree of coronal deformity correction
129
-
Pre-operative LLD
130
-
Pre-operative flexion contractures in the operative knee
4
ACCEPTED MANUSCRIPT 131
-
Post-operative flexion contractures in the operative knee
132
-
Polyethylene insert thickness
133
-
Other lower limb degenerative disease or arthroplasty: osteoarthritis of either hip joint resulting in asymmetric joint space loss, or prior THA on
135
either side were recorded. Kellgren and Lawrence Grades 3 or 4
136
osteoarthritis or TKA in the contralateral knee was also recorded. The
137
Kellgren and Lawrence classification is used to categorise radiographic
138
knee osteoarthritis based on reduction of joint space, sclerosis and
139
presence of osteophytes, with grade 4 being the most severe. [11]
140
SC
RI PT
134
Patient-reported outcome measures
142
A custom-designed leg length questionnaire was created and used to assess the
143
functional implications of leg lengthening six-month post-operation (Appendix
144
A). This questionnaire was used to derive frequencies and identify patients with
145
perceived LLD, as well as assess patient satisfaction with their surgery on a 5-
146
point scale from Poor to Excellent. All patients were also completed a Knee
147
Society Score (KSS) and the Knee Injury and Osteoarthritis Outcome Score
148
(KOOS) with its included subset Western Ontario and McMaster Universities
149
Arthritis Index (WOMAC) score pre-operatively and six months post-operatively.
150
The KSS is a widely used instrument that combines functional knee scores with
151
anatomical knee alignment [12]. Scores of 80-100 are considered excellent,
152
whilst scores below 60 are considered poor. The KOOS is a validated knee score
153
in the surgical setting, which assesses patients’ pain, symptoms, function and
154
quality of life in the preceding week [13]. It utilizes 5-point Likert scales
155
resulting in extrapolated percentage scores of 0 (extreme symptoms) to 100 (no
156
symptoms) [14]. WOMAC is an arthritis-specific, validated score that is
157
commonly used in assessing symptoms from arthritic hips and knees [15].
158
Changes in KOOS, KSS and WOMAC scores were used to compare groups.
AC C
EP
TE D
M AN U
141
159 160
Statistical analysis
161
All statistical analysis was performed using SPSS Statistics Software, version 20
162
(IBM Software). Differences in variables with continuous outcomes were
163
analyzed using the independent t-test. Categorical variables were tested using
5
ACCEPTED MANUSCRIPT 164
Chi-square analysis and Fisher’s exact test. Predictive factors for LLD and
165
perceived LLD were also analyzed with multivariate analysis using binomial
166
logistic regression. Statistical significance was set at p=0.05.
167 Ethics approval
169
Informed consent was obtained from all patients. Approval to conduct the study
170
was granted by the Hunter New England Research Ethics Committee (Reference
171
15/02/18/5.07).
RI PT
168
172 RESULTS
174
Population
175
116 patients undergoing TKA surgeries were enrolled in this study. 25 patients
176
were excluded (Figure 2), leaving 91 patients for analysis.
177
M AN U
SC
173
Baseline characteristics
179
Baseline demographics for the 91 included patients undergoing TKA (44 left, 47
180
right) surgeries are provided in Table 1. Pre-operative diagnoses included 83
181
osteoarthritis cases, 5 post-traumatic arthritis cases, 2 avascular necrosis cases
182
and 1 rheumatoid arthritis case. 89% of patients completed the leg length and
183
surgery satisfaction questionnaire.
TE D
178
184
186
2% of patients pre-operatively had a leg length discrepancy of 10mm or greater.
EP
185
Degree of correction
188
Mean postoperative HKA was 0.8 degrees (range -5.6 to 7.5 degrees, SD 2.5),
189
with an overall correction of 5.8 degrees (range 0.1 – 17.3, SD 3.6).
190
AC C
187
191
Magnitude of Lengthening
192
Mean overall lengthening in this study was 3.5mm (range -31.0 to 21.4mm, SD
193
8.4) with a mean post-operative LLD of 0.4mm longer (range -42.3 to 29.9mm,
194
SD 10.0). 77% of knees were lengthened post-operatively by a mean of 6.7mm
195
(range 0.0 to 21.4mm, SD 4.9). The 23% of knees that were shortened post-
6
ACCEPTED MANUSCRIPT 196
operatively had a mean leg length change of -7.4mm (range -31 to -0.6mm, SD
197
8.5).
198 Radiographic LLD
200
Post-operatively, 89% of patients had a LLD of less than 10mm (mean –1.4mm,
201
range – 42.3 to 9.7, SD 8.7) and were considered to have normal leg lengths for
202
the study groups. 11% of patients were considered to have a LLD with the
203
operative leg more than or equal to 10mm longer (mean 15.3, range 10.3 – 29.9,
204
SD 5.9).
RI PT
199
SC
205 Perceived LLD
207
Leg length questionnaire analysis revealed 16% of patients felt they had a LLD
208
pre-operatively. Immediately post-operatively, 21% of patients felt they had a
209
LLD with 88% of these patients perceiving their operative leg to be longer. 12%
210
of these patients wore a contralateral shoe raise for perceived lengthening of the
211
operative side. Six months post operatively, 14% of patients perceived a LLD
212
with 73% of these patients feeling they were lengthened on the operative side.
213
9% required a shoe raise and 45% of these patients (6% of total cohort)
214
reported being troubled by their perceived LLD. None of the patients with a
215
perceived LLD at six months post operation had a radiographic LLD of 10mm or
216
longer.
TE D
EP
217
M AN U
206
Predictive factors for Radiographic LLD
219
Pre-operative LLD was associated with increased post-operative leg length
220
discrepancy (p<0.001), with mean pre-operative LLD in the LLD group 7.9mm
221
versus –4.4mm in the normal group. No relationship was identified between
222
radiographic post-operative LLD and sex (p=1), age (p=0.53), BMI (p=0.27),
223
polyethylene insert thickness (p= 0.11), pre-operative FFD (p=0.74), post-
224
operative FFD (p=0.59), contralateral knee degenerative disease (p= 0.51) or
225
arthroplasty (p= 0. 2) or either hip arthroplasty (p= 1), severity of pre-operative
226
angular deformity (p=0.67), or degree of angular correction (p= 0.62).
227
There were no significant correlations between any of the above factors and
228
post-operative LLD on multivariate logistic regression analysis.
AC C
218
7
ACCEPTED MANUSCRIPT 229 Predictive factors for Perceived LLD
231
Female sex was significantly associated with perceived LLD in this series as 20%
232
of females had a perceived LLD at 6 months post operation versus 3% of males
233
(p=0.04). Perceived LLD was not associated with age (p=0.98), BMI (p=0.21),
234
polyethylene insert thickness (p= 0.59), pre-operative FFD (p=0.52), post-
235
operative FFD (p=0.3), contralateral knee degenerative disease (p= 0.53) or
236
arthroplasty (p = 1) or either hip arthroplasty (p= 0.26), severity of pre-
237
operative angular deformity (p=0.39), degree of angular correction (p= 0.83) or
238
pre-operative LLD (p=.0.42).
SC
RI PT
230
239
Multivariate regression analysis also found an association between female sex
241
and perceived LLD (OR 21.76, 95% CI 1.65 – 287.31, p=0.02). No patients with
242
perceived LLD had a radiographic LLD.
M AN U
240
243 Functional outcomes
245
There were no significant associations between changes in functional scores
246
(KOOS, KSS and WOMAC) and radiographic LLD (Table 2). There were
247
differences between outcomes for the KOOS Pain, KOOS Activities of Daily Living
248
(ADL), KOOS Quality of Life (QOL) and WOMAC functional scores, with
249
significantly lower improvements in functional scores for patients with
250
perceived LLD (Table 2).
EP
TE D
244
251
Patient satisfaction
253
Overall, 75% of patients were satisfied with their operation (rated very good or
254
excellent). There was a non-statistically significant association between patient
255
satisfaction and radiographic LLD with 77% of normal radiograph patients rating
256
their operation as very good or excellent compared with 57% of patients with
257
radiographic LLD (p=0.4). With regards to perception of LLD, 18% of patients
258
with perceived LLD rated their operation as very good or excellent, significantly
259
lower than the 84% of patients without perceived LLD (p<0.001).
AC C
252
260 261
DISCUSSION
8
ACCEPTED MANUSCRIPT 262
Our finding of 77% of patients undergoing overall lengthening following TKA is
263
similar to the finding of 83% by Lang et al[7]. However, 89% of our patients had
264
a discrepancy of less than 10mm longer post-operatively. These findings show
265
that the majority of patients undergoing TKA do not have radiographic evidence
266
of significant functional LLD.
RI PT
267 The finding on univariate analysis of correlation between pre-operative LLD and
269
increased post-operative LLD is not unexpected. We are unable to identify why
270
females were more likely to perceive LLD in this series. The finding of female sex
271
being associated with perceived LLD, as well as none of the patients with
272
perceived LLD having radiographic evidence of the same, may be useful in
273
counseling patients pre-operatively.
M AN U
274
SC
268
Patients with radiographic LLD following TKA showed lower satisfaction with
276
their operations, although this finding did not reach statistical significance in this
277
series. Furthermore, this was not associated with lower functional scores in this
278
group. Kim et al also found a non-significant result with less highly satisfied
279
patients who had radiographic LLD in their series [9]. Their results also showed
280
some significant functional differences with regards to ability on stairs and KSS
281
functional score, in favor of patients with radiographic LLD <15mm. As their
282
series only looked at pre-operative varus knees with a 15mm LLD cut-off, our
283
series investigating all pre-operative alignments and a 10mm cut-off for LLD,
284
may not be directly comparable.
EP
AC C
285
TE D
275
286
Patients with perceived LLD had significantly lower satisfaction scores in our
287
series. Additionally, our study found a significant correlation between perceived
288
LLD and lower KOOS pain, KOOS ADL, KOOS QOL and WOMAC functional knee
289
scores. The other KOOS subscales and KSS scores also showed decreased
290
functional outcomes in patients with perceived LLD, however these did not reach
291
statistical significance. Our findings are consistent with those of Konyves[3] and
292
Wylde[4] that perceived LLD is associated with poorer functional outcomes in
293
THA patients. This is in contrast to the finding by Mufty that patients with actual
294
and perceived lengthening following revision TKA had better pain functional
9
ACCEPTED MANUSCRIPT 295
scores and no differences in other scores.[8] This discrepancy may be due to
296
differing expectations in patients undergoing primary and revision TKA. It is
297
unclear whether patients who are unhappy with or feel limited following their
298
primary TKA may try and attribute this to other causes such as perceived LLD.
299 This study does have some limitations that need to be considered when
301
interpreting the results. There is potential underestimation of radiographic
302
lengthening in this series, as long leg radiographs post-operatively were done on
303
day 2 following surgery. Some patients may have been unable to achieve full
304
extension by this stage and this would understate their leg length on standing
305
long leg films. However as we did not identify a correlation between LLD and
306
patients with fixed flexion deformity, this is unlikely to be clinically significant.
307
Another limitation of our study is the small number of patients with extreme
308
degrees of coronal deformity, which may limit our ability to correlate this and
309
the degree of correction with LLD. Finally, this study utilized long leg
310
radiographs to assess radiographic LLD. Long leg radiographs have been shown
311
to be comparable to Computed Tomography (CT) imaging for calculation of
312
lower limb lengths.[16] CT imaging would have allowed calculation of actual
313
limb length discrepancy independent of coronal or sagittal deformity. However
314
the added radiation exposure and the fact that coronal and sagittal deformity
315
correction that is routine with most primary TKA is most likely to affect an
316
apparent LLD make CT imaging less beneficial in this setting.
SC
M AN U
TE D
EP
317
RI PT
300
Strengths of this study include the completion rate of the functional outcome,
319
LLD and satisfaction questionnaires and our ability to correlate this with
320
radiographic and perceived LLD. These findings will be useful in being able to
321
allow surgeons to inform patients of the impact of limb lengthening in primary
322
TKA. Patients can be counseled that although they may perceive LLD, it is
323
uncommon following primary TKA and usually not associated with radiographic
324
LLD. Patients can also be informed that even those with radiographic LLD have
325
similar functional scores to those without radiographic LLD and there are no
326
significant predictive factors for radiographic LLD. Further study into patients
AC C
318
10
ACCEPTED MANUSCRIPT 327
who are not satisfied with their TKA or have poorer functional scores or
328
perceived LLD is needed to identify the interplay between these findings.
329 CONCLUSIONS
331
The majority of patients undergoing primary TKA do not have significant pre or
332
post-operative LLD. Perceived LLD following TKA correlates with female sex,
333
poorer functional outcomes and decreased satisfaction but not with radiographic
334
LLD.
RI PT
330
335 REFERENCES
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
[1] Plaass C, Clauss M, Ochsner PE, Ilchmann T. Influence of leg length discrepancy on clinical results after total hip arthroplasty--a prospective clinical trial. Hip Int 2011;21:441–9. doi:10.5301/HIP.2011.8575. [2] Della Valle CJ, Di Cesare PE. Complications of total hip arthroplasty: neurovascular injury, leg-length discrepancy, and instability. Bull Hosp Jt Dis 2001;60:134–42. [3] Konyves A, Bannister GC. The importance of leg length discrepancy after total hip arthroplasty. J Bone Joint Surg Br 2005;87:155–7. [4] Wylde V, Whitehouse SL, Taylor AH, Pattison GT, Bannister GC, Blom AW. Prevalence and functional impact of patient-perceived leg length discrepancy after hip replacement. Int Orthop 2009;33:905–9. doi:10.1007/s00264-008-0563-6. [5] Röder C, Vogel R, Burri L, Dietrich D, Staub LP. Total hip arthroplasty: leg length inequality impairs functional outcomes and patient satisfaction. BMC Musculoskelet Disord 2012;13:95. doi:10.1186/1471-2474-13-95. [6] Maloney WJ, Keeney JA. Leg length discrepancy after total hip arthroplasty. J Arthroplasty 2004;19:108–10. [7] Lang JE, Scott RD, Lonner JH, Bono JV, Hunter DJ, Li L. Magnitude of limb lengthening after primary total knee arthroplasty. J Arthroplasty 2012;27:341–6. doi:10.1016/j.arth.2011.06.008. [8] Mufty S, van denneucker H, Bellemans J. The influence of leg length difference on clinical outcome after revision TKA. The Knee 2014;21:424–7. doi:http://dx.doi.org/10.1016/j.knee.2012.09.007. [9] Kim SH, Rhee S-M, Lim J-W, Lee H-J. The effect of leg length discrepancy on clinical outcome after TKA and identification of possible risk factors. Knee Surg Sports Traumatol Arthrosc 2016;24:2678–85. doi:10.1007/s00167015-3866-3. [10] Rösler J, Perka C. The effect of anatomical positional relationships on kinetic parameters after total hip replacement. Int Orthop 2000;24:23–7. [11] KELLGREN JH, LAWRENCE JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494–502. [12] Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;248:13–4.
AC C
EP
TE D
M AN U
SC
336
11
ACCEPTED MANUSCRIPT
SC
RI PT
[13] Collins NJ, Misra D, Felson DT, Crossley KM, Roos EM. Measures of knee function: International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form, Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Injury and Osteoarthritis Outcome Score Physical Function Short Form (KOOS-PS), Knee Outcome Survey Activities of Daily Living Scale (KOS-ADL), Lysholm Knee Scoring Scale, Oxford Knee Score (OKS), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), Activity Rating Scale (ARS), and Tegner Activity Score (TAS). Arthritis Care & Research 2011;63:S208–28. doi:10.1002/acr.20632. [14] Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS)--development of a self-administered outcome measure. J Orthop Sports Phys Ther 1998;28:88–96. [15] Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833– 40. [16] Guggenberger R, Pfirrmann CWA, Koch PP, Buck FM. Assessment of lower limb length and alignment by biplanar linear radiography: comparison with supine CT and upright full-length radiography. AJR Am J Roentgenol 2014;202:W161–7. doi:10.2214/AJR.13.10782.
M AN U
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391
AC C
EP
TE D
392
12
ACCEPTED MANUSCRIPT 393
APPENDIX A
394
Leg Length Questionnaire 1. Before your operation, did you feel you had a difference in the lengths of
396
your legs? YES / NO (circle one)
397
2. Initially after your knee replacement
402 403 404 405 406 407 408 409 410 411 412 413 414
RI PT
401
If Yes to Part a,
b. Did the leg with the knee replacement feel longer or shorter? LONGER / SHORTER (circle one)
c. Did you need to wear a shoe raise to even out your leg lengths? YES / NO (circle one) 3. Now after your knee replacement
M AN U
400
YES / NO (circle one)
a. Do your leg lengths feel equal now? YES / NO (circle one) If No to Part a,
b. Does the leg with the knee replacement feel longer or shorter? LONGER / SHORTER (circle one)
TE D
399
a. Did you feel you had a difference in the lengths of your legs?
c. Do you wear a shoe raise now? YES / NO (circle one)
d. Does the difference in leg lengths bother you? YES / NO (circle one)
EP
398
SC
395
Would you describe the results of your operation as
416
EXCELLENT / VERY GOOD / GOOD / FAIR / POOR (circle one)
AC C
415
13
ACCEPTED MANUSCRIPT ACKNOWLEDGMENTS The authors would like to acknowledge and thank Jo Maguire from Sydney Knee Specialists for her assistance in liaising with patients and collecting
AC C
EP
TE D
M AN U
SC
RI PT
questionnaires.
ACCEPTED MANUSCRIPT Table 1- Pre-operative patient demographics Characteristics
All
Varus (<0°)
Valgus (≥0°) P
n
91
66 (73%)
25 (27%)
HKA (°) mean (SD)
-3.1 (6.5)
-6.3 (3.9)
5.4 (3.6)
HKA range (°)
-17.4 to 12.2
-17.4 to -0.4
0 to 12.2
Pre-operative LLD (mm) -3.0 (8.0)
-2.4 (8.5)
-4.5 (6.3)
mean (SD) range
-23.3 to 38.4
-23.3 to 38.4 -15.1 to 6.8
Age (years)
70.2 (8.9) 50-89 70.6 (9.0) 52-89
RI PT
mean (SD) range
69.1 (7.6)
29.4 (5.0)
28.9 (4.3)
mean (SD) range
17.5–48.6
20.8–48.5
17.5–47.6
Sex (M:F)
34:57
25:41
9:16
30.8 (6.6)
SC
M AN U TE D EP
0.5
50-86
BMI (kg/m2)
AC C
0.3
0.1
0.9
ACCEPTED MANUSCRIPT Table 2- Comparison of changes in functional scores for radiographic and perceived leg length discrepancy at 6 months post operation Change in Score
p
Perceived LLD
Yes
No
Yes
No
(n=10)
(n=81)
(n=11)
(n=70)
KOOS Symptoms
18.1
17.3
Mean (SD)
(20.8)
(23.6)
KOOS Pain
40.9
37.1
Mean (SD)
(24.4)
(24.3)
KOOS ADL Mean
39.3
35.1
(SD)
(16.3)
(23.6)
KOOS Sport Mean
14.4
20.7
(SD)
(21.1)
KOOS QOL Mean
32.6
(SD)
(35.4)
WOMAC Mean (SD)
37.1
M AN U
Radiographic LLD
(16.0)
(SD) KSS Functional
(30.7) 30.8
(21.2) 29.2
(20.8)
(35.2)
29.4
23.6
(21.2)
(27.7)
AC C
EP
Mean (SD)
37.0
39.2
TE D
KSS Knee Mean
(30.7)
(28.1) 0.66 10.1
(24.4)
0.15
(22.5) 42.3
<0.
(21.7)
001
39.1
0.002
SC
0.61 14.2
19.4
RI PT
0.92 7.1
p
(24.3)
0.56 5.8
(28.5)
0.55 10.3
(30.7)
0.40 13.1
(22.6)
0.52 16.2
(26.2) 0.55 15.3 (28.9)
(21.0) 22.8
0.11
(29.4) 41.8
0.004
(29.3) 35.5
0.002
(18.8) 31.0
0.22
(34.1) 26.1 (27.1)
0.30
ACCEPTED MANUSCRIPT Figure 1- Measurement of leg lengths on long leg radiographs
AC C
EP
TE D
M AN U
SC
RI PT
Figure 2- Study exclusions
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Underwent Primary TKA 116
Incomplete study protocol radiographs=14 Pre-existing extra-articular deformity=4 - Pre-existing THA LLD >5mm=1 - Bilateral TKA = 6
SC
25
RI PT
Met Exclusion Criteria
Study Subjects
AC C
EP
TE D
M AN U
91
S0883-5403(17)30320-0
DOI:
10.1016/j.arth.2017.04.007
Reference:
YARTH 55809
To appear in:
The Journal of Arthroplasty
Received Date: 9 February 2017 Revised Date:
1 April 2017
Accepted Date: 5 April 2017
Please cite this article as: Chinnappa J, Chen DB, Harris IA, MacDessi SJ, Predictors and Functional Implications of Change in Leg Length Following Total Knee Arthroplasty, The Journal of Arthroplasty (2017), doi: 10.1016/j.arth.2017.04.007. 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
PREDICTORS AND FUNCTIONAL IMPLICATIONS OF CHANGE IN LEG LENGTH FOLLOWING TOTAL KNEE ARTHROPLASTY Dr Jason Chinnappa MBBS, MS (Ortho)1 Dr Darren B Chen MBBS (Hons) FRACS FAOrthA1,2
Dr Samuel J MacDessi MBBS (Hons) FRACS FAOrthA 1,2
RI PT
Professor Ian A Harris MBBS, FRACS(Ortho), FAHMS, MMed(Clin Epi), PhD3
1. Department of Orthopaedic Surgery, The Canterbury Hospital, Canterbury,
SC
NSW, Australia
2. Sydney Knee Specialists, St George Private Hospital, Sydney, NSW, Australia
School, UNSW, Australia
M AN U
3. Ingham Institute of Applied Medical Research, South Western Sydney Clinical
Please address all correspondence to
Dr Samuel J MacDessi
TE D
Sydney Knee Specialists
Suite 8, 19 Kensington Street
Kogarah NSW 2217 Australia
EP
Phone: +61 2 8307 0333 Fax: +61 2 8307 0334
AC C
Email: [email protected]
ACCEPTED MANUSCRIPT 1
PREDICTORS AND FUNCTIONAL IMPLICATIONS OF CHANGE IN LEG
2
LENGTH FOLLOWING TOTAL KNEE ARTHROPLASTY
3 ABSTRACT
5
Background: Leg lengthening occurs in 83% of primary TKA. The effects of leg
6
length discrepancy (LLD) on THA patients are well established. However, patient
7
function and satisfaction associated with LLD following primary TKA has not
8
been analyzed. This study aimed to quantify the magnitude of limb lengthening,
9
identify radiographic and perceived LLD, and correlate these with predictive
RI PT
4
factors and functional outcomes in a series of TKA patients.
11
Methods: Patients undergoing primary TKA who met inclusion criteria were
12
prospectively enrolled in this study. Leg length measurements were measured
13
on standardised pre and post-operative long leg radiographs. Patients completed
14
pre-operative and six month post-operative KSS and KOOS functional scores as
15
well as a post-operative satisfaction and customised leg length specific functional
16
questionnaire.
17
Results: 91 patients undergoing TKA surgeries were included. Mean overall
18
lengthening was 3.5mm (range -31.0 to 21.4mm, SD 8.4) with 77% of limbs
19
lengthened. 89% of patients had no LLD (defined as 10mm or greater) following
20
TKA. Post-operative radiographic LLD was associated with increased pre-
21
operative LLD (p<0.001). Perceived post-operative LLD was associated with
22
female gender (p=0.02), decreased satisfaction (18% vs. 84%, p<0.001) and
23
poorer functional score changes. Perceived LLD was not associated with
24
radiographic LLD.
25
Conclusions: Radiographic lengthened LLD is uncommon following primary TKA
26
(11%) and does not correlate with perceived LLD. Patients with perceived LLD
27
have decreased satisfaction and functional score improvements following TKA
28
surgery.
29
Level of Evidence: 2b
30
Keywords
31
Total Knee Replacement; Leg Length Inequality; Patient Reported Outcomes;
32
Patient Satisfaction
AC C
EP
TE D
M AN U
SC
10
1
ACCEPTED MANUSCRIPT INTRODUCTION
34
Leg length inequality is a well-described complication of hip arthroplasty (THA)
35
surgery. Patients with leg length discrepancy (LLD) after THA have been found to
36
have increased rates of gait disturbance, hip pain [1], low back pain and nerve
37
injury [2]. Furthermore, patients with leg length discrepancy have lower
38
functional hip scores [3,4], more dissatisfaction [5] and higher rates of litigation
39
than patients with equal leg lengths [6].
40
RI PT
33
Leg length discrepancy and its effects following total knee arthroplasty (TKA)
42
have not been extensively studied. A report by Lang et al found that leg
43
lengthening was common after TKA, occurring in 83% of patients [7]. It was
44
noted that the average lengthening following unilateral TKA was 6.3mm, with a
45
range from 11mm shorter to 24mm longer. However, their study did not
46
investigate the correlation of these findings with patient-reported outcomes.
47
Another recent study by Mufty et al found that 76% of patients underwent leg
48
lengthening following revision TKA [8]. Average lengthening was 5.3mm in their
49
series, resulting in a 3.8mm longer operative side (range from 24mm shorter to
50
35mm longer). They also found positive correlation between leg lengthening and
51
clinical outcomes whilst patients with perceived LLD did not have any
52
differences in clinical or functional outcomes. Kim et al have retrospectively
53
investigated both the quantitative changes in leg length as well as the functional
54
implications of this in primary TKA, however their series was specifically limited
55
to cases of varus knee osteoarthritis [9].
M AN U
TE D
EP
AC C
56
SC
41
57
The aim of our study was to investigate the magnitude of limb lengthening and
58
frequency of perceived limb length discrepancy after primary TKA and correlate
59
this with functional outcomes and predictive factors for these variables.
60 61
METHODS
62
Patients and setting
63
All patients who were undergoing elective primary TKA by two knee
64
arthroplasty surgeons (SJM, DBC) at a single institution were prospectively
65
selected for inclusion in the study. Patients with pre-existing extra-articular
2
ACCEPTED MANUSCRIPT deformity due to fracture or previous surgery were excluded from the study.
67
Patients with known leg length inequality due to other causes (radiographic leg
68
lengthening post THA >5mm and longstanding leg length inequality > 5 years
69
requiring orthotics) were excluded. Patients undergoing bilateral TKA (staged
70
within the study period or simultaneous) were excluded from the study. Finally
71
patients who did not undergo radiographs according to the protocol for the
72
study were also excluded from analysis.
73
RI PT
66
Surgical Procedure
75
All surgeries were performed using a medial parapatellar approach, with the aim
76
to restore a neutral mechanical axis followed by appropriate ligament balancing.
77
One of two cemented posterior-stabilised prostheses were used in all cases
78
(Smith and Nephew Legion, Memphis, TN, USA or Amplitude SCORE, Valence,
79
France). Bone resections were based on intramedullary jigs or computer
80
navigated alignment. All patients underwent the same post-operative
81
mobilisation and physiotherapy protocol.
82
M AN U
SC
74
Study variables and measurement
84
Patient demographic characteristics including age, sex, BMI, knee pathology and
85
operative side were recorded at time of initial pre-operative consultation. Pre
86
and post-operative flexion contractures were measured by the treating surgeon
87
using a goniometer and recorded when present.
EP
88
TE D
83
Radiographic Measurements
90
All patients underwent pre-operative and post-operative full length standing
91
radiographs at a single radiology centre. Pre-operative radiographs were
92
performed within six months of surgery, and post-operative radiographs were
93
performed two days post operation.
AC C
89
94 95
Leg length differences were measured from the top of the femoral head to the
96
center of the tibial plafond (Figure 1). The difference in leg lengths between the
97
operated and non-operated limb was recorded in millimeters (mm) with
3
ACCEPTED MANUSCRIPT 98
shortening of the operated side being recorded as a negative value and
99
lengthening as a positive value.
100 Pre-operative and post-operative hip-knee angle (HKA) was measured as the
102
angle between a line from the center of the femoral head to the center of the
103
distal femur and a line between the center of the proximal tibia and the center of
104
the tibial plafond. Varus angles were given negative values and valgus angles
105
were given positive values. All measurements were performed using the center’s
106
image processor (Medway, Healthcare Imaging Services, NSW, Australia) by one
107
of the authors (JC). The non-operative side leg length was used to calculate
108
magnification changes between pre and post-operative radiographs and
109
accordingly adjust post-operative leg length measurements. 10 patient pre-
110
operative leg length radiographic measurements were repeated to check the
111
intra-observer reliability of these values. The Intraclass Correlation Coefficient
112
(ICC) was 1.00 (p <0.001), indicating excellent reliability of these measurements.
113
M AN U
SC
RI PT
101
LLD and Perceived LLD
115
LLD for the purpose of comparison was defined as a leg length difference of
116
greater than or equal to 10mm and sub-analyzed as lengthened and non-
117
lengthened groups. These values were chosen as 10mm has been shown to have
118
biokinetic implications on gait [10] and been used in a prior study looking at hip
119
arthroplasty and leg length discrepancy [5].
EP
TE D
114
120
The frequency of perceived lengthened LLD at six months post-surgery was
122
obtained from the custom questionnaire (Appendix A- Question 3). Perceived
123
and actual LLD’s were correlated.
124 125
AC C
121
Predictive factors investigated for correlation with actual or perceived LLD were:
126
-
Patient baseline characteristics: age, sex, body mass index (BMI)
127
-
Pre-operative coronal deformity
128
-
Degree of coronal deformity correction
129
-
Pre-operative LLD
130
-
Pre-operative flexion contractures in the operative knee
4
ACCEPTED MANUSCRIPT 131
-
Post-operative flexion contractures in the operative knee
132
-
Polyethylene insert thickness
133
-
Other lower limb degenerative disease or arthroplasty: osteoarthritis of either hip joint resulting in asymmetric joint space loss, or prior THA on
135
either side were recorded. Kellgren and Lawrence Grades 3 or 4
136
osteoarthritis or TKA in the contralateral knee was also recorded. The
137
Kellgren and Lawrence classification is used to categorise radiographic
138
knee osteoarthritis based on reduction of joint space, sclerosis and
139
presence of osteophytes, with grade 4 being the most severe. [11]
140
SC
RI PT
134
Patient-reported outcome measures
142
A custom-designed leg length questionnaire was created and used to assess the
143
functional implications of leg lengthening six-month post-operation (Appendix
144
A). This questionnaire was used to derive frequencies and identify patients with
145
perceived LLD, as well as assess patient satisfaction with their surgery on a 5-
146
point scale from Poor to Excellent. All patients were also completed a Knee
147
Society Score (KSS) and the Knee Injury and Osteoarthritis Outcome Score
148
(KOOS) with its included subset Western Ontario and McMaster Universities
149
Arthritis Index (WOMAC) score pre-operatively and six months post-operatively.
150
The KSS is a widely used instrument that combines functional knee scores with
151
anatomical knee alignment [12]. Scores of 80-100 are considered excellent,
152
whilst scores below 60 are considered poor. The KOOS is a validated knee score
153
in the surgical setting, which assesses patients’ pain, symptoms, function and
154
quality of life in the preceding week [13]. It utilizes 5-point Likert scales
155
resulting in extrapolated percentage scores of 0 (extreme symptoms) to 100 (no
156
symptoms) [14]. WOMAC is an arthritis-specific, validated score that is
157
commonly used in assessing symptoms from arthritic hips and knees [15].
158
Changes in KOOS, KSS and WOMAC scores were used to compare groups.
AC C
EP
TE D
M AN U
141
159 160
Statistical analysis
161
All statistical analysis was performed using SPSS Statistics Software, version 20
162
(IBM Software). Differences in variables with continuous outcomes were
163
analyzed using the independent t-test. Categorical variables were tested using
5
ACCEPTED MANUSCRIPT 164
Chi-square analysis and Fisher’s exact test. Predictive factors for LLD and
165
perceived LLD were also analyzed with multivariate analysis using binomial
166
logistic regression. Statistical significance was set at p=0.05.
167 Ethics approval
169
Informed consent was obtained from all patients. Approval to conduct the study
170
was granted by the Hunter New England Research Ethics Committee (Reference
171
15/02/18/5.07).
RI PT
168
172 RESULTS
174
Population
175
116 patients undergoing TKA surgeries were enrolled in this study. 25 patients
176
were excluded (Figure 2), leaving 91 patients for analysis.
177
M AN U
SC
173
Baseline characteristics
179
Baseline demographics for the 91 included patients undergoing TKA (44 left, 47
180
right) surgeries are provided in Table 1. Pre-operative diagnoses included 83
181
osteoarthritis cases, 5 post-traumatic arthritis cases, 2 avascular necrosis cases
182
and 1 rheumatoid arthritis case. 89% of patients completed the leg length and
183
surgery satisfaction questionnaire.
TE D
178
184
186
2% of patients pre-operatively had a leg length discrepancy of 10mm or greater.
EP
185
Degree of correction
188
Mean postoperative HKA was 0.8 degrees (range -5.6 to 7.5 degrees, SD 2.5),
189
with an overall correction of 5.8 degrees (range 0.1 – 17.3, SD 3.6).
190
AC C
187
191
Magnitude of Lengthening
192
Mean overall lengthening in this study was 3.5mm (range -31.0 to 21.4mm, SD
193
8.4) with a mean post-operative LLD of 0.4mm longer (range -42.3 to 29.9mm,
194
SD 10.0). 77% of knees were lengthened post-operatively by a mean of 6.7mm
195
(range 0.0 to 21.4mm, SD 4.9). The 23% of knees that were shortened post-
6
ACCEPTED MANUSCRIPT 196
operatively had a mean leg length change of -7.4mm (range -31 to -0.6mm, SD
197
8.5).
198 Radiographic LLD
200
Post-operatively, 89% of patients had a LLD of less than 10mm (mean –1.4mm,
201
range – 42.3 to 9.7, SD 8.7) and were considered to have normal leg lengths for
202
the study groups. 11% of patients were considered to have a LLD with the
203
operative leg more than or equal to 10mm longer (mean 15.3, range 10.3 – 29.9,
204
SD 5.9).
RI PT
199
SC
205 Perceived LLD
207
Leg length questionnaire analysis revealed 16% of patients felt they had a LLD
208
pre-operatively. Immediately post-operatively, 21% of patients felt they had a
209
LLD with 88% of these patients perceiving their operative leg to be longer. 12%
210
of these patients wore a contralateral shoe raise for perceived lengthening of the
211
operative side. Six months post operatively, 14% of patients perceived a LLD
212
with 73% of these patients feeling they were lengthened on the operative side.
213
9% required a shoe raise and 45% of these patients (6% of total cohort)
214
reported being troubled by their perceived LLD. None of the patients with a
215
perceived LLD at six months post operation had a radiographic LLD of 10mm or
216
longer.
TE D
EP
217
M AN U
206
Predictive factors for Radiographic LLD
219
Pre-operative LLD was associated with increased post-operative leg length
220
discrepancy (p<0.001), with mean pre-operative LLD in the LLD group 7.9mm
221
versus –4.4mm in the normal group. No relationship was identified between
222
radiographic post-operative LLD and sex (p=1), age (p=0.53), BMI (p=0.27),
223
polyethylene insert thickness (p= 0.11), pre-operative FFD (p=0.74), post-
224
operative FFD (p=0.59), contralateral knee degenerative disease (p= 0.51) or
225
arthroplasty (p= 0. 2) or either hip arthroplasty (p= 1), severity of pre-operative
226
angular deformity (p=0.67), or degree of angular correction (p= 0.62).
227
There were no significant correlations between any of the above factors and
228
post-operative LLD on multivariate logistic regression analysis.
AC C
218
7
ACCEPTED MANUSCRIPT 229 Predictive factors for Perceived LLD
231
Female sex was significantly associated with perceived LLD in this series as 20%
232
of females had a perceived LLD at 6 months post operation versus 3% of males
233
(p=0.04). Perceived LLD was not associated with age (p=0.98), BMI (p=0.21),
234
polyethylene insert thickness (p= 0.59), pre-operative FFD (p=0.52), post-
235
operative FFD (p=0.3), contralateral knee degenerative disease (p= 0.53) or
236
arthroplasty (p = 1) or either hip arthroplasty (p= 0.26), severity of pre-
237
operative angular deformity (p=0.39), degree of angular correction (p= 0.83) or
238
pre-operative LLD (p=.0.42).
SC
RI PT
230
239
Multivariate regression analysis also found an association between female sex
241
and perceived LLD (OR 21.76, 95% CI 1.65 – 287.31, p=0.02). No patients with
242
perceived LLD had a radiographic LLD.
M AN U
240
243 Functional outcomes
245
There were no significant associations between changes in functional scores
246
(KOOS, KSS and WOMAC) and radiographic LLD (Table 2). There were
247
differences between outcomes for the KOOS Pain, KOOS Activities of Daily Living
248
(ADL), KOOS Quality of Life (QOL) and WOMAC functional scores, with
249
significantly lower improvements in functional scores for patients with
250
perceived LLD (Table 2).
EP
TE D
244
251
Patient satisfaction
253
Overall, 75% of patients were satisfied with their operation (rated very good or
254
excellent). There was a non-statistically significant association between patient
255
satisfaction and radiographic LLD with 77% of normal radiograph patients rating
256
their operation as very good or excellent compared with 57% of patients with
257
radiographic LLD (p=0.4). With regards to perception of LLD, 18% of patients
258
with perceived LLD rated their operation as very good or excellent, significantly
259
lower than the 84% of patients without perceived LLD (p<0.001).
AC C
252
260 261
DISCUSSION
8
ACCEPTED MANUSCRIPT 262
Our finding of 77% of patients undergoing overall lengthening following TKA is
263
similar to the finding of 83% by Lang et al[7]. However, 89% of our patients had
264
a discrepancy of less than 10mm longer post-operatively. These findings show
265
that the majority of patients undergoing TKA do not have radiographic evidence
266
of significant functional LLD.
RI PT
267 The finding on univariate analysis of correlation between pre-operative LLD and
269
increased post-operative LLD is not unexpected. We are unable to identify why
270
females were more likely to perceive LLD in this series. The finding of female sex
271
being associated with perceived LLD, as well as none of the patients with
272
perceived LLD having radiographic evidence of the same, may be useful in
273
counseling patients pre-operatively.
M AN U
274
SC
268
Patients with radiographic LLD following TKA showed lower satisfaction with
276
their operations, although this finding did not reach statistical significance in this
277
series. Furthermore, this was not associated with lower functional scores in this
278
group. Kim et al also found a non-significant result with less highly satisfied
279
patients who had radiographic LLD in their series [9]. Their results also showed
280
some significant functional differences with regards to ability on stairs and KSS
281
functional score, in favor of patients with radiographic LLD <15mm. As their
282
series only looked at pre-operative varus knees with a 15mm LLD cut-off, our
283
series investigating all pre-operative alignments and a 10mm cut-off for LLD,
284
may not be directly comparable.
EP
AC C
285
TE D
275
286
Patients with perceived LLD had significantly lower satisfaction scores in our
287
series. Additionally, our study found a significant correlation between perceived
288
LLD and lower KOOS pain, KOOS ADL, KOOS QOL and WOMAC functional knee
289
scores. The other KOOS subscales and KSS scores also showed decreased
290
functional outcomes in patients with perceived LLD, however these did not reach
291
statistical significance. Our findings are consistent with those of Konyves[3] and
292
Wylde[4] that perceived LLD is associated with poorer functional outcomes in
293
THA patients. This is in contrast to the finding by Mufty that patients with actual
294
and perceived lengthening following revision TKA had better pain functional
9
ACCEPTED MANUSCRIPT 295
scores and no differences in other scores.[8] This discrepancy may be due to
296
differing expectations in patients undergoing primary and revision TKA. It is
297
unclear whether patients who are unhappy with or feel limited following their
298
primary TKA may try and attribute this to other causes such as perceived LLD.
299 This study does have some limitations that need to be considered when
301
interpreting the results. There is potential underestimation of radiographic
302
lengthening in this series, as long leg radiographs post-operatively were done on
303
day 2 following surgery. Some patients may have been unable to achieve full
304
extension by this stage and this would understate their leg length on standing
305
long leg films. However as we did not identify a correlation between LLD and
306
patients with fixed flexion deformity, this is unlikely to be clinically significant.
307
Another limitation of our study is the small number of patients with extreme
308
degrees of coronal deformity, which may limit our ability to correlate this and
309
the degree of correction with LLD. Finally, this study utilized long leg
310
radiographs to assess radiographic LLD. Long leg radiographs have been shown
311
to be comparable to Computed Tomography (CT) imaging for calculation of
312
lower limb lengths.[16] CT imaging would have allowed calculation of actual
313
limb length discrepancy independent of coronal or sagittal deformity. However
314
the added radiation exposure and the fact that coronal and sagittal deformity
315
correction that is routine with most primary TKA is most likely to affect an
316
apparent LLD make CT imaging less beneficial in this setting.
SC
M AN U
TE D
EP
317
RI PT
300
Strengths of this study include the completion rate of the functional outcome,
319
LLD and satisfaction questionnaires and our ability to correlate this with
320
radiographic and perceived LLD. These findings will be useful in being able to
321
allow surgeons to inform patients of the impact of limb lengthening in primary
322
TKA. Patients can be counseled that although they may perceive LLD, it is
323
uncommon following primary TKA and usually not associated with radiographic
324
LLD. Patients can also be informed that even those with radiographic LLD have
325
similar functional scores to those without radiographic LLD and there are no
326
significant predictive factors for radiographic LLD. Further study into patients
AC C
318
10
ACCEPTED MANUSCRIPT 327
who are not satisfied with their TKA or have poorer functional scores or
328
perceived LLD is needed to identify the interplay between these findings.
329 CONCLUSIONS
331
The majority of patients undergoing primary TKA do not have significant pre or
332
post-operative LLD. Perceived LLD following TKA correlates with female sex,
333
poorer functional outcomes and decreased satisfaction but not with radiographic
334
LLD.
RI PT
330
335 REFERENCES
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
[1] Plaass C, Clauss M, Ochsner PE, Ilchmann T. Influence of leg length discrepancy on clinical results after total hip arthroplasty--a prospective clinical trial. Hip Int 2011;21:441–9. doi:10.5301/HIP.2011.8575. [2] Della Valle CJ, Di Cesare PE. Complications of total hip arthroplasty: neurovascular injury, leg-length discrepancy, and instability. Bull Hosp Jt Dis 2001;60:134–42. [3] Konyves A, Bannister GC. The importance of leg length discrepancy after total hip arthroplasty. J Bone Joint Surg Br 2005;87:155–7. [4] Wylde V, Whitehouse SL, Taylor AH, Pattison GT, Bannister GC, Blom AW. Prevalence and functional impact of patient-perceived leg length discrepancy after hip replacement. Int Orthop 2009;33:905–9. doi:10.1007/s00264-008-0563-6. [5] Röder C, Vogel R, Burri L, Dietrich D, Staub LP. Total hip arthroplasty: leg length inequality impairs functional outcomes and patient satisfaction. BMC Musculoskelet Disord 2012;13:95. doi:10.1186/1471-2474-13-95. [6] Maloney WJ, Keeney JA. Leg length discrepancy after total hip arthroplasty. J Arthroplasty 2004;19:108–10. [7] Lang JE, Scott RD, Lonner JH, Bono JV, Hunter DJ, Li L. Magnitude of limb lengthening after primary total knee arthroplasty. J Arthroplasty 2012;27:341–6. doi:10.1016/j.arth.2011.06.008. [8] Mufty S, van denneucker H, Bellemans J. The influence of leg length difference on clinical outcome after revision TKA. The Knee 2014;21:424–7. doi:http://dx.doi.org/10.1016/j.knee.2012.09.007. [9] Kim SH, Rhee S-M, Lim J-W, Lee H-J. The effect of leg length discrepancy on clinical outcome after TKA and identification of possible risk factors. Knee Surg Sports Traumatol Arthrosc 2016;24:2678–85. doi:10.1007/s00167015-3866-3. [10] Rösler J, Perka C. The effect of anatomical positional relationships on kinetic parameters after total hip replacement. Int Orthop 2000;24:23–7. [11] KELLGREN JH, LAWRENCE JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494–502. [12] Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;248:13–4.
AC C
EP
TE D
M AN U
SC
336
11
ACCEPTED MANUSCRIPT
SC
RI PT
[13] Collins NJ, Misra D, Felson DT, Crossley KM, Roos EM. Measures of knee function: International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form, Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Injury and Osteoarthritis Outcome Score Physical Function Short Form (KOOS-PS), Knee Outcome Survey Activities of Daily Living Scale (KOS-ADL), Lysholm Knee Scoring Scale, Oxford Knee Score (OKS), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), Activity Rating Scale (ARS), and Tegner Activity Score (TAS). Arthritis Care & Research 2011;63:S208–28. doi:10.1002/acr.20632. [14] Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS)--development of a self-administered outcome measure. J Orthop Sports Phys Ther 1998;28:88–96. [15] Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833– 40. [16] Guggenberger R, Pfirrmann CWA, Koch PP, Buck FM. Assessment of lower limb length and alignment by biplanar linear radiography: comparison with supine CT and upright full-length radiography. AJR Am J Roentgenol 2014;202:W161–7. doi:10.2214/AJR.13.10782.
M AN U
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391
AC C
EP
TE D
392
12
ACCEPTED MANUSCRIPT 393
APPENDIX A
394
Leg Length Questionnaire 1. Before your operation, did you feel you had a difference in the lengths of
396
your legs? YES / NO (circle one)
397
2. Initially after your knee replacement
402 403 404 405 406 407 408 409 410 411 412 413 414
RI PT
401
If Yes to Part a,
b. Did the leg with the knee replacement feel longer or shorter? LONGER / SHORTER (circle one)
c. Did you need to wear a shoe raise to even out your leg lengths? YES / NO (circle one) 3. Now after your knee replacement
M AN U
400
YES / NO (circle one)
a. Do your leg lengths feel equal now? YES / NO (circle one) If No to Part a,
b. Does the leg with the knee replacement feel longer or shorter? LONGER / SHORTER (circle one)
TE D
399
a. Did you feel you had a difference in the lengths of your legs?
c. Do you wear a shoe raise now? YES / NO (circle one)
d. Does the difference in leg lengths bother you? YES / NO (circle one)
EP
398
SC
395
Would you describe the results of your operation as
416
EXCELLENT / VERY GOOD / GOOD / FAIR / POOR (circle one)
AC C
415
13
ACCEPTED MANUSCRIPT ACKNOWLEDGMENTS The authors would like to acknowledge and thank Jo Maguire from Sydney Knee Specialists for her assistance in liaising with patients and collecting
AC C
EP
TE D
M AN U
SC
RI PT
questionnaires.
ACCEPTED MANUSCRIPT Table 1- Pre-operative patient demographics Characteristics
All
Varus (<0°)
Valgus (≥0°) P
n
91
66 (73%)
25 (27%)
HKA (°) mean (SD)
-3.1 (6.5)
-6.3 (3.9)
5.4 (3.6)
HKA range (°)
-17.4 to 12.2
-17.4 to -0.4
0 to 12.2
Pre-operative LLD (mm) -3.0 (8.0)
-2.4 (8.5)
-4.5 (6.3)
mean (SD) range
-23.3 to 38.4
-23.3 to 38.4 -15.1 to 6.8
Age (years)
70.2 (8.9) 50-89 70.6 (9.0) 52-89
RI PT
mean (SD) range
69.1 (7.6)
29.4 (5.0)
28.9 (4.3)
mean (SD) range
17.5–48.6
20.8–48.5
17.5–47.6
Sex (M:F)
34:57
25:41
9:16
30.8 (6.6)
SC
M AN U TE D EP
0.5
50-86
BMI (kg/m2)
AC C
0.3
0.1
0.9
ACCEPTED MANUSCRIPT Table 2- Comparison of changes in functional scores for radiographic and perceived leg length discrepancy at 6 months post operation Change in Score
p
Perceived LLD
Yes
No
Yes
No
(n=10)
(n=81)
(n=11)
(n=70)
KOOS Symptoms
18.1
17.3
Mean (SD)
(20.8)
(23.6)
KOOS Pain
40.9
37.1
Mean (SD)
(24.4)
(24.3)
KOOS ADL Mean
39.3
35.1
(SD)
(16.3)
(23.6)
KOOS Sport Mean
14.4
20.7
(SD)
(21.1)
KOOS QOL Mean
32.6
(SD)
(35.4)
WOMAC Mean (SD)
37.1
M AN U
Radiographic LLD
(16.0)
(SD) KSS Functional
(30.7) 30.8
(21.2) 29.2
(20.8)
(35.2)
29.4
23.6
(21.2)
(27.7)
AC C
EP
Mean (SD)
37.0
39.2
TE D
KSS Knee Mean
(30.7)
(28.1) 0.66 10.1
(24.4)
0.15
(22.5) 42.3
<0.
(21.7)
001
39.1
0.002
SC
0.61 14.2
19.4
RI PT
0.92 7.1
p
(24.3)
0.56 5.8
(28.5)
0.55 10.3
(30.7)
0.40 13.1
(22.6)
0.52 16.2
(26.2) 0.55 15.3 (28.9)
(21.0) 22.8
0.11
(29.4) 41.8
0.004
(29.3) 35.5
0.002
(18.8) 31.0
0.22
(34.1) 26.1 (27.1)
0.30
ACCEPTED MANUSCRIPT Figure 1- Measurement of leg lengths on long leg radiographs
AC C
EP
TE D
M AN U
SC
RI PT
Figure 2- Study exclusions
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Underwent Primary TKA 116
Incomplete study protocol radiographs=14 Pre-existing extra-articular deformity=4 - Pre-existing THA LLD >5mm=1 - Bilateral TKA = 6
SC
25
RI PT
Met Exclusion Criteria
Study Subjects
AC C
EP
TE D
M AN U
91