1 Introduction

Gait analysis is generally used for identifying issues related to musculoskeletal disorders, neurological conditions, or orthopedic injuries and for assessing the effectiveness of surgical interventions on gait and mobility. It is usually performed in specialized gait analysis laboratories. We set out to develop a comparatively simple method that can be performed in any clinical setting without a gait analysis laboratory and even in a normal doctor's office. To demonstrate the value of this novel tool, we analysed the gait patterns of patients having received a total hip arthroplasty and compared them with healthy individuals.

Primary total hip arthroplasty is a surgical procedure which is indicated for patients with severe hip osteoarthritis. Although arthroplasty has been shown to be effective in relieving pain and in restoring function,^1–5 patients often experience gait limitations following appropriate surgery.⁶ Thus, the spatial and temporal parameters of gait patterns are of clinical relevance for the assessment of motor pathologies, especially in orthopedics.

Here we present a novel gait analysis system, called Computed MyoGraphy (CMG), that not only assesses the kinematics of a movement but also the forces acting within the musculo-skeletal system, e.g. joint forces. In contrast to the traditional gait analysis performed in gait laboratories, CMG can also be used for daily clinical practice. The CMG system is based on state-of-the-art techniques by means of a special measuring suit (Xsens Awinda with MVN Record) and on a three-dimensional computer model (Myonardo, Predimo GmbH, Münster, Germany). It combines conventional gait analysis with a precise and temporally highly resolved analysis of the associated muscle activity and joint loads. Thus, in contrast to previous tools which only focus at the description and analysis of the movement patterns,^7,8 CMG opens a new venue for future studies, such as comparing the effectiveness of different surgical procedures, to significantly improve prosthesis designs and rehabilitation protocols in the future.

We have used CMG to analyse the joint forces at the hip after implantation of a primary total hip arthroplasty, comparing the results obtained before, after the surgery and at the end of the rehabilitation process and with healthy individuals.

Our hypotheses are the following: The total and shear joint forces are reduced for patients before and after the surgery and may recover during rehabilitation. Furthermore, we assume that the forces in the affected joint are lower compared to the unaffected body side of the patients.

2 Methods

In order to generate a tool for objective gait analysis including underlying muscle activity and joint loads, we obtained kinematic data, which were then analysed using Computed Myography (CMG) based on a 3D musculoskeletal computer model Myonardo (Predimo GmbH, Münster, Germany) to calculate total and shear joint contact forces.

2.1

2.1 Kinematic assessment

Participants were given a suit equipped with inertial measurement units (IMUs). The corresponding hardware, Xsens MVN Awinda (Movella Inc., 3535 Executive Terminal Drive, Suite 110, Henderson, NV 89052), consisted of 17 wireless inertial measurement units (IMUs), which were attached to predetermined locations on relevant body segments by Velcro straps (Fig. 1). The lengths of the individual segments were measured for each participant prior to recording. Each IMU contains a gyroscope, accelerometer, barometer, and magnetometer whose data is transduced into radio signals.⁹ All 17 IMU's signals were received by a USB antenna plugged into a computer running the MVN software.

Close

Fig. 1 MVN system and Xsens suit with MVN motion system. MVN system as an overview display on the computer showing the individual sensors (connectivity status, green dots) as well as (exemplary) segment length. The representation of the data is three-dimensional with the vectors x, y and z (A1). Test person with x-sense motion capture system-motion trackers in red, (A2) in front view and (A3) in back view.

Export to PPT

Kinematic data was acquired using the software MVN Record 2021.2 (Xsens Technologies B.V., 2021) with the following settings: (1) Suit Configuration: Full body; (2) Accept System: Awinda; (3) Scenario: Single level (4) Maximum Update Rate: 60 Hz. MVN Analyse was used as a 64-bit application for Windows 10. Each recorded file was HD-processed, i.e. the data was analysed over a longer period, to increase the accuracy and consistency of the estimated position and orientation by eliminating artifacts such as skin movement, for each body segment.¹⁰ Then each file was trimmed to contain only the relevant movements.

2.2

2.2 Experimental procedure

Patients were asked to perform the following movements while wearing the suit (Fig. 2):(a)Walk back and forth along the station corridor (6–8 m)(b)Squat down and up three time to a maximum of 90° knee-flexion, if possible.

Close

Fig. 2 Movement patterns of Myonardo. Different movement patterns as a computerized three-dimensional human model, called Myonardo. Representation with bones and muscles during the execution of squats (A1) and walking (A2). In the visualization the muscles are clearly visible when they are loaded in comparison to unloaded muscles, which appear translucent. According to the level of loading of the joints the color of those joints appears green, yellow or red.

Export to PPT

2.5

2.5 Patient collective for study

This prospective study included patients with primary hip osteoarthritis, aged 18–80, treated with primary total hip arthroplasty between 06/2022 and 06/2023 at a university clinic (n = 18; mean age: 62.94 ± 8.61). See Fig. 3 for a flow chart of the recruiting process. Excluded were patients with additional or different joint prostheses, secondary interventions, gait-altering pathologies, or malpositioned fractures. The present study was approved by a university clinic's ethics committee (approval number: AN 15/3/22) in compliance with the Helsinki Declaration. The average body mass index (BMI) of the healthy probands was 24.17 ± 4.82 kg/m² and 29.10 ± 5.38 kg/m² of the osteoarthritis patients, respectively. The severity of osteoarthritis was graded on the basis of preoperative anterior-posterior (a.p.) and axial native radiological imaging using the Kellgren-Lawrence classification¹⁵ (Table 1). Implantation of the hip arthroplasty was always performed by the same, experienced surgeon via a modified minimally invasive surgical approach according to Watson and Jones. The participants should not have any further surgery and other joint prosthesis implanted. Exclusion criteria were defined as implantations in other joints (knee prosthesis, already existing other joint prosthesis), secondary interventions with an already inserted hip prosthesis after, for example, prosthesis infection or secondary loosening, patients with pathologies that change the physiological gait pattern and patients with previously known fractures that have healed in a malposition state. Healthy individuals (n = 20) were selected on the basis of a physiological gait pattern (age between 18 and 80 years; mean age: 55.89 ± 17.69; 11 females and 9 males).

Close

Fig. 3 Flowchart about the study overview. After applying the exclusion criteria, 21 patients were integrated into the final examination.

Export to PPT

Table 1 Baseline characteristics of the participants' collective.

Description	Control	Patients with osteoarthritis
Number of all patients	20	18
Age range [years]	23–83	52–80
Age mean [years] ± SD	55.89 ± 17.69	62.94 ± 8.61
Body weight [kg] ± SD	72.03 ± 14.63	83.26 ± 14.45
BMI [kg/m2] ± SD	23.77 ± 4.25	28.44 ± 4.50
Gender
Female [n]	11	10
Male [n]	9	8
Number of included patients	10	11
Age range [years]	27–83	50–69
Age mean [years] ± SD	60.40 ± 19.22	58.82 ± 5.95
Body weight [kg] ± SD	72.14 ± 12.14	84.21 ± 15.48
BMI [kg/m2] ± SD	24.17 ± 4.82	29.10 ± 5.38
Gender
Female [n]	6	7
Male [n]	4	4
Kellgren-Lawrence Classification		Osteoarthritis hip side
Patients [#]		right	left
1		–	2
2		2	–
3		–	2
4		3	–
5		4	–
6		–	2
7		–	3
8		4	–
9		4	–
10		2	–
11		2	–
Mean ± SD	Total: 2.73 ± 0.90	3.00 ± 1.00	2.25 ± 0.50
Mean female [n = 7] ± SD		3.20 ± 1.10	2.50 ± 0.71
Mean male [n = 4] ± SD		2.50 ± 0.71	2.00 ± 0.00

Eleven patients (mean age: 58.82 ± 5.95; seven females and four males) were included in the study who were treated with primary total hip arthroplasty and 10 healthy reference control individuals (mean age: 60.40 ± 19.22; 6 females and 4 males).

The movement patterns of the patients before and after the implantation of a primary total hip arthroplasty were determined at three different time points, i.e. preoperatively (t1), at the fifth postoperative day (t2) and after the completion of a rehabilitation approximately four months after the operation (t3). The gait pattern of the healthy control individuals was measured once.

Data from 17 individuals (7 patients with primary hip osteoarthritis and 10 in the control group) were not evaluated due to a dropout of patients (n = 1) and due to operating errors such as misapplication of sensors (n = 2), incorrectly stored data (n = 3), and mismatch of sensors (n = 11) (Fig. 3). It should be noted that these errors did not occur as part of the CMG evaluation but as part of the measurement preparation.

3 Results

3.1

3.1 Total and shear hip joint contact forces during squatting

In patients, total joint contact forces (JCFs) on the ipsilateral side were significantly lower than those of healthy controls during both the pre- and post-operative phases, while there was no significant difference between these two phases after the rehabilitation of the patients and the healthy control subjects. On the contralateral side, there only was a significant decrease of JCFs between healthy controls and patients in the post-operative phase. For both sides, after rehabilitation the JCFs increased significantly to a level that is no longer significantly different from that of healthy controls (Fig. 4a, Table 2). A similar situation occurs for the shear JCFs, except that there is an additional significant difference between the pre- and the post-operative phase (Fig. 4b–Table 2).

Close

Fig. 4 Violin Plots of the total and shear contact forces (JCF) of the hip joint on the ipsi- and contralateral side during squatting. Representation of the data sets of healthy and surgically treated individuals at different timepoints. Healthy patients collective in blue, the osteoarthritis (= affected) collective preoperative (preOP; t1) in red, postoperatively (postOp; t2) in yellow and after completion of rehabilitation (postReha; t3) in violet. The force in multiples of the body weight [BW] are indicated on the y-axis, the patient's collective is presented on the x-axis. Small white circles represent estimated marginal means (EMMs), error bars correspond to their respective 95 % confidence intervals.

Export to PPT

Table 2 Estimated marginal means (EMMs) and their corresponding 95 % confidence intervals, as well as the p-values of pairwise comparisons, for the squatting task. In the table, ∗ indicates a significance level of p < 0.05, ∗∗ indicates p < 0.01, ∗∗∗ indicates p < 0.001. (PreOP is pre-operative, PostOP is post-operative, PostReha is post rehabilitation, BW is bodyweight, CI is confidence interval, IPSI is ipsilateral, CONTRA is contralateral, JCF is joint contract force).

SQUATTING: total JCF
IPSI	Healthy	PreOP	PostOP	PostReha
BW (CI)	2.90 (2.33, 3.84)	1.63 (1.30, 2.17)	1.36 (1.13, 1.70)	2.10 (1.67, 2.82)
Healthy	–	0.018∗	0∗∗∗	0.215
PreOP		–	0.197	0.004∗∗
PostOP			–	0∗∗∗
PostReha				–
CONTRA	Healthy	PreOP	PostOP	PostReha
BW (CI)	3.06 (2.45, 4.06)	2.13 (1.68, 2.90)	1.71 (1.39, 2.24)	2.55 (2.14, 3.16)
Healthy	–	0.24	0.014∗	0.248
PreOP		–	0.175	0.348
PostOP			–	0.011∗
PostReha				–
SQUATTING: shear JCF
IPSI	Healthy	PreOP	PostOP	PostReha
BW (CI)	2.37 (1.37, 8.45)	0.57 (0.48, 0.69)	0.29 (0.25, 0.35)	1.14 (0.81, 1.92)
Healthy	–	0∗∗∗	0∗∗∗	0.105
PreOP		–	0∗∗∗	0∗∗∗
PostOP			–	0∗∗∗
PostReha				–
CONTRA	Healthy	PreOP	PostOP	PostReha
BW (CI)	2.64 (1.59, 7.91)	0.68 (0.53, 0.96)	0.27 (0.23, 0.33)	1.3 (0.93, 2.16)
Healthy	–	0∗∗∗	0∗∗∗	0.053
PreOP		–	0∗∗∗	0∗∗∗
PostOP			–	0∗∗∗
PostReha				–

3.2

3.2 Total and shear hip joint contact forces during walking

During walking, a significant reduction in total JCFs was observed on both the ipsi- and contralateral side from healthy controls as compared to patients both before (t1) and after surgery (t2) (Fig. 5, Table 3). Rehabilitation significantly (t3) increased JCFs from post-operative to post-rehabilitation for both sides to the point where they were no longer significantly different from those of healthy controls. For shear forces, both the ipsi- and contralateral sides showed a significant decrease between healthy subjects and patients in the post-operative phase. Rehabilitation again led to a significant increase in shear forces from post-operative (t2) to post-rehabilitation (t3), where no longer a significant difference to healthy controls could be observed (Fig. 5, Table 3).

Close

Fig. 5 Violin Plots of the total and shear contact forces (JCF) of the hip joint on the ipsi- and contralateral side during walking. Representation of the data sets of healthy and surgically treated individuals at different timepoints. Healthy patients collective in blue, the osteoarthritis (= affected) collective preoperative (preOP; t1) in red, postoperatively (postOp; t2) in yellow and after completion of rehabilitation (postReha; t3) in violet. The force in multiples of the body weight [BW] are indicated on the y-axis, the patient's collective is presented on the x-axis. Small white circles represent estimated marginal means (EMMs), error bars correspond to their respective 95 % confidence intervals.

Export to PPT

Table 3 Estimated marginal means (EMMs) and their corresponding 95 % confidence intervals, as well as the p-values of pairwise comparisons, for the walking task. In the table, ∗ indicates a significance level of p < 0.05, ∗∗ indicates p < 0.01, ∗∗∗ indicates p < 0.001. (PreOP is pre-operative, PostOP is post-operative, PostReha is post rehabilitation, BW is bodyweight, CI is confidence interval, IPSI is ipsilateral, CONTRA is contralateral, JCF is joint contract force).

WALKING: total JCF
IPSI	Healthy	PreOP	PostOP	PostReha
BW (CI)	5.64 (5.27, 6.07)	4.6 (4.28, 4.97)	3.89 (3.27, 4.79)	4.91 (4.33, 5.68)
Healthy	–	0.001∗∗	0.014∗	0.267
PreOP		–	0.31	0.426
PostOP			–	0∗∗∗
PostReha				–
CONTRA	Healthy	PreOP	PostOP	PostReha
BW (CI)	5.63 (5.06, 6.34)	4.61 (4.25, 5.04)	3.63 (3.18, 4.23)	4.76 (4.34, 5.26)
Healthy	–	0.032∗	0∗∗∗	0.117
PreOP		–	0.054	0.342
PostOP			–	0.009∗∗
PostReha				–
WALKING: shear JCF
IPSI	Healthy	PreOP	PostOP	PostReha
BW (CI)	1.41 (1.22, 1.69)	1.19 (1.00, 1.47)	0.86 (0.75, 1.00)	1.19 (1.08, 1.31)
Healthy	–	0.366	0∗∗∗	0.186
PreOP		–	0.056	0.99
PostOP			–	0∗∗∗
PostReha				–
CONTRA	Healthy	PreOP	PostOP	PostReha
BW (CI)	1.44 (1.21, 1.78)	1.00 (0.90, 1.12)	0.77 (0.64, 0.96)	1.13 (1.01, 1.29)
Healthy	–	0.004∗∗	0∗∗∗	0.153
PreOP		–	0.088	0.13
PostOP			–	0∗∗∗
PostReha				–

4 Discussion

Traditional gait analysis methods are often subjective and lack detailed insight into joint forces and the individual load profiles. CMG addresses these limitations by allowing detailed gait analysis in routine clinical practice without requiring specialized laboratory facilities. CMG also enables a functional insight and an objective analysis not only of the gait patterns but addresses also the joint forces acting under individual load profiles as shown here. It therefore seems to be adaptable to possible medical areas and to possibly optimize rehabilitation measures after a hip arthroplasty surgery. Traditional gait analysis methods are often subjective and lack detailed insight into joint contact forces. CMG addresses these limitations by allowing detailed gait analysis in routine clinical practice without requiring specialized laboratory facilities.

Five days after the surgery, there was a significant difference between healthy individuals and hip arthroplasty patients with respect to performing squats and walking (see Figs. 4 and 5). For the patients, in both activities, force distribution was significantly lower on both the contralateral and the ipsilateral sides. There were also significant reductions observed in the walk total analysis in both “ipsi” and “contra” sides of patients as compared to healthy individuals at the preoperative postoperative stage. This finding indicates a decline in load-bearing capacity at the pre- and post-surgery stage. The difference can be explained by postoperative pain symptoms and a compensating response of the patients which causes a relief with respect to the affected side. The largest differences between patients and healthy subjects can be seen with the shear forces (see Figs. 4 and 5). A plausible explanation is that shear forces cause particular pain, leading patients to adapt their movements to minimize discomfort.

The data indicate that patients with primary hip arthroplasty have significantly lower hip joint contact forces on both sides of the body, with the effect being pronounced during both squatting and walking (Figs. 4 and 5, Tables 2 and 3). Initially, patients exhibited a minimal loss of strength on the ipsilateral side during squatting postoperatively (t2), while the strength on the contralateral side increased slightly likely due to the compensatory mechanism. Upon completing rehabilitation, joint strength increased substantial on both bodysides, nearing the level of healthy individuals. In addition, all measurements for squatting showed better values when the preoperative status of the patients was compared to post rehabilitation stage. With the exception of squat hip total JCF, all values were statistically significant.

Five days postoperatively (t2), patients exhibited slight strength loss on both bodysides while walking likely due to pain response. After completion of the rehabilitation measures (t3), however, an increase in joint strength was found on both body sides. This increase, as observed with squatting, reached almost the level of healthy individuals. This observation suggests that after rehabilitation, total hip arthroplasty patients nearly reach the functional level of healthy subjects and patients are therefore in better condition than before the surgery.

Our studies also revealed a time-dependent increase in joint forces on both sides from five days postoperatively (t2) to post-rehabilitation (t3). A possible explanation for the initially less strain on the hips is that the movement and loading of the affected hip causes pain. Patients may reduce the underlying painful joint contact forces by reducing muscle activation and/or movement velocity, with the latter being causally related to the former.

Mendiolagoitia et al.¹⁶ already reported in an earlier study that the gait pattern after total hip arthroplasty improves compared to the preoperative state, although there are still minimal deficits compared to healthy individuals. Their findings are consistent with the results described here demonstrating that osteoarthritis patients undergoing total hip arthroplasty show significant improvements with respect to pain reduction and function.¹⁷ In fact, there is ample evidence that total hip arthroplasty is indeed effective in reducing pain and improving function^16,17 and early improvements in spatial-temporal and kinematic gait patterns were observed compared to preoperative values.¹⁸ After 12 months, only minor deficits were still noted with such patients when compared to healthy individuals.¹⁸

Although this aspect was not addressed in our study, it is interesting to note that also obese patients showed a significant functional improvement after total hip arthroplasty implantation. This aspect was previously assessed on the basis of their gait speed.¹⁹ Even if the patients did not fully reach the level of a healthy control person, the functional gain was comparable regardless of the body mass index.¹⁷ Mak et al.²⁰ noted a psoas muscle atrophy on the implant side compared to the non-operative side of post-unilateral implant patients. CMG could therefore support the postoperative rehabilitation process through targeted exercises which strengthen the psoas muscle to improve the overall functional outcome. Thus, patients could possibly benefit from a clinical use of this new analysis tool, an aspect that needs further investigation.

Peng et al.²¹ reported that female patients with unilateral hip arthroplasty had significantly increased adduction of the implanted hip compared to the native hip. This effect is partly due to a superior center of rotation of the femoral shaft. Understanding gender differences in kinematic patterns may benefit female patients through targeted preoperative planning and postoperative rehabilitation which was not addressed in our study. This issue requires future investigations to explore questions concerning muscular imbalances, which can now be examined by CMG in everyday clinical practice.

Notably, the results of Rosenlund et al.²² indicate that measures to improve hip muscle strength can moderately improve the general quality of gait in patients with primary hip osteoarthritis. Their study showed that the patient population exhibited better gait quality with better developed hip abductors and hip flexors.

In summary, patients with primary hip arthroplasty exerted less force on their hip joints after surgery than healthy subjects. The reduction of force on the affected side appears to be more pronounced during squatting than during walking, but there is an improvement towards normal during the rehabilitation process.

The initial reductions of joint contact forces before (t1) and five days after the surgery (t2) could be a result of pain in the affected hip of the patients, an effect which may significantly reduce their quality of life. Our study shows that hip arthroplasty results in an improvement in strength development that almost reaches the level of healthy subjects after the rehabilitation period (t3). A final important outcome of our study suggests the possibility to apply CMG as a tool for standard clinical practice. CMG could reduce or even replace the corresponding costly and time-consuming examinations in a gait laboratory. Furthermore, muscular imbalances can be detected already at an early stage after the surgery and thus be treated in time.

One essential limitation of our pilot study is the relatively small number of patients. However, our study is a cross-sectional analysis and therefore exploratory in nature. With a larger number of patients, differences among individuals, including existing pathologies, the type of surgery performed, and postoperative treatments, additional information may become apparent which may impact the outcome of the surgery.

5 Conclusion

Patients with hip osteoarthritis are limited in their walking and squatting performance compared to healthy subjects, most likely due to pain avoidance. These limitations are reflected in lower total and shear hip joint contact forces as measured by the new method of Computed Myography (CMG). The reduction in hip joint contact forces disappears as patients progress through surgery and subsequent rehabilitation to the level of healthy subjects. CMG has been shown to be a useful diagnostic tool in patients with hip osteoarthritis to identify limitations in squatting and walking after total hip arthroplasty. It provides an objective tool to assist the documentation of recovery after surgery and subsequent rehabilitation. In the future, CMG may offer a way to compare surgical procedures, improve prosthesis designs, and enhance rehabilitation protocols. This new method could replace complex gait lab analyses and support data-driven, individualized therapy.

CRediT authorship contribution statement

Katharina Jäckle: Conceptualization, Methodology, data recording, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, preparation, Writing – review & editing, Visualization, Project administration. Tim Alexander Walde: data recording. Marc-Pascal Meier: Writing – review & editing. Thelonius Hawellek: Writing – review & editing. Paul-Jonathan Roch: Writing – review & editing. Colja Homann: Validation, Formal analysis, Supervision. William H.M. Castro: Writing – review & editing, Supervision. Heiko Wagner: Formal analysis, Writing – review & editing, statistics, Visualization. Kim Boström: Formal analysis, Writing – review & editing, statistics, Visualization. Wolfgang Lehmann: Resources, Writing – review & editing, Supervision. Lukas Weiser: Validation, Writing – review & editing, Supervision, Project administration, All authors have read and agreed to the published version of the manuscript.

Ethical approval

“All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.”

Ethics approval and consent to participate

The present study was approved by the ethics committee of the University Medical Center Göttingen (approval number: AN 15/3/22). “Additional informed consent was obtained from all individual participants for whom identifying information is included in this article.” Written informed consent will be obtained from all participants.

Availability of data and material

“The datasets supporting the conclusions of this article are included within the article (and its additional files).”

Funding

This research received no external funding. Open access funding enabled and organized by the Open Access Publication Funds of the Göttingen University. This study was supported by the AO Nachwuchsförderungspreis (AO Young Talent Award) 2022.