Analysis of Spatiotemporal Gait Variables before and after Unilateral Total Knee Arthroplasty

Abstract

This study aimed (a) to evaluate the spatiotemporal gait variables of total knee arthroplasty (TKA) before (pre-) and after the procedure (post-), and (b) to investigate the influence of the surgical side on these variables. Twenty-one volunteers (13 females and 8 males) participated, undergoing assessments pre-surgery and nine to 12 months post-surgery. Clinical tests indicated significant reductions in knee pain and improvements in active and passive extension post-surgery. TKA resulted in decreased pain, extension deficits, and functional assessments, with lower scores on the WOMAC questionnaire. A gait analysis showed post-surgery improvements in gait speed (5.8%), cycle time (−4.8%), step time (4.4%), double limb support time (−11.1%), step (4.4%) and stride (6.3%) lengths, and step (5.1%) and stride (5.0%) cadences. Comparisons between surgical side and limb dominance indicated significant differences in gait speed, stance, swing and step times, double limb support time, step and stride lengths, and step cadence. The non-dominant limb demonstrated greater improvements across most parameters compared to the dominant limb. These findings emphasize the importance of considering the limb dominance of patients with knee osteoarthritis when evaluating post-TKA function. These conclusions can be helpful for personalized rehabilitation programs, allowing tailored interventions for individuals undergoing knee surgery.

1. Introduction

Knee osteoarthrosis (KOA) is the degenerative disease that causes the greatest loss of mobility in the elderly [1]. In Portugal, the prevalence is about 15.8% in females and 8.6% in males [2]. Patients with KOA have limited mobility, reflected as a reduced range of motion and functional impairment [3]. The risk of developing knee osteoarthritis involves a complex interaction between intrinsic factors, such as bone density and morphology, sex hormones, and trauma, and mechanical factors, such as joint overload. However, the greatest risk factors are age and obesity, the increasing rates of which in the populations of developed countries make the increase in knee osteoarthritis an expected phenomena [4].

TKA is the key surgical treatment for severe KOA because it effectively reduces pain and functional impairment, improves quality of life, and enables autonomy in daily activities [5,6]. Despite reports of gait improvement after TKA, it also has been suggested that those are not sufficient to accomplish values comparable to healthy age-matched subjects [1,6,7,8,9,10], raising the need to better understand the nature and extent of these gait improvements [9,11,12].

However, after unilateral TKA, altered gait mechanics, including increased joint loading and reduced knee flexion excursion, may contribute to the progression of contralateral KOA. After TKA, individuals often favor the contralateral limb during weight-bearing tasks, potentially leading to increased forces and adduction moments at the contralateral knee [10,13,14]. This increased dependence on the contralateral knee may be a contributing factor to the radiographic progression of KOA in that knee. Further exploration is warranted to fully understand these dynamics and develop strategies to mitigate the risk of contralateral KOA progression [1,10].

Gait analysis has become an important tool for evaluating KOA and TKA patients, as it provides objective measures of kinematic and kinetic parameters that relate to the patient’s functional capacity [9]. Spatiotemporal parameters, such as gait velocity and cycle length, and the duration of its phases, are easily obtained and provide valuable insights into gait patterns. While spatiotemporal and kinematic gait parameters are generally believed to improve in the medium to long-term following TKA [6,8,15,16,17], the incomplete recovery of normal knee joint function and impaired gait pattern are often reported when compared to asymptomatic controls [1,7,9,10,14,18,19]. Specifically, studies have shown that KOA patients exhibit slower walking velocity, shortened step length, and lower cadence than healthy individuals [3,6,15,18,20]. However, improvements in stride time and length were reported after TKA [21]. Nevertheless, knowledge gaps and limitations in post-TKA gait assessment research have been previously identified, and few published works have studied the effect of TKA on the gait parameters of the non-surgical limb [17].

TKA outcomes are influenced by multiple factors [10], making essential a comprehensive post-TKA assessment that considers both general improvements and biomechanical adaptations. While evidence regarding the effect of TKA in the surgical and non-surgical limb is available [1,22], the effect of limb dominance is a less explored aspect that deserves analysis, as it can guide rehabilitation strategies and improve postoperative care. Thus, the aims of this study were (a) to evaluate the spatiotemporal gait variables of patients before (pre-) and after (post-) TKA surgery, and (b) to investigate the influence of the surgical side dominance on these variables.

2. Materials and Methods

2.1. Experimental Approach

This is a prospective study designed to compare the spatiotemporal variables of gait in TKA patients. The participants underwent two assessments, one prior to surgery (19 ± 14 days) (pre-) and another 9 to 12 months after surgery (post-). In both assessments, a set of sociodemographic, clinical, and functional parameters were recorded for further sample characterization, followed by a gait analysis. All participants were evaluated pre- and post-surgery, and subsequently assessed based on limb dominance, with both the dominant and non-dominant operated limbs being analyzed.

2.2. Participants

Twenty-one volunteers (13 females and 8 males) participated in the study. The volunteers were all right limb dominant, with 66.05 ± 6.34 years, 1.60 ± 0.64 m, 80.28 ± 12.55 kg, 31.33 ± 4.89 kg/m² body mass indexes, presenting a Kellgren–Lawrence ipsilateral score of 2.76 ± 0.70 and a contralateral score of 2.33 ± 0.65. The sample comprised participants who were operated on their right (n = 10) and left (n = 11) lower limb.

Adults scheduled for primary TKA from November 2020 to May 2021 at one of the local University Hospitals were invited to participate in the study. All the patients who met the inclusion criteria were contacted by telephone and have provided verbal consent for the use of demographic and clinical data. Exclusion criteria were defined as having previous lower limb orthopedic surgery (except knee arthroscopy or foot surgeries), contralateral total knee arthroplasty, other indications for total knee arthroplasty rather than knee osteoarthritis, vestibular disease requiring medication, severe visual impairment, and/or neurological diseases that may cause gait impairment.

A researcher was responsible for listing the patients who were eligible for study participation. Exclusion criteria were applied and data regarding gender, age, surgery laterality, and KOA radiological severity for both knees (evaluated by a named researcher with practice in KOA radiology using the Kellgren–Lawrence scale) were collected from the electronic clinical records. A first in-person evaluation before TKA was scheduled with the remaining patients. The patients who were unable to attend were excluded. The same group of patients were contacted to schedule the second in-person evaluation 9 to 12 months after TKA. Exclusion criteria were applied again and patients who were unable to attend the evaluation or who had postoperative complications (prosthetic infection, periprosthetic fracture, or revision surgery for any reason) were excluded.

The sample was evaluated in outpatient consultations by the same researcher. At first evaluation, each patient received a written informed consent document about the objectives of the study, which had to be signed to confirm participation. The study was conducted in accordance with the Declaration of Helsinki and approved by the Centro Hospitalar S. João/FMUP Ethics Committee (approval number 452/2020).

2.3. Data Collection Procedures

The demographic parameters included age, body mass (Inbody 230, InBody Co., Ltd., Seoul, Korea), height (Seca 206, Seca GmbH, Hamburg, Germany), lower limb dominance (by handedness laterality), the number of falls in the last month, the laterality of surgery, pain in the last week in both knees as assessed by the visual analogue scale, and the radiological severity of knee arthroplasty for both knees (assessed by a researcher with practice in knee arthroplasty radiology) using the Kellgren–Lawrence scale. The functional parameters included the measurement of the active and passive range of motion (ROM) with a handheld goniometer (Enraf-Nonius, Rotterdam, The Netherlands), and knee extension force, measured with a MicroFET2 (Hogan Health Industries, Draper, UT, USA) handheld dynamometer. These tests were performed for both knees by an experienced clinician. Additionally, the functional section of the Knee Society Score (2011KSS) [23,24,25] and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) [26,27,28], were used for the characterization of the function and quality of life, respectively. For a more detailed description of the clinical assessment, see Almeida e Reis et al. [29].

After the aforementioned evaluations, all participants underwent a gait analysis, after a brief familiarization period. A 12-camera motion capture system, composed of Oqus cameras (Qualisys AB, Göteborg, Sweden) operating at 200 Hz sampling frequency, was used for the kinematic gait analysis. The camera system was calibrated with an error below 0.6 mm on the central 6 m of a 10 m walkway. Retroreflective markers, with a 12 mm diameter, were placed on the participants according to the IOR lower body marker set [30], as well as 4-marker clusters on the thigh and shank of both lower limbs. Participants were instructed to walk at a comfortable, self-selected speed along the 10 m walkway, with at least ten valid attempts being recorded in each assessment session [31,32]. Kinematic data was recorded using Qualisys Track Manager 2022.2 build 7710 (Qualisys AB, Göteborg, Sweden).

2.4. Data Analysis Procedures

The Qualisys Clinical Gait plugin version 2.1.3+694 (Qualisys AB, Göteborg, Sweden) was used to build a 6 degrees of freedom (6DoF) model in Visual3D 2022.11.3 (C-Motion, Germantown, MD, USA) and perform the necessary calculations to obtain the gait linear kinematics for each lower limb, including: gait cycle, stance, swing, and step times, as well as step and stride length, and gait cadence expressed as steps and strides per minute. Time-dependent parameters were normalized to the gait cycle duration of the corresponding lower limb, while distances were normalized to the participants’ stature.

2.5. Statistical Analysis

Data normality was tested with the Shapiro–Wilk test. The clinical parameters and spatiotemporal gait variables were analyzed for pre- vs. post-surgery comparisons using the Wilcoxon test. To test the difference between pre- and post-surgery of the surgical limb, the differences were first calculated by subtracting the pre-surgery values from the post-surgery values, and then applying the Mann–Whitney test to these differences. The results are presented as the median and inter-quartile range (IQR). The effect size was calculated using G*Power 3.1.7 software (University of Düsseldorf Kiel, Germany), and expressed as Cohen’s d, with the corresponding interpretation criteria (small: > 0.2; moderate: > 0.50; large: > 0.80) [33]. All the statistical tests were performed using SPSS Statistics version 29 (IBM Corporation, Armonk, NY, USA) and α = 0.05.

3. Results

The results of the clinical tests performed pre- and post-surgery can be observed in

Table 1. Median and IQR of the comparison of clinical parameters pre- vs. post-surgery.

Clinical Parameter				Pre	Post	Z-Value	p-Value	dcohen
Number of falls in last month				0.0 (1.0)	0.0 (0.0)	−1.15	0.24	0.52
Knee Pain	Ipsilateral			6.0 (2.0)	2.0 (3.0)	−3.83	<0.001 *	3.05
	Contralateral			5.0 (4.0)	4.0 (5.0)	−0.06	0.95	0.02
Joint ROM	Ipsilateral	Extension	Active	9.0 (6.0)	4.0 (4.0)	−3.52	<0.001 *	2.41
			Passive	0.0 (6.0)	0.0 (1.0)	−2.10	0.03 *	1.03
		Flexion	Active	102.0 (18.0)	102.0 (10.0)	−1.17	0.87	0.52
			Passive	114.0 (15.0)	114.0 (12.0)	−1.58	0.79	0.73
	Contralateral	Extension	Active	4.0 (8.0)	4.0 (8.0)	−0.15	0.24	0.06
			Passive	0.0 (2.0)	0.0 (6.0)	−0.26	0.11	0.11
		Flexion	Active	112.0 (15.0)	114.0 (12.0)	−1.16	0.24	0.52
			Passive	118.0 (18.0)	124.0 (15.0)	−1.64	0.10	0.77
Dynamometry	Ipsilateral			208.2 (67.7)	221.5 (84.7)	−1.44	0.14	0.66
	Contralateral			221.1 (68.7)	243.3 (72.0)	−1.61	0.10	0.75
2011KSS				15.0 (12.0)	24.0 (14.0)	−3.61	<0.001 *	2.57
WOMAC				61.0 (21.0)	30.0 (23.0)	−3.98	<0.001 *	3.51

Note: An asterisk (*) indicates a significant difference (p < 0.05) between pre- and post-surgery. ROM: range of motion; 2011KSS: Knee Society Score; WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index.

. In short, after TKA, the patients experienced less pain, improved knee extension, better functionality according to the 2011KSS, and a better quality of life according to the WOMAC questionnaire. Despite these improvements, force production did not significantly increase.

As reported in

Table 2. Median and (IQR) of the spatiotemporal gait variables for surgical knee participants pre- and post-surgery.

Variables	Pre	Post	Z-Value	p-Value	dcohen
Gait Speed (m/s)	0.51 (0.16)	0.54 (0.15)	−5.95	<0.001 *	0.95
Cycle Time (s)	1.23 (0.22)	1.17 (0.21)	−5.93	<0.001 *	0.94
Stance Time (%GC)	62.97 (4.16)	63.45 (2.84)	−0.69	0.49	0.11
Swing Time (%GC)	37.02 (4.16)	36.54 (2.84)	−0.69	0.49	0.10
Step Time (%GC)	50.74 (1.73)	50.05 (1.44)	−6.08	<0.001 *	0.97
Double Limb Support Time (%GC)	14.93 (4.14)	13.27 (3.45)	−6.82	<0.001 *	1.13
Step Length (%S)	31.51 (7.05)	32.91 (5.37)	−5.56	<0.001 *	0.87
Stride Length (%S)	60.54 (12.34)	64.37 (11.02)	−5.03	<0.001 *	0.77
Cadence (Steps/Minute)	98.36 (17.07)	103.35 (17.82)	−7.23	<0.001 *	1.22
Cadence (Strides/Minute)	48.81 (9.15)	51.23 (9.68)	−5.83	<0.001 *	0.94

Note: An asterisk (*) indicates a significant difference (p < 0.05). %GC: percentage of the participant’s gait cycle duration; %S: percentage of the participant’s stature.

, the gait presents the significant differences after TKA surgery in terms of linear kinematics in all parameters (p < 0.001), with the exception of stance and swing time (p = 0.49). After TKA surgery, the gait speed increased by 5.8%, which was promoted by a shorter cycle time (−4.8%) and longer step (4.4%) and stride (6.3%) lengths. These changes enabled a higher gait cadence, with increases in step (5.1%) and strides (5.0%) per minute. Additionally, possibly due to a greater stability and confidence in gait, there were reductions in the double support time (−11.1%) and the step time (−1.4%).

These alterations to gait linear kinematic parameters have a significant impact in the ability to ambulate, as summarized in Figure 1. In this figure, a comparison between the initial position during a randomly selected gait trial pre-TKA (red) and post-TKA (white) is presented, and compared to the position reached by the participant after the same period of time.

Despite the changes observed post-TKA surgery, Table 2 does not differentiate the surgical outcome considering if the operated limb was the dominant or the non-dominant.

Table 3. Median and (IQR) of the difference between the spatiotemporal variables of gait pre- and post-surgery of the operated dominant and non-dominant limbs. Results in square brackets represent the relative percentual variation from pre-surgery.

Variables	Non-Dominant	Dominant	U-Value	p-Value	dcohen
Gait Speed (m/s)	0.08 (0.19) [18.6%]	0.03 (0.17) [6.4%]	3323	<0.001 *	0.49
Cycle Time (s)	−0.07 (0.16) [−3.83%]	−0.06 (0.17) [−4.84%]	4053	0.151	0.20
Stance Time (%GC)	−0.56 (4.22) [0.2%]	0.46 (3.13) [1.0%]	3584	0.008 *	0.39
Swing Time (%GC)	0.56 (4.22) [−0.4%]	0.51 (3.20) [-1.8%]	3584	0.008 *	0.39
Step Time (%GC)	−1.01 (2.67) [−2.1%]	−0.43 (1.68) [−0.8%]	3580	0.008 *	0.39
Double Limb Support Time (%GC)	−1.50 (4.09) [−12.7%]	−0.98 (3.54) [−8.3%]	3821	0.041 *	0.29
Step Length (%S)	2.96 (6.37) [7.8%]	1.57 (5.88) [4.8%]	3538	0.006 *	0.40
Stride Length (%S)	5.07 (13.12) [10.5%]	1.67 (12.20) [4.5%]	3086	<0.001 *	0.59
Cadence (Steps/Minute)	8.90 (15.13) [7.5%]	5.51 (15.31) [6.6%]	3701	0.019 *	0.34
Cadence (Strides/Minute)	3.42 (7.01) [3.9%]	2.83 (7.11) [5.0%]	3927	0.078	0.25

Note: An asterisk (*) indicates a significant difference (p < 0.05). %GC: percentage of the participant’s gait cycle duration; %S: percentage of the participant’s stature.

presents the differences between the post-surgery and pre-surgery linear kinematics, differentiating the sample according to the dominance of the operated limb. Results are generally in agreement with those found in Table 2; however, it is highlighted that the magnitude of the outcome differs considering the dominance of the operated limb.

Comparatively to the dominating limb, performing TKA to the non-dominant limb significantly improves gait speed (18.6%, p < 0.001) and reduces the double limb-support time (−12.7%, p = 0.041), while also increasing the stride length (10.5%, p < 0.001), allowing increased step cadence (7.5%, p = 0.019). Although there is also a statistical difference between the limbs in terms of stance, swing, and step time, these are very small changes, usually under 1% of the total gait cycle duration, and may lack clinical significance. Overall, participants who underwent surgery on the non-dominant limb demonstrated greater improvements in most parameters compared to the dominant (right) limb.

4. Discussion

This study compared the spatiotemporal gait parameters of the TKA, pre- and post-surgery, and examined the influence of the surgical side and participant’s lateral dominance on these variables. After TKA, the surgical limb showed improvements in all variables related to gait, except stance time and swing time. Furthermore, the study revealed that better gait performance was obtained when surgery was performed on the non-dominant limb.

Six months after surgery, participants reported improvements in extension ROM, daily activities, and reduced pain and stiffness, based on clinical parameter results. These findings are in accordance with the literature [6,18,29,34]. According to Notarnicola et al. [35] after TKA, patients experience diminished pain levels and heightened daily functioning, resulting in decreased reaction time and improved movement control. Additionally, it is crucial to acknowledge that TKA contributes to the correction of axes deformities and joint alignment, with the rehabilitation program playing a pivotal role in modifying sensorimotor function.

As expected, after surgery, participants demonstrated greater confidence when walking, evidenced by improvements in spatiotemporal variables, except for stance and swing times. However, it is important to consider the time after the surgery as a factor that can influence these results. In this study, the aim was to evaluate patients at a time frame set to allow the real examination of the surgery’s clinical and functional impact. Mandeville et al. [6] found that walking speed improvement was evident 6 months after surgery. Also, Abbasi-Bafghi et al. [15] reported improvement in walking speed not sooner than 6 months after TKA. This persisted for up to 13 months after surgery, after which time a decline started to show [15]. Other studies showed that, despite improvements in patients’ perception about their walking abilities, pain, and ROM, objective gait measures did not improve between both three and 18 months after surgery [12,14]. Based on the presented literature, it can be suggested that walking speed may take several months to improve after TKA and this benefit may be lost over time. However, this is a significant positive finding because even minimal improvement of 0.1 m/s in walking speed has been associated with better functional outcomes [17].

Another improvement presented in this study is related to stride length and stride cadence which have been shown to both be reduced in KOA patients [18]. According to the KOA severity scale of Elbaz et al. [20], a shorter stride length with a lower cadence is indicative of a higher functional severity grade. Our findings are supported by others, like Mandeville et al. [6], that found a significant increase in stride length after TKA. Since both characteristics were improved by TKA in both limbs in our study, we can sustain that a primary TKA can have a positive functional outcome.

When considering the dominance of the lower limb (right for all the participants) and the surgical side, our findings showed an improvement in the gait parameters, especially when the non-dominant limb was operated. The literature has considered laterality as an additional explanation for the functional behavior of the lower limbs during the gait, with the gait asymmetry reflecting a natural functional difference between limbs [22]. Given that both characteristics were improved by TKA in our study, we can postulate that TKA may produce positive functional outcomes in both limbs, potentially contributing to a reduction in the severity of KOA.

The findings of our study are clinically relevant, as they show that gait analysis is an important tool for evaluating patients before and after TKA, which can help tailor rehabilitation programs to each patient’s specific gait abnormalities. Several studies have focused on the role of specific rehabilitation programs in changes to gait parameters changes after TKA [3,14,18]. Improving gait stability with targeted rehabilitation programs can be considered an important objective during postoperative rehabilitation [14], with more intensive programs leading to clinically significant improvements in the later stages of recovery after TKA.

A limitation of this study is the small sample size, which was constrained by the number of patients available for elective surgery at the local hospital. Another limitation concerns the absence of a control group of healthy patients of the same age, which could have provided a more robust comparison of the post-TKA improvements. However, the proportion between male and female participants in this study does reflect that of KOA patients in Portugal, and the non-pathological gait parameters established in the literature were utilized to interpret our findings. Additionally, the lack of standardized rehabilitation programs for participants might have influenced the results, as variations in post-surgery rehabilitation can affect outcomes. Another potential limitation is recall bias, particularly concerning the number of falls reported in the month before the assessment. Nonetheless, this did not impact our results significantly, as no differences were observed between the groups in this regard.

5. Conclusions

Our findings suggest that after TKA, there are improvements in all variables related to gait in the limb undergoing surgery, except stance and swing time. Furthermore, results indicated that improved gait performances are observed when surgery is performed on the non-dominant lower limb. As such, this study points to gait analysis as a safe, effective, and reliable evaluation method to obtain objective data on KOA and post-TKA knee function, which may be important to personalize rehabilitation programs and to maximize functionality after TKA.