Total knee arthroplasty for patients with medial knee osteoarthritis improves trunk movement during gait

Abstract

BACKGROUND:

Previous studies have indicated that the kinematics of the knee joint affect the trunk and pelvis during gait. However, the factors that influence trunk movement in knee osteoarthritis patients during gait after surgery remain unclear.

OBJECTIVE:

To examine the effect of total knee arthroplasty (TKA) on trunk movement during gait by comparing knee osteoarthritis patients with healthy controls.

METHODS:

Fourteen medial knee osteoarthritis patients who underwent initial unilateral TKA and 11 controls participated in this study. Knee and hip joint flexion and trunk and pelvic tilts during gait were acquired using a three-dimensional motion analysis system. Knee joint range of motion, pain, and kinematic data were collected preoperatively and 1 year postoperatively for knee osteoarthritis patients.

RESULTS:

Knee extension limitation and pain significantly improved postoperatively compared with preoperative stages. Preoperatively, the peak anterior trunk tilt during the stance phase was significantly larger in osteoarthritis patients than in controls. The peak anterior trunk tilt during the stance phase was significantly smaller postoperatively than at preoperative stages.

CONCLUSIONS:

These results suggest that after TKA, the trunk movements of knee osteoarthritis patients were approximately equal to those of controls, with improvement in clinical outcomes such as knee extension limitation and pain.

Keywords

Knee osteoarthritis total knee arthroplasty gait analysis spinal alignment pelvic alignment sagittal alignment

1. Introduction

As the population continues to age rapidly, the number of patients with knee osteoarthritis (OA) is also increasing. Yoshimura et al. [1] estimated that 25.3 million people older than 40 years in Japan have radiographic findings of knee OA and that 7.8 million people have symptoms, such as pain.

The correlation between knee OA and low back pain has recently gained attention. Previous studies have reported that low back pain was present in 54.6–57.4% of patients with knee OA [2, 3]. Several studies proposed a “knee-spine syndrome” in radiological research [4, 5]. Murata et al. [5] reported that knee extension limitation assessed using a goniometer was greater in patients with lumbar lordosis $\leqslant$ 30 ${}^{\circ}$ and that lumbar lordosis was reduced in patients with a knee extension limitation $>$ 5 ${}^{\circ}$ . Additionally, in patients with knee OA who underwent total knee arthroplasty (TKA), the sacral slope in the static standing position was increased after TKA [6]. These findings demonstrated that changes in the knee joint alignment affected orientation of the spine and pelvis in the standing position.

Some studies investigated whether the kinematics of the knee influenced those of the spine or pelvis during gait. Harato et al. [7] reported that a simulated knee flexion contracture of 30 ${}^{\circ}$ led to increased anterior inclination of the trunk and pelvis during gait in healthy females. A recent study reported that spinal alignment parameters in the sagittal plane during gait of patients with knee OA after high tibial osteotomy (HTO; open-wedge method) were closer to those of healthy young subjects than to those of patients before HTO [8]. However, clinical outcomes including pain, knee joint range of motion (ROM), and knee kinematics during gait were not described in that study. Furthermore, the mean patient age in that study was 55.0 years. However, studies on older patients who underwent TKA are lacking. It is unclear whether TKA for older patients with knee OA also improves spinal movements during gait, and if so, what factors lead to changes in spinal movement during gait.

Accordingly, the purpose of this study was to identify the factors that led to changes in trunk and pelvic movements based on three-dimensional data analysis and clinical outcomes of patients with severe knee OA before and after TKA. We hypothesized that patients with knee OA would walk with an anterior inclination of the trunk and pelvis before TKA and that their trunk and pelvic kinematics during gait would be similar to those of healthy subjects, with improvement in clinical symptoms such as pain and function of the knee joint after TKA.

2. Methods

2.1 Participants

Fourteen patients with medial knee OA who underwent initial unilateral TKA (Scorpio NRG; Stryker, Kalamazoo, MI, USA) performed by an experienced knee surgeon between July 2011 and February 2015 in our university hospital were selected as the OA group (Table 1). All the patients had a knee flexion contracture of $>$ 5 ${}^{\circ}$ on the operatory knee before TKA. Twelve of these patients had symptomatic bilateral knee OA, and two patients had knee OA only on their left sides. Posterior stabilized implants were used for all patients. The exclusion criteria were spinal surgery before or after TKA, morphological change in the vertebral body due to traumatic or osteoporotic compression fractures, hip and low back pains during gait, history of central nervous system disease, serious heart or lung disease, rheumatoid arthritis, and inability to walk independently. Knee OA severity was determined by an orthopaedic surgeon, based on the classification of an anteroposterior knee radiograph by using the Kellgren-Lawrence grading scale [9]. None of the patients was classified under Grade I or II knee OA. Eight patients had Grade III and six patients had Grade IV knee OA. The control group was composed of 11 age-matched healthy volunteers without knee, hip and low back pains while walking (Table 1). Their knee joints had full range of motion in both flexion and extension. This study was approved by our institution’s Epidemiology Research Ethical Review Board (approval number: E Epd-204-3). All patients received an explanation of the study’s purpose and design, and each participant provided written informed consent.

Table 1
Demographic data for the OA and control groups

	OA		Control		$p$ value
Age (years)	74.8	(6.2)	74.4	(6.1)	0.866
Body height (cm)	152.3	(5.4) ${}^{*}$	158.3	(6.5)	0.019
Body weight (kg)	58.0	(9.2)	57.5	(10.6)	0.907
Body mass index (kg/m ${}^{2}$ )	25.0	(3.7)	22.8	(3.3)	0.139
Sex (male:female)	2:12		5:6		0.085
Operated sides (left:right)	8:6

Values are the mean (SD). OA, knee osteoarthritis group ( $n=$ 14); Control, control group ( $n=$ 11). ${}^{*}$ Significant difference from the Control at $p<$ 0.05.

2.2 Outcome measures

The following parameters were measured preoperatively and 1 year postoperatively for the knee OA group: (1) knee joint ROM, (2) the Japanese Orthopaedic Association (JOA) score to assess clinical symptoms, (3) the Numeric Rating Scale (NRS) to evaluate knee pain, and (4) gait analysis to examine kinematics during gait.

For the control group, only a gait analysis was performed. Clinical outcomes were not measured.

2.2.1 Measurement of ROM

Bilateral knee joint ROM was measured passively by one knee surgeon at 5 ${}^{\circ}$ intervals using a goniometer.

2.2.2 Clinical knee score

The JOA score of knee OA was used to evaluate knee joint treatment and clinical outcomes [10]. The JOA score consists of four parts: (1) pain on walking, (2) pain on ascending and descending stairs, (3) contracture/ROM, and (4) joint effusion. The JOA score ranges from 0 to 100 points, with a higher score indicating better knee joint function.

2.2.3 Pain evaluation

Knee pain during gait was evaluated using the NRS [11]. The NRS allows the patient to grade current knee pain by using an 11-point scale ranging from 0 to 10, with 0 indicating no pain and 10 indicating the worst pain imaginable.

2.2.4 Gait analysis

For gait analysis, a three-dimensional motion analysis system (VICON MX; Vicon Motion Systems, Oxford, UK) with 16 infrared cameras (operating at 100 Hz; Vicon Motion Systems) and eight force plates (1000 Hz; AMTI, Watertown, MI, USA) was used. Ground reaction force data were synchronized with marker coordinate data. Thirty-five passive reflective markers (14 mm diameter) were placed according to a commercially available kinematic model (Plug-in-Gait; Vicon ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ Peak, Oxford, UK). The participants were instructed to walk barefoot at a self-selected speed along a 10-m walkway three times.

Marker coordinate data were low-pass filtered (Butterworth 4 ${}^{\text{th}}$ -order filter; cut-off frequency, 6 Hz) using a plug-in software (ButterPlug; Vaquita Software, Zaragoza, Spain). Gait velocity, stride length, and cadence were extracted from the biomechanical model output, and the stride length was normalized to the subject’s height. All reference segments were collected using the plug-in gait marker set and model, which has been previously described in detail [12, 13]. Using the local coordinate system, the knee and hip movements were defined as movements of the tibial segment to the femoral segment and the femoral segment to the pelvic segment, respectively. On the global segment, trunk and pelvic movements were defined as movements of the thoracic and pelvic segments. Kinematic data were represented as “ $+$ ” (flexion/anterior tilt) and “ $-$ ” (extension/posterior tilt).

The magnitude of each angle was averaged across the 2-s static standing trial to calculate resting alignment. The gait kinematics of the trunk and pelvis were expressed relative to the participant’s standing alignment [14]. Initial contact was identified by the force plates detecting $>$ 10 N, while toe off was defined as the detection of $<$ 10 N. We defined one gait cycle from the initial contact of the instrumented leg to the next initial contact of the same leg. We then divided the gait cycle into the stance and swing phases. For the control group, the instrumented legs were selected to match the proportions of the left and right limbs evaluated in the OA group. These data were time-normalized to 100 data points. Kinematic data included the maximum and minimum values of each data point during the stance and swing phases and the values of range of motion during one gait cycle. For each participant, the average values of three trials were used in the statistical analyses.

2.3 Statistical analysis

Statistical analysis was performed using SPSS for Windows (version 23.0; IBM Japan, Tokyo, Japan). Data are expressed as the mean (standard deviation). To analyse the normality of data distribution, we first performed the Shapiro-Wilk test. Independent $t$ -test and chi-square test were used to compare the demographic data, while a paired $t$ -test or Wilcoxon signed-rank test was used to compare the clinical outcomes and kinematic data of the knee and hip on the non-operated side during gait in the OA group according to the normality of the distribution. The level of statistical significance was $p<$ 0.05. For gait parameters and kinematic data, independent $t$ -test or Mann-Whitney U test were used for independent data (controls versus before surgery and controls versus after surgery), and a paired $t$ -test or Wilcoxon signed-rank test were used for dependent data (before surgery versus after surgery). Since we performed a Bonferroni correction for $p$ values, we set that the level of statistical significance was $p<$ 0.017 for gait parameters and kinematic data.

Figure 1.

Temporal changes in knee and hip joint flexion as well as trunk and pelvic tilting movements during one gait cycle. One gait cycle was defined as the period from the initial contact of the instrumented leg to the next initial contact of the same leg. The dashed and dotted lines show the mean trace taken preoperatively and that taken 1 year postoperatively in the knee osteoarthritis group, respectively. The smooth line shows the mean trace of the control group.

3. Results

3.1 Demographic data

The body height was significantly smaller in the OA group than in the control group (Table 1). No significant differences between groups were observed in age, body weight, body mass index, or sex.

3.2 Clinical outcomes of the OA group

On the operated sides, knee joint extension ROM improved 1 year postoperatively compared to preoperative stages, whereas flexion ROM did not improve (Table 2). On the non-operated sides, no significant differences were observed in the ROM of flexion (118.5 [19.7] vs. 113.0 [16.2] ${}^{\circ}$ , $p=$ 0.240) or extension ( $-$ 11.5 [8.9] vs. $-$ 10.0 [7.8] ${}^{\circ}$ , $p=$ 0.193) in the OA group 1 year postoperatively compared to preoperative stages. The JOA score was significantly higher 1 year postoperatively than at the preoperative stage. The NRS score for the knee joint was significantly lower 1 year postoperatively than at the preoperative stage.

Table 2
Clinical outcomes on the operated knees in the OA group preoperatively versus 1 year postoperatively

	Pre-op		Post-op		$p$ value
Flexion ROM ( ${}^{\circ}$ )	112.1	(17.2)	113.6	(10.3)	0.738
Extension ROM ( ${}^{\circ}$ )	$-$ 13.6	(8.9)	$-$ 2.1	(5.4) ${}^{{\dagger}{\dagger}}$	0.001
JOA score (points)	48.2	(7.7)	73.2	(13.7) ${}^{{\dagger}{\dagger}}$	$<$ 0.001
NRS (points)	7.0	(2.3)	0.7	(0.9) ${}^{{\dagger}{\dagger}}$	0.001

Values are the mean (SD). OA, knee osteoarthritis; Pre-op, preoperatively; Post-op, 1 year postoperatively; ROM, range of motion; JOA score, The Japanese Orthopaedic Association score for knee osteoarthritis; NRS, Numeric Rating Scale. ${}^{{\dagger}{\dagger}}$ Significant difference from Pre-op at $p<$ 0.01.

3.3 Velocity, stride length, and cadence during gait

Compared to the control group, velocity and stride length during gait were significantly lower in the OA group both preoperatively and 1 year postoperatively (Table 3). Cadence in the OA group was significantly lower preoperatively than that in the control group, but not at 1 year postoperatively. No significant differences were observed in velocity, cadence, or stride length during gait in the OA group 1 year postoperatively compared with preoperative stages.

Table 3
Velocity, stride length and cadence during gait

	OA				Control
	Pre-op		Post-op
Velocity (m/min)	43.5	(9.3) ${}^{*}$	50.1	(8.9) ${}^{*}$	68.6	(13.7)
Stride length (% body height)	0.55	(0.10) ${}^{*}$	0.61	(0.12) ${}^{*}$	0.73	(0.10)
Cadence (step/min)	103.7	(12.6) ${}^{*}$	109.1	(11.1)	118.3	(10.7)

Values are the mean (SD). OA, knee osteoarthritis group; Control, control group; Pre-op, preoperatively; Post-op, 1 year postoperatively. ${}^{*}$ Significant difference from Control at $p<$ 0.017.

Table 4

Kinematic maximum and minimum values and range of motion during gait for the knee and hip joints on the operated sides or the trunk and pelvis

	OA				Control
	Pre-op		Post-op
Knee max flexion stance ( ${}^{\circ}$ )	24.1	(10.3) ${}^{*}$	22.1	(8.8) ${}^{*}$	37.0	(8.8)
Knee min flexion stance ( ${}^{\circ}$ )	9.6	(9.1)	4.8	(5.6)	6.2	(6.5)
Knee max flexion swing ( ${}^{\circ}$ )	43.0	(13.2) ${}^{*}$	48.2	(9.2)	56.5	(10.8)
Knee range of motion ( ${}^{\circ}$ )	33.6	(12.3) ${}^{*}$	44.7	(11.6) ${}^{{\dagger}}$	51.6	(6.9)
Hip max flexion stance ( ${}^{\circ}$ )	30.3	(7.5)	29.0	(7.6)	28.4	(4.0)
Hip max extension stance ( ${}^{\circ}$ )	$-$ 3.9	(7.6) ${}^{*}$	$-$ 6.7	(7.8)	$-$ 11.2	(3.7)
Hip range of motion ( ${}^{\circ}$ )	36.1	(3.7) ${}^{*}$	37.5	(8.5)	41.4	(5.1)
Trunk max anterior tilt stance ( ${}^{\circ}$ )	11.4	(4.1) ${}^{*}$	6.9	(2.8) ${}^{{\dagger}}$	6.8	(2.0)
Trunk min anterior tilt stance ( ${}^{\circ}$ )	7.8	(4.1) ${}^{*}$	4.2	(2.9) ${}^{{\dagger}}$	4.1	(1.5)
Trunk range of motion ( ${}^{\circ}$ )	4.2	(1.3)	3.3	(0.5)	3.3	(1.1)
Pelvis max anterior tilt stance ( ${}^{\circ}$ )	5.2	(2.8) ${}^{*}$	4.0	(2.1) ${}^{*}$	0.4	(2.6)
Pelvis min anterior tilt stance ( ${}^{\circ}$ )	2.6	(3.3) ${}^{*}$	1.5	(2.3) ${}^{*}$	$-$ 1.8	(2.7)
Pelvis range of motion ( ${}^{\circ}$ )	3.2	(1.1)	3.1	(1.0)	2.7	(1.2)

Values are the mean (SD). OA, knee osteoarthritis group; Control, control group; Pre-op, preoperatively; Post-op, 1 year postoperatively; $+$ , flexion/anterior tilt; $-$ , extension/posterior tilt; max, maximum; min, minimum; stance, stance phase; swing, swing phase. ${}^{*}$ Significant difference from Control at $p<$ 0.017. ${}^{{\dagger}}$ Significant difference from Pre-op at $p<$ 0.017.

3.4 Kinematics during gait (Fig. 1)

3.4.1 Intergroup differences (OA versus control group)

Preoperatively, maximum knee flexion during the stance and swing phases and knee range of motion during one gait cycle and maximum hip extension during the stance phase were significantly smaller in the OA group than in the control group (Table 4). Maximum and minimum anterior tilts of the trunk and pelvis during the stance phase were significantly larger in the OA group than in the control group. No significant differences were observed in the other kinematic values between the preoperative OA patients and controls.

After TKA, maximum knee flexion during the stance phase was significantly smaller in the OA group than in the control group. In addition, maximum and minimum anterior tilts of the pelvis during the stance phase were significantly larger in the OA group than in the control group. No significant differences in the other kinematic values that had significant differences before surgery, such as maximum knee flexion during the swing phase, knee range of motion during one gait cycle, maximum hip extension during the stance phase, and maximum and minimum anterior tilts of the trunk, were observed postoperatively between the OA patients and controls.

3.4.2 Intragroup differences (preoperative versus 1 year postoperative)

Knee range of motion during one gait cycle was significantly larger 1 year postoperatively than at preoperative stages (Table 4). Maximum and minimum trunk anterior tilts during the stance phase were significantly smaller 1 year postoperatively than at the preoperative stage. No significant differences in the other kinematic values were observed between preoperative and 1 year postoperative stages.

On the non-operated sides in the OA group, the 1-year postoperative values of maximum (20.8 [7.9] vs. 26.0 [12.3] ${}^{\circ}$ , $p=$ 0.088) and minimum (6.0 [7.0] vs. 4.9 [5.9] ${}^{\circ}$ , $p=$ 0.522) knee flexion during the stance phase, maximum knee flexion during the swing phase (47.9 [9.4] vs. 46.8 [13.1] ${}^{\circ}$ , $p=$ 0.733), knee range of motion during one gait cycle (47.9 [9.4] vs. 46.8 [13.1] ${}^{\circ}$ , $p=$ 0.733), maximum hip flexion (30.3 [4.7] vs. 27.6 [4.0] ${}^{\circ}$ , $p=$ 0.210) and extension ( $-$ 4.6 [6.9] vs. $-$ 7.6 [7.9] ${}^{\circ}$ , $p=$ 0.239), and hip range of motion during one gait cycle (36.4 [5.7] vs. 35.6 [7.1] ${}^{\circ}$ , $p=$ 0.593) showed no significant differences from those in the preoperative stages.

4. Discussion

There were two important findings in the present study. First, preoperatively, the anterior trunk and pelvic tilts of patients with knee OA were larger than those of the controls during gait. Second, 1 year postoperatively, the anterior tilt of the trunk during gait was smaller than the preoperative measurements, and there was an improvement in clinical outcomes such as knee extension ROM and pain; thus, trunk kinematics of patients with knee OA were approximately equal to those of the healthy subjects.

Patients with knee OA had larger anterior tilts of the trunk and pelvis before surgery than the controls, as was hypothesized. This result suggested that “knee-spine syndrome” occurred during both standing and gait in the patients with knee OA. A previous study reported that when a 30 ${}^{\circ}$ unilateral knee flexion contracture was simulated in healthy females using a knee brace, anterior inclination of the trunk and pelvis during walking increased, but they were not observed at a 15 ${}^{\circ}$ knee flexion contracture [7]. Patients with severe knee OA, such as the participants of this study, might have been susceptible to anterior trunk tilt. Konno et al. [15] reported that as spinal alignment changed from lordosis to kyphosis, the intramuscular pressure increased and blood flow decreased; as a result, patients with a kyphotic spinal alignment had low back pain. One reason for the high prevalence of low back pain in the patients with knee OA might be the increased anterior trunk tilt during gait shown in this study.

The most important findings in this study were that the maximum and minimum anterior trunk tilts decreased 1 year postoperatively compared to preoperative stages, and the trunk movement of the patients with knee OA after TKA was approximately equal to that of the controls. Before TKA, the limited knee joint extension caused a posterior shift in the centre of gravity; thus, an induced anterior trunk tilt might have been a compensatory mechanism. However, after TKA, the extension ROM improved in the operated knees but not the non-operated knees. These results suggested that improvements in functions such as ROM observed in only the unilateral knee joint after TKA created the changes in trunk kinematics during gait.

A different study, however, showed that there were no significant differences in the maximal trunk angle in the sagittal, coronal, and horizontal plane during gait between before and after TKA in the knee OA patients [16]. The postoperative measurements were performed at 4 weeks in this previous study, whereas they were performed at 1 year in the present study. These results might suggest that trunk movement during gait of the knee OA patients improved over a long period postoperatively.

On the operated sides, the knee range of motion during one gait cycle was larger after TKA than preoperative values. Several studies [17, 18, 19] reported that the knee joint kinematics and kinetics in the sagittal plane following TKA had not changed despite pain and function improvements. The result of the previous study [19], which showed that preoperative knee range of motion during gait was 46.7 (12.4) ${}^{\circ}$ , was larger than that of our study (33.6 [12.3] ${}^{\circ}$ ). In the present study, our patients with knee OA had more severe deformities or symptoms before surgery than those in previous reports. Therefore, knee kinematics might be easily changed.

In contrast, hip extension during the stance phase, anterior pelvic tilt during gait, and gait parameters such as velocity and stride length had not changed 1 year postoperatively in our OA group. These hip joint and pelvic results were similar to those of previous studies [8, 19]. In older people, the lack of hip extension during walking was associated with greater energy cost of walking, as reflected by poorer functional self-reports of older adults with mobility disability [20]. To increase normal gait patterns in patients with knee OA, it might be necessary to increase hip extension during gait.

This study has several limitations. First, the number of study participants was limited, which may lead to type II statistical errors. However, in this study, a power showed $>$ 0.80 for all the significant parameters from a post-hoc power analysis. Second, the use of a self-selected velocity is a potential confounding variable since participants in the control group walked faster than those in the knee OA group both pre and postoperatively. One previous study reported that the range of motion of the lumbar region and pelvis during gait were larger at fast speeds than at preferred speeds [21]. However, no significant change in gait velocity was observed in the OA group 1 year postoperatively; thus, velocity was unlikely to be a confounding variable in this study. Third, there were no kinematic data of the ankle joint in the present study. Chang et al. [16] reported that at the 4th week post-TKA, ankle range of motion during gait on the non-operated side increased. Changes in the kinematics of the knee joint might have influenced the kinematics of the ankle joint. Third, because the control participants had full range of motion and no deformity in their knee joint and no pain during gait, we did not confirm the presence of the radiological knee OA in them. Considering ages of the control participants, they might have a slight degeneration in their knee joints. However, since they almost had a normal function of their knee, we adopted them for the control group. Finally, we did not assess low back pain in the patients with knee OA. Thus, the correlation between low back pain and spinal or pelvic alignment should be investigated in future studies.

5. Conclusions

Patients with knee OA walked with greater anterior tilts of the trunk and pelvis before surgery compared to controls. Furthermore, as clinical symptoms such as unilateral knee joint ROM and pain improved, the anterior trunk tilt during gait decreased 1 year postoperatively, and the trunk movement of the patients with knee OA after TKA was approximately equal to that of the healthy subjects. Future studies with regard to the correlation between low back pain and spinal alignment may enhance our understanding of the high prevalence of low back pain in patients with knee OA.

Footnotes

Acknowledgments

The authors thank Professor Hiroaki Kimura, Department of Rehabilitation, Hiroshima University Hospital for his assistance in the preparation of this manuscript.

Conflict of interest

The authors declare no conflicts of interest associated with this study.

References

Yoshimura

Muraki

Oka

Mabuchi

En-Yo

Yoshida

, et al. Prevalence of knee osteoarthritis, lumbar spondylosis, and osteoporosis in Japanese men and women: the research on osteoarthritis/osteoporosis against disability study. J. Bone Miner. Metab. 2009; 27: 620-628.

Wolfe

Hawley

Peloso

Wilson

Anderson

. Back pain in osteoarthritis of the knee. Arthritis Care Res. 1996; 9: 376-383.

Suri

Morgenroth

Kwoh

Bean

Kalichman

Hunter

. Low back pain and other musculoskeletal pain comorbidities in individuals with symptomatic osteoarthritis of the knee: data from the osteoarthritis initiative. Arthritis Care Res. 2010; 62: 1715-1723.

Tsuji

Matsuyama

Goto

Yimin

Sato

Hasegawa

, et al. Knee-spine syndrome: correlation between sacral inclination and patellofemoral joint pain. J. Orthop. Sci. 2002; 7: 519-523.

Murata

Takahashi

Yamagata

Hanaoka

Moriya

. The knee-spine syndrome. Association between lumbar lordosis and extension of the knee. J. Bone Joint Surg. Br. 2003; 85: 95-99.

Lee

Yoon

Moon

Lee

Seo

. Effect of correction of the contractured flexed osteoarthritic knee on the sagittal alignment by total replacement. Asian Spine J. 2013; 7: 204-211.

Harato

Nagura

Matsumoto

Otani

Toyama

Suda

. A gait analysis of simulated knee flexion contracture to elucidate knee-spine syndrome. Gait Posture. 2008; 28: 687-692.

Kim

Dorj

Han

Kim

Nha

. Improvements in spinal alignment after high tibial osteotomy in patients with medial compartment knee osteoarthritis. Gait Posture. 2016; 48: 131-136.

Kellgren

Lawrence

. Radiological assessment of osteo-arthrosis. Ann. Rheum. Dis. 1957; 16: 494-502.

10.

Koshino

. Japanese Orthopaedic Association knee scoring system. J. Jpn. Orthop. Assoc. 1988; 62: 900-902 (in Japanese).

11.

Hartrick

Kovan

Shapiro

. The numeric rating scale for clinical pain measurement: a ratio measure? Pain Pract. 2003; 3: 310-316.

12.

Schwartz

Rozumalski

Trost

. The effect of walking speed on the gait of typically developing children. J. Biomech. 2008; 41: 1639-1650.

13.

Thummerer

von Kries

Marton

Beyerlein

. Is age or speed the predominant factor in the development of trunk movement in normally developing children? Gait Posture. 2012; 35: 23-28.

14.

Gombatto

Brock

DeLork

Jones

Madden

Rinere

. Lumbar spine kinematics during walking in people with and people without low back pain. Gait Posture. 2015; 42: 539-544.

15.

Konno

Kikuchi

Nagaosa

. The relationship between intramuscular pressure of the paraspinal muscles and low back pain. Spine. 1994; 19(19): 2186-2189.

16.

Chang

Sohmiya

Wada

Tazawa

Sato

Yanagisawa

, et al. Alternation of trunk movement after arthroplasty in patients with osteoarthritis of the knee. J. Orthop. Sci. 2011; 16: 382-388.

17.

McClelland

Webster

Feller

. Gait analysis of patients following total knee replacement: a systematic review. Knee. 2007; 14: 253-263.

18.

Milner

. Is gait normal after total knee arthroplasty? Systematic review of the literature. J. Orthop. Sci. 2009; 14: 114-120.

19.

Levinger

Menz

Morrow

Feller

Bartlett

Bergman

. Lower limb biomechanics in individuals with knee osteoarthritis before and after total knee arthroplasty surgery. J. Arthroplasty. 2013; 28: 994-999.

20.

Wert

Brach

Perera

Van Swearingen

. Gait biomechanics, spatial and temporal characteristics, and the energy cost of walking in older adults with impaired mobility. Phys. Ther. 2010; 90: 977-985.

21.

Crosbie

de Faria Negrão

Nascimento

Ferreira

. Coordination of spinal motion in the transverse and frontal planes during walking in people with and without recurrent low back pain. Spine. 2013; 38: E286-292.