- Email: [email protected]
Does Patients' Perception of Leg Length After Total Hip Arthroplasty Correlate With Anatomical Leg Length?
The Journal of Arthroplasty xxx (2017) 1e5
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Does Patients' Perception of Leg Length After Total Hip Arthroplasty Correlate With Anatomical Leg Length? Jean Y. Lazennec, MD, PhD a, b, Dominique Folinais, MD c, Caroline Florequin, MS b, Aidin E. Pour, MD d, * ^pitaux de Paris, Paris, France Department of Orthopaedic and Trauma Surgery, La Piti e-Salp^ etri ere Hospital, Assistance Publique des Ho Anatomy Department, Universit e Pierre et Marie Curie (UPMC), Paris, France c RIM Maussins-Nollet, Paris, France d Department of Orthopaedic Surgery, University of Michigan, Ann Arbor, MI a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 July 2017 Received in revised form 4 December 2017 Accepted 5 December 2017 Available online xxx
Background: This study assessed the correlation among the patients' perception of the leg length discrepancy (LLD) after total hip arthroplasty (THA) and the anatomical and functional leg length, pelvic and knee alignments, and foot height. Methods: Patients without significant spinal pathology or previous spine or lower extremity surgery who underwent primary THA (101 patients) were evaluated using EOS images obtained in standing position. All 3-dimensional measures were evaluated and compared for the repeatability and reproducibility and correlation with patients' perception of leg length. Results: In our study, the anatomical femoral length (odds ratio [OR] 0.9, P ¼ .732) did not correlate with patients' perception of the LLD, but other variables like the distance from the middle of the tibial plafond to ground (OR 14.3, P ¼ .003), sagittal knee alignment (OR 1.07, P ¼ .021), and pelvic obliquity (OR 1.05, P ¼ .021) were correlated with the patients’ perception of LLD. Conclusion: The LLD is a multifactorial complication. We found that the anatomical femoral length as the factor that can be modified with THA technique or choice of prosthesis is not the only important factor. We recommend comprehensive physical examination to investigate spinal deformities, pelvic obliquity, abductor muscle weakness, sagittal and coronal knee alignment, and foot deformity in patients who complain of LLD after THA. © 2017 Elsevier Inc. All rights reserved.
Level of Evidence: Level IV diagnostic case series Keywords: leg length three-dimensional radiography knee alignments total hip arthroplasty
Leg length discrepancy (LLD) is a common postoperative complication after total hip arthroplasty (THA) [1e3]. It can result in patient dissatisfaction, limping, need for shoe lift, low back pain, hip instability, and revision surgery [4e10]. According to a survey study of the members of the American Association of Hip and Knee Surgeons by Upadhyay et al [11], LLD was the second most common
No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2017.12.004. -Salpe ^trie re Hospital Ethical review: Study approval was obtained from La Pitie review board. -Salpe ^trie re Hospital. Location of Study: La Pitie * Reprint requests: Aidin E. Pour, MD, Department of Orthopaedic Surgery, University of Michigan, 2912 Taubman Center, SPC 5328, 1500 E. Medical Center Dr., Ann Arbor, MI 48109. https://doi.org/10.1016/j.arth.2017.12.004 0883-5403/© 2017 Elsevier Inc. All rights reserved.
reason for litigation, and 8% of surgeons had been a defendant in a legal case secondary to this complication. Among the factors important in the LLD, femoral length and offset are the 2 main factors that a surgeon can modify with surgery. The femoral length can be affected by the choice of the stem design and fixation technique and the length of the prosthetic femoral head. The increased offset due to the use of high offset stems or prosthetic head with longer neck might also result in patients' perception of tightness in the abductor muscles, pelvic obliquity, and as a result a perception of LLD [12]. Regular anteroposterior radiographs of the pelvis are 2 dimensional only and they cannot accurately measure the leg length. Despite its better accuracy, computed tomography measures the anatomical length in the supine position instead of the standing position. The standing position is the position in which LLD is perceived by patients. Unlike the anatomical length, the functional length integrates the lengths of the femur and tibia, hip flexion contracture, and the coronal and sagittal knee alignments. Pelvic obliquity and axial
2
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
automatically measured all the variables used in this study (Table 1). The femoral offset is the distance between the center of the femoral head and the greater trochanter. Axial pelvic rotation is the angle between the line passing through the centers of both hips and the X-ray beam, with negative rotation being rotation toward the operative hip and positive rotation being rotation toward the nonoperative hip. If the pelvis rotates toward one side, one hip will be in more flexion (the side that the pelvis turns toward) which can give a perception of shorter leg. The opposite hip will be in more extension (the side that the pelvis rotates away from) which can give a perception of longer leg [14]. The pelvic obliquity was measured as the distance between the horizontal line and acetabular roof that was lower in a coronally tilted pelvis. Other variables included the hip-knee-ankle angle and the femoral and tibial mechanical axes (FMA, TMA). The FMA is the line connecting the center of the femoral head to the center of the femoral notch, and the TMA is the line from the center of the tibial plateau (interspinous intercruciate midpoint) extending distally to the center of the tibial plafond. Hip-knee-ankle angle is the angle between the FMA and TMA with neutral angle being zero. Tibial and femoral rotation and knee flexion/hyperextension angle were also derived from the 3D reconstruction [15]. We used the following definitions for measuring the length (Fig. 1):
pelvic rotation which affect the patients' perception of leg length will also be taken into account in this position. In this study, these questions were asked: (1) Does the perception of LLD correlate better with the anatomical or functional length? and (2) Does the perception of LLD correlate with other variables including the femoral offset, pelvic obliquity, sacral slope, axial pelvic rotation, and coronal or sagittal knee alignment and foot and ankle height? Our hypotheses were that the functional length would correlate with the perception of LLD better than the anatomical length. We also hypothesized that the perception of LLD correlates with coronal and sagittal knee alignment, axial pelvic rotation, and obliquity. Methods After obtaining Institutional Review Board approval, 101 consecutive patients (aged 18-80 years) who presented to our clinic between January 2014 and December 2016 were reviewed in this retrospective study. Patients had primary THA for degenerative hip arthritis. All the patients with symptomatic spinal pathology, previous spinal surgery, lower extremity open reduction and internal fixation, osteotomy, or arthroplasty (knee, ankle) were excluded. Thirty-eight hips were performed through an anterolateral approach, 37 through a posterior approach, and 26 through a direct anterior approach. All procedures utilized cementless implants. The bearing surface was ceramic-on-ceramic in 56 patients and metal or ceramic-on-polyethylene in 45 patients. All patients were evaluated using standing 3-dimensional (3D) imaging system for spine-related pain and lower extremityerelated pain (EOS; EOS Imaging SA, Paris, France). They were asked about their perception of the leg lengths. Each patient stood comfortably in the EOS machine and the position was specifically checked to avoid superimposition of anatomical structures on the lateral view because this would make 3D reconstruction impossible. The simultaneous biplanar acquisition was used to perform stereoradiographic 3D modeling of each lower extremity using specialized software (sterEOS 3D; EOS Imaging SA) according to a previously described method [13]. The software used the bony landmarks to determine the femoral and tibial torsions in 3D images. After 3D modeling, the software
Anatomical femoral length: distance between the center of the femoral head (A) and the center of the trochlea (B). Anatomical tibial length: distance between the center of the tibial spine (intercondylar eminence) (C) and the center of the ankle joint (D). Functional length: distance between the center of the femoral head to the center of the ankle joint (AD). Anatomical length: sum of the anatomical femoral and tibial lengths (AB þ CD). Foot deformities cannot be assessed in the EOS images thoroughly. For this study, the distance between the middle of the tibial plafond and the ground was considered the height of the foot as a possible factor affecting the LLD. The patients were divided into 2 groups for simple comparison of the means for all the measured variables. The first group (case group) included the patients with a
Table 1 Measurements of Study Variables in Operative and Nonoperative Sides in the Cohort. Variable
No perception of LLD Femoral offset (cm) Femoral neck-shaft angle ( ) Anatomical femoral length (cm) Anatomical tibial length (cm) Anatomical leg length (cm) Functional leg length (cm) Knee varus/valgus angle ( ) Hip-knee-ankle angle ( ) Knee flexum/recurvatum ( ) With perception of LLD Femoral offset (cm) Femoral neck-shaft angle ( ) Anatomical femoral length (cm) Anatomical tibial length (cm) Anatomical leg length (cm) Functional leg length (cm) Knee varus/valgus angle ( ) Hip-knee-ankle angle ( ) Knee flexum/recurvatum ( )
Operative
Nonoperative
P Value
Mean
Minimum
Maximum
Mean
Minimum
Maximum
4.4 132.4 41.9 35.1 77 77.6 0.1 4.5 6.6
3.3 126.3 37.7 31.8 69.7 69.9 6.7 0.6 7.88
5.6 139.3 47.3 40 86.1 87.2 12.1 7.9 30.2
4.3 124.2 41.5 35.4 77 77.4 1.6 5 5.9
3.4 111.9 37.4 31.8 69.8 70.2 10 2.3 6.4
5.1 140.6 46.3 40.3 85.8 86.4 4.52 7.8 27.3
.667 <.001 .005 <.001 .976 .008 .008 .09 .554
4 133.8 41.3 34.8 76.1 76.7 0.3 4.6 6.3
2.4 121.2 34.5 29.8 65.6 66.3 7.7 0.8 11.2
5.6 149.5 47.3 40.6 87.9 88.5 15.3 10.2 26.9
4.1 125.9 41 35.1 76.1 76.5 0.5 4.9 1.8
3.2 109.9 33.9 23 65.7 65.5 12.8 1.7 11.4
5.7 135.9 47.6 42.1 89.6 90.3 8.5 11.2 28.9
.06 <.001 .011 <.001 .687 .036 .825 .269 <.001
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
3
perception of LLD. The second group (control group) included the patients without perception of LLD. Statistical Analysis The statistical analysis was performed with MedCalc (version 11.2.0.0, Mariakerke, Belgium) and Stata 14.1 (StataCorp LP, College Station, TX). A post hoc power analysis was conducted using G*Power v. 3.1.9 to determine the power of the study considering the current sample size. Using an alpha of 0.05 and a medium effect size (d ¼ 0.5), a total sample size of 100 provided a study power of 0.81. Repeatability and reproducibility were assessed by calculating the interobserver and intraobserver intraclass correlation coefficient (ICC) and its 95% confidence interval. The ICC is defined as the ratio between explained variances (variance attributable to the cause of variation: observer factor; repetition of measurement) and overall variance (explained variance þ error variance). Comparison of repeatability and reproducibility of each parameter was performed using the Fisher-Snedecor test for comparison of variances. Quantitative variables were described using the mean (M), the mean difference (d), the standard deviation (SD), and the ICC with its confidence interval. We considered an ICC of >0.90 as high, between 0.80 and 0.90 as moderate, and <0.80 as insufficient. Normal distribution of the values was checked by means of the Shapiro-Wilk normality test for each series of measurements. For data with normal distribution, paired Student's t-test was used for analysis. For data without normal distribution, related samples Wilcoxon signed rank test was used for the analysis. The significance level was set at 5%. Multiple logistic regression was used to investigate the correlation between the patients' perception of LLD and variables found significant in the univariate analysis. The significance level was set at 0.05. Results The detailed data are presented in Tables 1 and 2. In univariate and multiple logistic regression (Table 3), no correlation was found among the perception of LLD and difference in anatomical femoral length (P ¼ .393), anatomical leg length (P ¼ .458), and functional leg length (P ¼ .472). Among those whose operative legs were shorter, 65.7% (23 of 35) had a perception of LLD as compared to 56% of patients in whom the operative side was longer (37 of 66) (P ¼ .347). Tables 1 and 2 present the data regarding other variables that could correlate with the patients' perception of limb length. In univariate and multiple logistic regression (Table 3), association was found among the patients’ perception of LLD with pelvic obliquity (odds ratio [OR] 1.06, P ¼ .032), difference in knee flexum/ recurvatum (OR 1.07, P ¼ .034), and the difference in distance between the middle of tibial plafond and ground (OR 7.3, P ¼ .031). Discussion
Fig. 1. Anatomic vs functional length. (1) Anatomical femoral length: distance between the center of the femoral head (A) and the center of the trochlea (B). (2) Anatomical tibial length: distance between the center of the tibial spine (intercondylar eminence) (C) and the center of the ankle joint (D). (3) Functional length: distance between the center of the femoral head to the center of the ankle joint (AD). (4) Anatomical length: sum of the anatomical femoral and tibial lengths (AB þ CD).
This retrospective study investigated the correlation between the patients’ perception of LLD and different variables measured in EOS Imaging after primary THA. In the final multiple regression analysis, the perception of LLD was associated with the pelvic obliquity, difference in knee flexum/recurvatum angle, and difference in distance from the middle of the tibial plafond and ground. No correlation was found among the perception of LLD and anatomical femoral length, anatomical leg length, and functional leg length.
4
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
Table 2 Difference Between the Operative and Nonoperative Side Variables. No Perception of LLD
Difference in anatomical femoral length (cm) Difference in anatomical tibial length (cm) Difference in functional length (cm) Difference in anatomical leg length (cm) Difference in femoral offset (cm) Difference in neck-shaft angle ( ) Difference in knee varus/valgus angle ( ) Difference in knee flexum/recurvatum angle ( ) Difference in hip-knee-ankle angle ( ) Difference in femoral torsion ( ) Difference in the middle of tibial plafond to ground distance (cm) Pelvic incidence ( ) Sacral slope ( ) Pelvic version ( ) Axial pelvic rotation ( ) Pelvic obliquity (mm)
Perception of LLD
P Value
Mean
Minimum
Maximum
Mean
Minimum
Maximum
0.4 0.4 0.3 0.1 0.8 8.5 1.5 0.7 0.4 2.1 0.1 58.1 39.4 18.7 1.2 0.9
1.7 1.2 1.7 2.2 19.3 1.5 5.6 20.9 4.9 62.7 0.6 33.5 14.4 3.8 6.6 20.6
1.4 0.7 1.4 1.4 12 20.1 11.9 21.9 3.3 31.7 0.6 86.8 59.7 37.6 4.9 11.1
0.4 0.3 0.3 0.1 0.7 8.3 0.1 4.5 0.25 0.93 0.1 61.3 43 18.3 1.7 5.3
2 1.6 2.8 3.2 14.5 6.8 8.4 21 4.2 40.8 0.9 33.5 22.1 3.5 10 20.3
3 0.6 2.6 2.5 15 24.5 12.4 27.9 2.9 35.2 0.8 106.7 73.9 43.6 11.3 26.6
This study has limitations. Preoperative EOS Imaging was not available for these patients. As a result, we cannot compare the preoperative and postoperative EOS images in terms of changes in the leg length caused by THA. The sample size is rather small but the post hoc power calculation showed that current sample size provided enough power (0.81) to detect the difference between the 2 groups. We also did not investigate the degree of any foot deformities in patients radiographically as flat foot deformity may potentially affect the patients' perception of the leg length. Instead, the distance between the middle of the tibial plafond and the ground was used as the foot arch height. LLD is a one of the complications of THA which sometimes results in early revision arthroplasty or medicolegal litigation [16e18]. In a study, 30% of the patients who underwent THA perceived different leg length [19]. Only 36% of these patients had a true LLD in imaging studies. Imaging studies performed in supine position cannot measure the true functional leg length accurately. The prevalence of anatomical LLD is diversely described in the literature [20e22]. In assessment of the normal population without any perception of the LLD, different authors reported that there was an anatomical LLD of 5 mm or more in 90% of the population [20e22]. Change in the anatomical femoral length can be controlled by the operating surgeon to some extent via the use of smaller or larger prosthetic stem which will seat more or less prominent in the femoral canal. Also, the use of prosthetic heads with shorter or longer neck can affect the anatomical length of the
.734 .176 .777 .756 .921 .883 .082 .015 .586 .715 .0012 .28 .08 .831 .554 .017
femur and femoral offset. Change in the femoral offset can potentially place more tension on the abductor muscles, cause pelvic obliquity, and cause a perception of LLD. Figure 2 shows the cumulative number of patients who have a perception of LLD and the distribution of the anatomical femoral leg length. This figure shows that regardless of the difference in the anatomical femoral length between the operative and nonoperative sides, around 50%-60% of the patients will have a perception of LLD. This is true even when the anatomical femoral length discrepancy is only 1 mm. As this figure shows, other factors like coronal and sagittal knee alignment, pelvic obliquity, or certain foot deformities (eg, flat foot deformity) are affecting the patients' perception of the limb length. These factors are not extensively studied and frequently ignored when assessing the patients. In a previously published study among patients before THA, it was pointed out that perception of LLD did not correlate with anatomical leg length [23]. Instead correlation was reported with other variables like coronal and sagittal knee alignment. Patients who had more than 10 of flexum/recurvatum were 2.1 times more likely to percept LLD. Also, those who had irreducible varus/valgus deformity of the knee were 4 times more likely to have a perception of LLD. Similar to that study, we found correlation between LLD and knee flexum/recurvatum deformities in the study cohort. Coronal and sagittal knee alignment can be easily assessed during the physical examination and should not be missed as one of the causes of LLD after THA.
Table 3 Results of Multiple Logistic Regressions, With Perception of Leg Length Discrepancy as Outcome Variable.
Pelvic incidence ( ) Sacral slope ( ) Pelvic version ( ) Pelvic obliquity ( ) Pelvic axial rotation ( ) Difference in anatomical femoral length (cm) Difference in anatomical tibial length (cm) Difference in functional leg length (cm) Difference in anatomical leg length (cm) Difference in femoral offset (cm) Difference in neck shaft angle ( ) Difference in femoral torsion ( ) Difference in knee varus/valgus angle ( ) Difference in knee flexum/recurvatum angle ( ) Difference in hip-knee-ankle angle ( ) Difference in the middle of tibial plafond to floor distance (cm)
Odds Ratio
Coefficient
Standard Error
P Value
95% confidence Interval
1.016 1.036 0.995 1.056 0.971 0.919 1.864 1.062 1.065 1.002 0.994 1.004 0.91 1.071 1.07 14.343
0.015 0.035 0.004 0.054 0.028 0.083 0.622 0.06 0.063 0.002 0.005 0.004 0.093 0.069 0.068 2.663
0.014 0.021 0.02 0.237 0.047 0.243 0.464 0.21 0.201 0.029 0.035 0.012 0.054 0.029 0.124 0.89
.279 .089 .829 .021 .551 .732 .18 .775 .754 .921 .881 .712 .087 .021 .583 .003
0.012 to 0.044 0.005 to 0.077 0.438 to 0.351 0.008-0.101 0.122 to 0.065 0.56 to 0.393 0.286 to 1.532 0.352 to 0.472 0.331 to 0.458 0.055 to 0.061 0.075 to 0.065 0.019 to 0.029 0.201 to 0.013 0.01-0.127 0.175 to 0.311 0.918-4.408
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
5
Fig. 2. Regardless of the difference in the anatomical leg length, always 50%-60% of the patients complain of LLD, even when the difference is only 1 mm or less.
One of the potential variables that was not fully investigated in this study is axial pelvic rotation [14]. The axial rotation can result in more flexion of the hip which axial rotation is toward (the hip in more posterior position). This will result in more extension of the hip which the axial rotation is away from (the hip in more anterior position). Patients may have a perception of the shorter leg with the hip in flexion and longer leg with the hip in extension. This can potentially affect patients' perception of the leg length after surgery in extreme cases and requires investigation in future studies. We do not have the preoperative EOS images in all these patients, as a result we cannot assess if the pelvic rotation was caused by LLD or it was an adaptation of the posture and pelvic rotation for LLD. The effect of foot deformities on perception of LLD also requires more investigation. The concept of functional length as measured by sterEOS software has limitations as it is not possible to thoroughly assess the foot deformities in EOS anteroposterior and lateral images. Conclusion Other factors than anatomical femoral length can affect patients' perception of the LLD after THA. In this study, the anatomical femoral length did not correlate with patients' perception of the LLD but other variables like distance from the middle of the tibial plafond and ground, sagittal knee alignment, and pelvic obliquity were correlated with the patients' perception of LLD. These important factors can be found during regular physical examination and discussed with patients before THA. A comprehensive physical examination to investigate spinal deformities, pelvic obliquity, abductor muscle weakness, sagittal and coronal knee alignment, and foot deformity in these patients is critical. EOS 3D imaging in standing position including the foot view can be used to find the true cause of LLD in a medicolegal case or if the surgeon is planning to perform a revision THA. References [1] Blackley HR, Howell GE, Rorabeck CH. Planning and management of the difficult primary hip replacement: preoperative planning and technical considerations. Instr Course Lect 2000;49:3e11.
[2] Edeen J, Sharkey PF, Alexander AH. Clinical significance of leg-length inequality after total hip arthroplasty. Am J Orthop (Belle Mead NJ) 1995;24:347e51. [3] McCrory JL, White SC, Lifeso RM. Vertical ground reaction forces: objective measures of gait following hip arthroplasty. Gait Posture 2001;14:104e9. [4] Austin MS, Hozack WJ, Sharkey PF, Rothman RH. Stability and leg length equality in total hip arthroplasty. J Arthroplasty 2003;18:88e90. [5] Hofmann AA, Skrzynski MC. Leg-length inequality and nerve palsy in total hip arthroplasty: a lawyer awaits! Orthopedics 2000;23:943e4. [6] Parvizi J, Sharkey PF, Bissett GA, Rothman RH, Hozack WJ. Surgical treatment of limb-length discrepancy following total hip arthroplasty. J Bone Joint Surg Am 2003;85-A:2310e7. [7] Pritchett JW. Nerve injury and limb lengthening after hip replacement: treatment by shortening. Clin Orthop Relat Res 2004:168e71. [8] Rand JA, Ilstrup DM. Comparison of Charnley and T-28 total hip arthroplasty. Clin Orthop Relat Res 1983:201e5. [9] Turula KB, Friberg O, Lindholm TS, Tallroth K, Vankka E. Leg length inequality after total hip arthroplasty. Clin Orthop Relat Res 1986:163e8. [10] Williamson JA, Reckling FW. Limb length discrepancy and related problems following total hip joint replacement. Clin Orthop Relat Res 1978:135e8. [11] Upadhyay A, York S, Macaulay W, McGrory B, Robbennolt J, Bal BS. Medical malpractice in hip and knee arthroplasty. J Arthroplasty 2007;22:2e7. [12] Flecher X, Ollivier M, Argenson JN. Lower limb length and offset in total hip arthroplasty. Orthop Traumatol Surg Res 2016;102:S9e20. [13] Dubousset J, Charpak G, Dorion I, Skalli W, Lavaste F, Deguise J, et al. [A new 2D and 3D imaging approach to musculoskeletal physiology and pathology with low-dose radiation and the standing position: the EOS system]. Bull Acad Natl Med 2005;189:287e97. discussion 297e300. [14] Cummings G, Scholz JP, Barnes K. The effect of imposed leg length difference on pelvic bone symmetry. Spine (Phila Pa 1976) 1993;18:368e73. [15] Than P, Szuper K, Somoskeoy S, Warta V, Illes T. Geometrical values of the normal and arthritic hip and knee detected with the EOS imaging system. Int Orthop 2012;36:1291e7. [16] Gurney B. Leg length discrepancy. Gait Posture 2002;15:195e206. [17] Maloney WJ, Keeney JA. Leg length discrepancy after total hip arthroplasty. J Arthroplasty 2004;19:108e10. [18] Mannello DM. Leg length inequality. J Manipulative Physiol Ther 1992;15:576e90. [19] 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;33:905e9. [20] Guichet JM, Spivak JM, Trouilloud P, Grammont PM. Lower limb-length discrepancy. An epidemiologic study. Clin Orthop Relat Res 1991:235e41. [21] Knutson GA. Anatomic and functional leg-length inequality: a review and recommendation for clinical decision-making. Part I, anatomic leg-length inequality: prevalence, magnitude, effects and clinical significance. Chiropr Osteopat 2005;13:11. [22] Knutson GA. Anatomic and functional leg-length inequality: a review and recommendation for clinical decision-making. Part II. The functional or unloaded leg-length asymmetry. Chiropr Osteopat 2005;13:12. [23] Lazennec JY, Brusson A, Rousseau MA, Robbins CB, Pour AE. Do patients' perceptions of leg length correlate with standing 2- and 3-dimensional radiographic imaging? J Arthroplasty 2016;31:2308e13.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Does Patients' Perception of Leg Length After Total Hip Arthroplasty Correlate With Anatomical Leg Length? Jean Y. Lazennec, MD, PhD a, b, Dominique Folinais, MD c, Caroline Florequin, MS b, Aidin E. Pour, MD d, * ^pitaux de Paris, Paris, France Department of Orthopaedic and Trauma Surgery, La Piti e-Salp^ etri ere Hospital, Assistance Publique des Ho Anatomy Department, Universit e Pierre et Marie Curie (UPMC), Paris, France c RIM Maussins-Nollet, Paris, France d Department of Orthopaedic Surgery, University of Michigan, Ann Arbor, MI a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 July 2017 Received in revised form 4 December 2017 Accepted 5 December 2017 Available online xxx
Background: This study assessed the correlation among the patients' perception of the leg length discrepancy (LLD) after total hip arthroplasty (THA) and the anatomical and functional leg length, pelvic and knee alignments, and foot height. Methods: Patients without significant spinal pathology or previous spine or lower extremity surgery who underwent primary THA (101 patients) were evaluated using EOS images obtained in standing position. All 3-dimensional measures were evaluated and compared for the repeatability and reproducibility and correlation with patients' perception of leg length. Results: In our study, the anatomical femoral length (odds ratio [OR] 0.9, P ¼ .732) did not correlate with patients' perception of the LLD, but other variables like the distance from the middle of the tibial plafond to ground (OR 14.3, P ¼ .003), sagittal knee alignment (OR 1.07, P ¼ .021), and pelvic obliquity (OR 1.05, P ¼ .021) were correlated with the patients’ perception of LLD. Conclusion: The LLD is a multifactorial complication. We found that the anatomical femoral length as the factor that can be modified with THA technique or choice of prosthesis is not the only important factor. We recommend comprehensive physical examination to investigate spinal deformities, pelvic obliquity, abductor muscle weakness, sagittal and coronal knee alignment, and foot deformity in patients who complain of LLD after THA. © 2017 Elsevier Inc. All rights reserved.
Level of Evidence: Level IV diagnostic case series Keywords: leg length three-dimensional radiography knee alignments total hip arthroplasty
Leg length discrepancy (LLD) is a common postoperative complication after total hip arthroplasty (THA) [1e3]. It can result in patient dissatisfaction, limping, need for shoe lift, low back pain, hip instability, and revision surgery [4e10]. According to a survey study of the members of the American Association of Hip and Knee Surgeons by Upadhyay et al [11], LLD was the second most common
No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2017.12.004. -Salpe ^trie re Hospital Ethical review: Study approval was obtained from La Pitie review board. -Salpe ^trie re Hospital. Location of Study: La Pitie * Reprint requests: Aidin E. Pour, MD, Department of Orthopaedic Surgery, University of Michigan, 2912 Taubman Center, SPC 5328, 1500 E. Medical Center Dr., Ann Arbor, MI 48109. https://doi.org/10.1016/j.arth.2017.12.004 0883-5403/© 2017 Elsevier Inc. All rights reserved.
reason for litigation, and 8% of surgeons had been a defendant in a legal case secondary to this complication. Among the factors important in the LLD, femoral length and offset are the 2 main factors that a surgeon can modify with surgery. The femoral length can be affected by the choice of the stem design and fixation technique and the length of the prosthetic femoral head. The increased offset due to the use of high offset stems or prosthetic head with longer neck might also result in patients' perception of tightness in the abductor muscles, pelvic obliquity, and as a result a perception of LLD [12]. Regular anteroposterior radiographs of the pelvis are 2 dimensional only and they cannot accurately measure the leg length. Despite its better accuracy, computed tomography measures the anatomical length in the supine position instead of the standing position. The standing position is the position in which LLD is perceived by patients. Unlike the anatomical length, the functional length integrates the lengths of the femur and tibia, hip flexion contracture, and the coronal and sagittal knee alignments. Pelvic obliquity and axial
2
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
automatically measured all the variables used in this study (Table 1). The femoral offset is the distance between the center of the femoral head and the greater trochanter. Axial pelvic rotation is the angle between the line passing through the centers of both hips and the X-ray beam, with negative rotation being rotation toward the operative hip and positive rotation being rotation toward the nonoperative hip. If the pelvis rotates toward one side, one hip will be in more flexion (the side that the pelvis turns toward) which can give a perception of shorter leg. The opposite hip will be in more extension (the side that the pelvis rotates away from) which can give a perception of longer leg [14]. The pelvic obliquity was measured as the distance between the horizontal line and acetabular roof that was lower in a coronally tilted pelvis. Other variables included the hip-knee-ankle angle and the femoral and tibial mechanical axes (FMA, TMA). The FMA is the line connecting the center of the femoral head to the center of the femoral notch, and the TMA is the line from the center of the tibial plateau (interspinous intercruciate midpoint) extending distally to the center of the tibial plafond. Hip-knee-ankle angle is the angle between the FMA and TMA with neutral angle being zero. Tibial and femoral rotation and knee flexion/hyperextension angle were also derived from the 3D reconstruction [15]. We used the following definitions for measuring the length (Fig. 1):
pelvic rotation which affect the patients' perception of leg length will also be taken into account in this position. In this study, these questions were asked: (1) Does the perception of LLD correlate better with the anatomical or functional length? and (2) Does the perception of LLD correlate with other variables including the femoral offset, pelvic obliquity, sacral slope, axial pelvic rotation, and coronal or sagittal knee alignment and foot and ankle height? Our hypotheses were that the functional length would correlate with the perception of LLD better than the anatomical length. We also hypothesized that the perception of LLD correlates with coronal and sagittal knee alignment, axial pelvic rotation, and obliquity. Methods After obtaining Institutional Review Board approval, 101 consecutive patients (aged 18-80 years) who presented to our clinic between January 2014 and December 2016 were reviewed in this retrospective study. Patients had primary THA for degenerative hip arthritis. All the patients with symptomatic spinal pathology, previous spinal surgery, lower extremity open reduction and internal fixation, osteotomy, or arthroplasty (knee, ankle) were excluded. Thirty-eight hips were performed through an anterolateral approach, 37 through a posterior approach, and 26 through a direct anterior approach. All procedures utilized cementless implants. The bearing surface was ceramic-on-ceramic in 56 patients and metal or ceramic-on-polyethylene in 45 patients. All patients were evaluated using standing 3-dimensional (3D) imaging system for spine-related pain and lower extremityerelated pain (EOS; EOS Imaging SA, Paris, France). They were asked about their perception of the leg lengths. Each patient stood comfortably in the EOS machine and the position was specifically checked to avoid superimposition of anatomical structures on the lateral view because this would make 3D reconstruction impossible. The simultaneous biplanar acquisition was used to perform stereoradiographic 3D modeling of each lower extremity using specialized software (sterEOS 3D; EOS Imaging SA) according to a previously described method [13]. The software used the bony landmarks to determine the femoral and tibial torsions in 3D images. After 3D modeling, the software
Anatomical femoral length: distance between the center of the femoral head (A) and the center of the trochlea (B). Anatomical tibial length: distance between the center of the tibial spine (intercondylar eminence) (C) and the center of the ankle joint (D). Functional length: distance between the center of the femoral head to the center of the ankle joint (AD). Anatomical length: sum of the anatomical femoral and tibial lengths (AB þ CD). Foot deformities cannot be assessed in the EOS images thoroughly. For this study, the distance between the middle of the tibial plafond and the ground was considered the height of the foot as a possible factor affecting the LLD. The patients were divided into 2 groups for simple comparison of the means for all the measured variables. The first group (case group) included the patients with a
Table 1 Measurements of Study Variables in Operative and Nonoperative Sides in the Cohort. Variable
No perception of LLD Femoral offset (cm) Femoral neck-shaft angle ( ) Anatomical femoral length (cm) Anatomical tibial length (cm) Anatomical leg length (cm) Functional leg length (cm) Knee varus/valgus angle ( ) Hip-knee-ankle angle ( ) Knee flexum/recurvatum ( ) With perception of LLD Femoral offset (cm) Femoral neck-shaft angle ( ) Anatomical femoral length (cm) Anatomical tibial length (cm) Anatomical leg length (cm) Functional leg length (cm) Knee varus/valgus angle ( ) Hip-knee-ankle angle ( ) Knee flexum/recurvatum ( )
Operative
Nonoperative
P Value
Mean
Minimum
Maximum
Mean
Minimum
Maximum
4.4 132.4 41.9 35.1 77 77.6 0.1 4.5 6.6
3.3 126.3 37.7 31.8 69.7 69.9 6.7 0.6 7.88
5.6 139.3 47.3 40 86.1 87.2 12.1 7.9 30.2
4.3 124.2 41.5 35.4 77 77.4 1.6 5 5.9
3.4 111.9 37.4 31.8 69.8 70.2 10 2.3 6.4
5.1 140.6 46.3 40.3 85.8 86.4 4.52 7.8 27.3
.667 <.001 .005 <.001 .976 .008 .008 .09 .554
4 133.8 41.3 34.8 76.1 76.7 0.3 4.6 6.3
2.4 121.2 34.5 29.8 65.6 66.3 7.7 0.8 11.2
5.6 149.5 47.3 40.6 87.9 88.5 15.3 10.2 26.9
4.1 125.9 41 35.1 76.1 76.5 0.5 4.9 1.8
3.2 109.9 33.9 23 65.7 65.5 12.8 1.7 11.4
5.7 135.9 47.6 42.1 89.6 90.3 8.5 11.2 28.9
.06 <.001 .011 <.001 .687 .036 .825 .269 <.001
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
3
perception of LLD. The second group (control group) included the patients without perception of LLD. Statistical Analysis The statistical analysis was performed with MedCalc (version 11.2.0.0, Mariakerke, Belgium) and Stata 14.1 (StataCorp LP, College Station, TX). A post hoc power analysis was conducted using G*Power v. 3.1.9 to determine the power of the study considering the current sample size. Using an alpha of 0.05 and a medium effect size (d ¼ 0.5), a total sample size of 100 provided a study power of 0.81. Repeatability and reproducibility were assessed by calculating the interobserver and intraobserver intraclass correlation coefficient (ICC) and its 95% confidence interval. The ICC is defined as the ratio between explained variances (variance attributable to the cause of variation: observer factor; repetition of measurement) and overall variance (explained variance þ error variance). Comparison of repeatability and reproducibility of each parameter was performed using the Fisher-Snedecor test for comparison of variances. Quantitative variables were described using the mean (M), the mean difference (d), the standard deviation (SD), and the ICC with its confidence interval. We considered an ICC of >0.90 as high, between 0.80 and 0.90 as moderate, and <0.80 as insufficient. Normal distribution of the values was checked by means of the Shapiro-Wilk normality test for each series of measurements. For data with normal distribution, paired Student's t-test was used for analysis. For data without normal distribution, related samples Wilcoxon signed rank test was used for the analysis. The significance level was set at 5%. Multiple logistic regression was used to investigate the correlation between the patients' perception of LLD and variables found significant in the univariate analysis. The significance level was set at 0.05. Results The detailed data are presented in Tables 1 and 2. In univariate and multiple logistic regression (Table 3), no correlation was found among the perception of LLD and difference in anatomical femoral length (P ¼ .393), anatomical leg length (P ¼ .458), and functional leg length (P ¼ .472). Among those whose operative legs were shorter, 65.7% (23 of 35) had a perception of LLD as compared to 56% of patients in whom the operative side was longer (37 of 66) (P ¼ .347). Tables 1 and 2 present the data regarding other variables that could correlate with the patients' perception of limb length. In univariate and multiple logistic regression (Table 3), association was found among the patients’ perception of LLD with pelvic obliquity (odds ratio [OR] 1.06, P ¼ .032), difference in knee flexum/ recurvatum (OR 1.07, P ¼ .034), and the difference in distance between the middle of tibial plafond and ground (OR 7.3, P ¼ .031). Discussion
Fig. 1. Anatomic vs functional length. (1) Anatomical femoral length: distance between the center of the femoral head (A) and the center of the trochlea (B). (2) Anatomical tibial length: distance between the center of the tibial spine (intercondylar eminence) (C) and the center of the ankle joint (D). (3) Functional length: distance between the center of the femoral head to the center of the ankle joint (AD). (4) Anatomical length: sum of the anatomical femoral and tibial lengths (AB þ CD).
This retrospective study investigated the correlation between the patients’ perception of LLD and different variables measured in EOS Imaging after primary THA. In the final multiple regression analysis, the perception of LLD was associated with the pelvic obliquity, difference in knee flexum/recurvatum angle, and difference in distance from the middle of the tibial plafond and ground. No correlation was found among the perception of LLD and anatomical femoral length, anatomical leg length, and functional leg length.
4
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
Table 2 Difference Between the Operative and Nonoperative Side Variables. No Perception of LLD
Difference in anatomical femoral length (cm) Difference in anatomical tibial length (cm) Difference in functional length (cm) Difference in anatomical leg length (cm) Difference in femoral offset (cm) Difference in neck-shaft angle ( ) Difference in knee varus/valgus angle ( ) Difference in knee flexum/recurvatum angle ( ) Difference in hip-knee-ankle angle ( ) Difference in femoral torsion ( ) Difference in the middle of tibial plafond to ground distance (cm) Pelvic incidence ( ) Sacral slope ( ) Pelvic version ( ) Axial pelvic rotation ( ) Pelvic obliquity (mm)
Perception of LLD
P Value
Mean
Minimum
Maximum
Mean
Minimum
Maximum
0.4 0.4 0.3 0.1 0.8 8.5 1.5 0.7 0.4 2.1 0.1 58.1 39.4 18.7 1.2 0.9
1.7 1.2 1.7 2.2 19.3 1.5 5.6 20.9 4.9 62.7 0.6 33.5 14.4 3.8 6.6 20.6
1.4 0.7 1.4 1.4 12 20.1 11.9 21.9 3.3 31.7 0.6 86.8 59.7 37.6 4.9 11.1
0.4 0.3 0.3 0.1 0.7 8.3 0.1 4.5 0.25 0.93 0.1 61.3 43 18.3 1.7 5.3
2 1.6 2.8 3.2 14.5 6.8 8.4 21 4.2 40.8 0.9 33.5 22.1 3.5 10 20.3
3 0.6 2.6 2.5 15 24.5 12.4 27.9 2.9 35.2 0.8 106.7 73.9 43.6 11.3 26.6
This study has limitations. Preoperative EOS Imaging was not available for these patients. As a result, we cannot compare the preoperative and postoperative EOS images in terms of changes in the leg length caused by THA. The sample size is rather small but the post hoc power calculation showed that current sample size provided enough power (0.81) to detect the difference between the 2 groups. We also did not investigate the degree of any foot deformities in patients radiographically as flat foot deformity may potentially affect the patients' perception of the leg length. Instead, the distance between the middle of the tibial plafond and the ground was used as the foot arch height. LLD is a one of the complications of THA which sometimes results in early revision arthroplasty or medicolegal litigation [16e18]. In a study, 30% of the patients who underwent THA perceived different leg length [19]. Only 36% of these patients had a true LLD in imaging studies. Imaging studies performed in supine position cannot measure the true functional leg length accurately. The prevalence of anatomical LLD is diversely described in the literature [20e22]. In assessment of the normal population without any perception of the LLD, different authors reported that there was an anatomical LLD of 5 mm or more in 90% of the population [20e22]. Change in the anatomical femoral length can be controlled by the operating surgeon to some extent via the use of smaller or larger prosthetic stem which will seat more or less prominent in the femoral canal. Also, the use of prosthetic heads with shorter or longer neck can affect the anatomical length of the
.734 .176 .777 .756 .921 .883 .082 .015 .586 .715 .0012 .28 .08 .831 .554 .017
femur and femoral offset. Change in the femoral offset can potentially place more tension on the abductor muscles, cause pelvic obliquity, and cause a perception of LLD. Figure 2 shows the cumulative number of patients who have a perception of LLD and the distribution of the anatomical femoral leg length. This figure shows that regardless of the difference in the anatomical femoral length between the operative and nonoperative sides, around 50%-60% of the patients will have a perception of LLD. This is true even when the anatomical femoral length discrepancy is only 1 mm. As this figure shows, other factors like coronal and sagittal knee alignment, pelvic obliquity, or certain foot deformities (eg, flat foot deformity) are affecting the patients' perception of the limb length. These factors are not extensively studied and frequently ignored when assessing the patients. In a previously published study among patients before THA, it was pointed out that perception of LLD did not correlate with anatomical leg length [23]. Instead correlation was reported with other variables like coronal and sagittal knee alignment. Patients who had more than 10 of flexum/recurvatum were 2.1 times more likely to percept LLD. Also, those who had irreducible varus/valgus deformity of the knee were 4 times more likely to have a perception of LLD. Similar to that study, we found correlation between LLD and knee flexum/recurvatum deformities in the study cohort. Coronal and sagittal knee alignment can be easily assessed during the physical examination and should not be missed as one of the causes of LLD after THA.
Table 3 Results of Multiple Logistic Regressions, With Perception of Leg Length Discrepancy as Outcome Variable.
Pelvic incidence ( ) Sacral slope ( ) Pelvic version ( ) Pelvic obliquity ( ) Pelvic axial rotation ( ) Difference in anatomical femoral length (cm) Difference in anatomical tibial length (cm) Difference in functional leg length (cm) Difference in anatomical leg length (cm) Difference in femoral offset (cm) Difference in neck shaft angle ( ) Difference in femoral torsion ( ) Difference in knee varus/valgus angle ( ) Difference in knee flexum/recurvatum angle ( ) Difference in hip-knee-ankle angle ( ) Difference in the middle of tibial plafond to floor distance (cm)
Odds Ratio
Coefficient
Standard Error
P Value
95% confidence Interval
1.016 1.036 0.995 1.056 0.971 0.919 1.864 1.062 1.065 1.002 0.994 1.004 0.91 1.071 1.07 14.343
0.015 0.035 0.004 0.054 0.028 0.083 0.622 0.06 0.063 0.002 0.005 0.004 0.093 0.069 0.068 2.663
0.014 0.021 0.02 0.237 0.047 0.243 0.464 0.21 0.201 0.029 0.035 0.012 0.054 0.029 0.124 0.89
.279 .089 .829 .021 .551 .732 .18 .775 .754 .921 .881 .712 .087 .021 .583 .003
0.012 to 0.044 0.005 to 0.077 0.438 to 0.351 0.008-0.101 0.122 to 0.065 0.56 to 0.393 0.286 to 1.532 0.352 to 0.472 0.331 to 0.458 0.055 to 0.061 0.075 to 0.065 0.019 to 0.029 0.201 to 0.013 0.01-0.127 0.175 to 0.311 0.918-4.408
J.Y. Lazennec et al. / The Journal of Arthroplasty xxx (2017) 1e5
5
Fig. 2. Regardless of the difference in the anatomical leg length, always 50%-60% of the patients complain of LLD, even when the difference is only 1 mm or less.
One of the potential variables that was not fully investigated in this study is axial pelvic rotation [14]. The axial rotation can result in more flexion of the hip which axial rotation is toward (the hip in more posterior position). This will result in more extension of the hip which the axial rotation is away from (the hip in more anterior position). Patients may have a perception of the shorter leg with the hip in flexion and longer leg with the hip in extension. This can potentially affect patients' perception of the leg length after surgery in extreme cases and requires investigation in future studies. We do not have the preoperative EOS images in all these patients, as a result we cannot assess if the pelvic rotation was caused by LLD or it was an adaptation of the posture and pelvic rotation for LLD. The effect of foot deformities on perception of LLD also requires more investigation. The concept of functional length as measured by sterEOS software has limitations as it is not possible to thoroughly assess the foot deformities in EOS anteroposterior and lateral images. Conclusion Other factors than anatomical femoral length can affect patients' perception of the LLD after THA. In this study, the anatomical femoral length did not correlate with patients' perception of the LLD but other variables like distance from the middle of the tibial plafond and ground, sagittal knee alignment, and pelvic obliquity were correlated with the patients' perception of LLD. These important factors can be found during regular physical examination and discussed with patients before THA. A comprehensive physical examination to investigate spinal deformities, pelvic obliquity, abductor muscle weakness, sagittal and coronal knee alignment, and foot deformity in these patients is critical. EOS 3D imaging in standing position including the foot view can be used to find the true cause of LLD in a medicolegal case or if the surgeon is planning to perform a revision THA. References [1] Blackley HR, Howell GE, Rorabeck CH. Planning and management of the difficult primary hip replacement: preoperative planning and technical considerations. Instr Course Lect 2000;49:3e11.
[2] Edeen J, Sharkey PF, Alexander AH. Clinical significance of leg-length inequality after total hip arthroplasty. Am J Orthop (Belle Mead NJ) 1995;24:347e51. [3] McCrory JL, White SC, Lifeso RM. Vertical ground reaction forces: objective measures of gait following hip arthroplasty. Gait Posture 2001;14:104e9. [4] Austin MS, Hozack WJ, Sharkey PF, Rothman RH. Stability and leg length equality in total hip arthroplasty. J Arthroplasty 2003;18:88e90. [5] Hofmann AA, Skrzynski MC. Leg-length inequality and nerve palsy in total hip arthroplasty: a lawyer awaits! Orthopedics 2000;23:943e4. [6] Parvizi J, Sharkey PF, Bissett GA, Rothman RH, Hozack WJ. Surgical treatment of limb-length discrepancy following total hip arthroplasty. J Bone Joint Surg Am 2003;85-A:2310e7. [7] Pritchett JW. Nerve injury and limb lengthening after hip replacement: treatment by shortening. Clin Orthop Relat Res 2004:168e71. [8] Rand JA, Ilstrup DM. Comparison of Charnley and T-28 total hip arthroplasty. Clin Orthop Relat Res 1983:201e5. [9] Turula KB, Friberg O, Lindholm TS, Tallroth K, Vankka E. Leg length inequality after total hip arthroplasty. Clin Orthop Relat Res 1986:163e8. [10] Williamson JA, Reckling FW. Limb length discrepancy and related problems following total hip joint replacement. Clin Orthop Relat Res 1978:135e8. [11] Upadhyay A, York S, Macaulay W, McGrory B, Robbennolt J, Bal BS. Medical malpractice in hip and knee arthroplasty. J Arthroplasty 2007;22:2e7. [12] Flecher X, Ollivier M, Argenson JN. Lower limb length and offset in total hip arthroplasty. Orthop Traumatol Surg Res 2016;102:S9e20. [13] Dubousset J, Charpak G, Dorion I, Skalli W, Lavaste F, Deguise J, et al. [A new 2D and 3D imaging approach to musculoskeletal physiology and pathology with low-dose radiation and the standing position: the EOS system]. Bull Acad Natl Med 2005;189:287e97. discussion 297e300. [14] Cummings G, Scholz JP, Barnes K. The effect of imposed leg length difference on pelvic bone symmetry. Spine (Phila Pa 1976) 1993;18:368e73. [15] Than P, Szuper K, Somoskeoy S, Warta V, Illes T. Geometrical values of the normal and arthritic hip and knee detected with the EOS imaging system. Int Orthop 2012;36:1291e7. [16] Gurney B. Leg length discrepancy. Gait Posture 2002;15:195e206. [17] Maloney WJ, Keeney JA. Leg length discrepancy after total hip arthroplasty. J Arthroplasty 2004;19:108e10. [18] Mannello DM. Leg length inequality. J Manipulative Physiol Ther 1992;15:576e90. [19] 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;33:905e9. [20] Guichet JM, Spivak JM, Trouilloud P, Grammont PM. Lower limb-length discrepancy. An epidemiologic study. Clin Orthop Relat Res 1991:235e41. [21] Knutson GA. Anatomic and functional leg-length inequality: a review and recommendation for clinical decision-making. Part I, anatomic leg-length inequality: prevalence, magnitude, effects and clinical significance. Chiropr Osteopat 2005;13:11. [22] Knutson GA. Anatomic and functional leg-length inequality: a review and recommendation for clinical decision-making. Part II. The functional or unloaded leg-length asymmetry. Chiropr Osteopat 2005;13:12. [23] Lazennec JY, Brusson A, Rousseau MA, Robbins CB, Pour AE. Do patients' perceptions of leg length correlate with standing 2- and 3-dimensional radiographic imaging? J Arthroplasty 2016;31:2308e13.