Obstacle Crossing in Older Adults with Total Knee Arthroplasty at the Initial Swing Phase

Abstract

After undergoing a total knee arthroplasty (TKA) procedure, patients are at a high risk of falling because they present with ineffective mobility within a complex environment, especially during obstacle crossing. Toe clearance (TC) is an important factor to quantify the risks of trip-related falls. The study aimed to investigate TC height and toe trajectory and joint kinematic changes occurring in the lower limb following TKA during obstacle crossing at the initial swing phase. Twenty TKA patients, including those in preoperative and postoperative stages (three and six months), performed obstacle-crossing tasks to compare their performance with 20 healthy controls. Participants walked at self-pace along an 8 m walkway with 2.5, 5, and 10 cm obstacles positioned along the center of the path. For each participant, body segment motions were traced using reflective markers and the kinematics of lower extremity, toe clearance, and gait parameters were analyzed using a 3D-motion analysis system. TKA patients had lower TC height and toe trajectory at six months, slower toe elevation than controls when swing toe crossed 5 and 10 cm obstacles (p < 0.05), and decreased hip and knee flexion (p < 0.05). These altered gait patterns with decreased TC height and toe trajectory were identified as tripping factors as the toe trajectory was close to the ground surface. TKA patients had acquired different lower limb kinematics to maintain adequate TC. At long-term follow-up, there was an increasing trend for patients to trip after surgery. Therefore, more focus is needed on the exercise prescription for rehabilitation programs to improve muscle strength and stepping control.

1. Introduction

Knee osteoarthritis (OA) causes deformities of the joints, stiffness, muscle weakness, and pain during walking [1]. More than 50% of patients with knee OA report having one or more falls every year, which suggests that chronic knee pain is a risk factor for falling. [2]. Crucial factors of falls are accident- or environment-related (31%) and gait, balance disorders or weakness (17%) [3]. These walking problems are often stimulated by environmental factors, e.g., an uneven or slippery surface, height, and slopes [4]. One common cause of falls leading to mortality among the elderly [5] is tripping while attempting to cross a barrier. Knee osteoarthritis (OA) has been proven to increase the probability of tripping over obstacles [6].

TKA is a common surgical intervention for treating end-stage knee osteoarthritis which provides pain relief, improves physical functioning, corrects deformities, and improves the quality of life [7]. A principal goal for TKA patients is the recovery of gait function that allows the patients to independently perform various activities of their daily lives (e.g., household work, shopping) and helps them regain an active lifestyle, which enables them to restore lower-limb functions after surgery [8]. Additionally, patients who underwent TKA also improved their walking. However, they have decreased obstacle-navigation skills and are prone to trip-related falls [9]; their walking ability resulted in lesser total knee motion during gait, with lesser knee flexion during swing than the controls [10]. Interestingly, the rate of falling in TKA patients has been shown to range between 12% and 38%, with tripping on an object being the most common reason [11,12].

The main cause of falls is followed by tripping or stumbling (21%) and slipping (3%) [13]. Tripping occurs when the foot clearance during the swing phase of gait is interrupted by an obstacle or a sudden change in surface level [14] and toe contact with the ground was influenced by a decrease in toe clearance. [15]. Inadequate toe clearance increases the chance of tripping while walking on uneven ground, contributing to falls [16], especially in persons with very fluctuating foot trajectories. Therefore, toe clearance is critical to determine the risk of tripping-associated forward falls [17,18]. Walking on level ground (24% of all falls) and walking on uneven ground (24% of all falls) were the two most common activities among elderly individuals at the time of the fall [19]. Regarding walking on uneven ground, obstacle crossing is a multi-joint movement that requires accurate swing foot control and increased inter-joint coordination of swing legs [20]. There are a few examples in the literature demonstrating knee kinematics during various functional activities (e.g., level walking at different speeds, semi-squat, mid-squat, step ascent and descent, and sit to stand) three and six months, one year, or more following TKA [21,22,23]. These studies show significant decrease in the range of motion of the operated side during gait and walking speed, a deficit in strength, and altered movement patterns. These changes could lead to altered mechanical functions of the joint during walking [24]. One study indicated that knee OA patients had decreased knee flexion and increased ankle dorsiflexion when they swung the toe above the obstacles [25]. Another study reported that patients with TKA displayed a decrease in active knee flexion and an increase in hip hiking to compensate knee flexor work while crossing obstacles [26]. Theses studied focused on the kinematics of swing limb when the toe was directly above the obstacles.

Despite the apparent success of TKA, many patients cannot attain the normal joint function of walking. Concerning trip-related falls, toe clearance during the initial swing phase is important because the marker of the toe position is close to the walking surface. Thus, the toe could come in contact with the obstacles and lead to tripping. Although several studies have evaluated the level walking in TKA patients, the TC and lower limb kinematics of these patients during obstacle crossing at initial swing phase remains unclear. Therefore, this study aims to investigate how patients six months post-TKA adapt their TC height and toe trajectory, lower limb joint kinematics in the sagittal plane, and stepping characteristics of the operated limb during obstacle crossing of various heights at the initial swing phase compared to healthy controls. We hypothesized that the changes in TC height and toe trajectory of affected limbs and the changes in the lower limb kinematics and toe elevation during the initial swing phase in postoperative patients are close to those in healthy control groups.

2. Materials and Methods

2.1. Study Design and Participants

This study is a prospective descriptive-analytical study of patients evaluated pre-operatively and three and six months post-operatively compared to healthy adults. All participants completed the approved informed consent procedures mandated by the Institutional Review Board on Human Research of the Faculty of Medicine, Chulalongkorn University, Thailand (COA No.029/2019). Twenty female patients who underwent unilateral TKA and 20 healthy females were enrolled in this study. Patients undergoing TKA were consecutively recruited by three orthopedic surgeons at the King Chulalongkorn Memorial Hospital from October 2018 to January 2020. Inclusion criteria included those (1) aged between 65 and 85 years, (2) with osteoarthritic changes in the tibiofemoral joint (osteophytes and/or joint space narrowing) at grade 3 or 4 knee OA on the Kellgren-Lawrence scale, (3) able to walk along an 8 m walkway without assistive devices, and (4) having normal vision. Exclusion criteria were as follows: (1) demonstration of any neurological disease or any history affecting gait ability, (2) diagnosis of rheumatoid or other systemic inflammatory arthritis, (3) intraarticular corticosteroid injection in the preceding two months, (4) having lower limb surgeries in the prior six months, and (5) morbid obesity (BMI > 40 kg/m²). The same criteria used to evaluate TKA participants were also used for control participants, who were in general good health. The lower limb of the control participants was matched to the operated side of knee OA patients as described by McClelland et al. [21].

2.2. Total Knee Arthroplasty

Patients scheduled for unilateral TKA were obtained using the following protocol: radiographic severity of OA was assessed using the K-L scale by an experienced senior orthopedic surgeon. All surgeries were performed under spinal anesthesia by a single senior surgeon with prior experience in over 1000 cases. The arthroplasty was carried out using the standard medial parapatellar approach with patellar resurfacing under tourniquet control. All patients received a cemented posterior-stabilized knee system—Scorpio NRG (Stryker Corporation, Kalamazoo, MI, USA), Nexgen LPS-Flex (Zimmer, Warsaw, IN, USA), and VEGA (B. Braun Melsungen AG, Melsungen, Germany) [27]—and physiotherapy including enhanced operative education, range of motion exercise, muscle strengthening, and gait education. The follow-up visits were scheduled for three and six months. The Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire assessed the patients’ perceptions of symptoms and stiffness, pain, and quality of life [28]. The score was divided into five scored subscales: pain, symptoms, function in daily living (ADL), function in sport and recreation (sport/rec), and knee-related quality of life (QOL). Each patient was assessed immediately before surgery and post-surgery.

2.3. Knee Active Joint Position Sense Test

The test was performed while seated and was recorded using motion capture cameras (OptiTrack, OR, USA) and reflexive markers (B & L Engineering, CA, USA). The MotiveBody 2.0.0 Final software was used to process the data. The absolute error (AE) was calculated as the difference between the target angle (TA) and repositioning angle. All participants were acquainted with the procedure before the assessment through explanation and practice. They were barefoot and wore compression pants. Reflexive markers were attached to anatomical landmarks—the lateral malleolus, head of fibula, femoral lateral epicondyle, and mid-point between the femoral lateral epicondyle and greater trochanter [29]. Participants sat on a treatment table, their lower legs relaxed and their trunks supported. The knee joint was maintained in individual flexion as the starting position (Figure 1). They were blindfolded and informed to passively extend their legs from the starting position to the TA of 15° [30]. The TA was established using a Universal Goniometer (Sammons Preston, Bolingbrook, IL, USA): the axis aligned with the lateral epicondyle of the femur, the stationary arm with the greater trochanter, and the moving arm aligned with the lateral malleolus. The measurement of knee motion with a goniometer was reliable (ICC = 0.993). After TA, a modified H-frame, an instrument used as a range of motion guide, was positioned to ensure the same TA. [29]. The uprights were made of steel pipes and the crossbar with elastic stretch to tighten around the uprights, which was located until the crossbar reached the anterior ankle joint line while participants held the target angles. They were instructed to hold at the TA for 5 s while concentrating on the angle with a modified H-frame. Then, they moved their leg to the starting position. After a relaxation period of 5 s, they were advised to actively reposition the legs at the same TA and maintained the position for 5 s without the H-frame. Then, they returned their legs to the starting position. Each test was repeated three times.

2.4. Lower Limb Strength Test

A modified load cell strain gauge isometric dynamometer (maximum 120 kg-force (kgf), error ± 0.05% kgf) was used to assess leg muscle strength. It was used to measure the peak force of the knee extensor and flexor muscles on an isometric module using a biopac MP 100 system with an acqKnowLedged software version 3.7.3 (Biopac System Inc., Montréal, QC, Canada). This study demonstrated excellent test-retest reliability (ICC = 0.969, ICC = 0.920, respectively) in the measurement of the knee extensor and flexor strength. All participants were informed about the procedure, which was performed in a sitting position with armed folded across their chest at a trunk-thigh angle of 90° and knee flexion at 90° [31,32]. A strap was secured around the trunk, waist, and distal part of the thigh and shin pad fixed to a modified isometric dynamometer was positioned perpendicular to the anterior or posterior aspect of the tibia and 5 cm proximal of the medial malleolus [33]. The measurement consisted of three maximal voluntary isometric contractions flexion and extension trials for 3 s, with 60 s pauses. The value of maximum peak force was measured in kgf and normalized to body mass (kg).

2.5. Gait Evaluation

Participants visited a gait and motion analysis laboratory where a 10-camera motion analysis system (OptiTrack, OR, USA) was used to track the motion of the lower body segments during gait. Two force plates (Bertec Corporation, OH, USA) embedded in the floor detected gait cycle events. The kinematics and the force plate data were synchronized and sampled at 100 Hz. Thirty reflective markers (B & L Engineering, CA, USA) were placed at anatomical landmarks, including the pelvis (Anterior-Superior Iliac Spine: ASISs and Posterior-Superior Iliac Spine: PSISs), thighs (greater trochanter, mid-thigh, medial and lateral epicondyles), shank (head of the fibula, tibial tuberosity, medial and lateral malleolus), and foot (first and fifth metatarsal base, lateral side of big toe, tip of the big toe, and heel) [25]. Raw data of reflexive markers were employed to generate an anatomical reference frame and foot segments. These data were also smoothed using a Butterworth low-pass filter with a cut-off frequency of 6 Hz and processed with the MotiveBody 2.0.0 Final software with composite accuracy of ±2 mm in each of the three coordinate directions. Participants walked barefoot at a self-selected comfortable pace on an 8 m walkway in the absence of obstacles (0 cm obstacle) and the presence of obstacles measuring 2.5, 5, and 10 cm in height, placed randomly in the middle of the walkway (Figure 2). The obstacles (length: 70 cm, depth: 5 cm) were made of soft sponges to prevent falls associated with contact [34]. Two reflective markers were placed on either end of the tube to define the location of the obstacle. The participants were allowed to familiarize themselves with the walkway and three successful trials for each condition were obtained. To reduce the effects of fatigue, they were provided appropriate time for rest. Obstacle crossing was analyzed by the swing phase divided into three: the initial phase (0–34%), mid-swing phase (34–65%), and terminal swing of the swing phase (65–100%) [35]. At level walking, the TC clearance (minimum toe height) was measured between the distal inferior surface of the toe and the ground during swing phase [36]. At obstacle crossing, the vertical position of the toe at the end of the initial swing phase while crossing the obstacle was defined as toe clearance. For each participant, data points from the relevant trials were combined to form averages for each walking condition.

2.6. Statistical Analysis

All data were analyzed using SPSS ver. 22.0 and demonstrated normal distribution tested by Shapiro-Wilk goodness of fit test with an α level of p < 0.05 (IBM Corp., New York, NY, USA). It was determined that all data were normally distributed and suited for parametric testing. Sample size for this research was calculated with an α of 0.5 and 80% power using data from Chen et al. [37] The sample size of this study is estimated by using two–sample independent groups. Each group required a minimum sample size of 18 participants. As a result, 20 participants per group should be sufficient to detect a difference between groups. The marker data were used to calculate TC height and toe trajectory and lower limb joint kinematics. Toe clearance of the affected limb was calculated as the distance at which the affected limb reached its height between tip of the toe and the ground during the swing phase. The slope of the toe trajectories was measured relative to the change in vertical position of the rising toe and the change in horizontal distance as the tip of toe traveled between 0% and 34% of the initial swing phase. The knee characteristics of preoperative patients were demonstrated in numbers and percentages. For comparing the differences between the two groups, the independent t-test was performed considering demographic data, spatiotemporal variables, TC at level walk, knee active joint position sense, lower limb strength, TC height when crossing obstacles, the slope of toe trajectory, percentage of swing phase during crossing obstacles, joint angle of lower limb in sagittal plane at level walk and crossing obstacles. The differences within the TKA patient groups (preoperative, three- and six-month postoperative) were analyzed with repeated measures ANOVA as follows; KOOS score, spatiotemporal variables, TC at level walk, knee active joint position sense, lower limb strength, TC height when crossing obstacles, the slope of toe trajectory, joint angle of lower limb in sagittal plane at level walk. Bonferroni’s adjusted tests were used to assess the differences when a significant interaction effect was observed.

3. Results

3.1. Participant Characteristics and Gait Parameters

Forty participants took part in the study. Twenty underwent TKA before surgery and they visited the lab for check-ups until their sixth postoperative month. Excluding the three who withdrew from participation, the final analysis engaged thirty-seven participants. The control group included twenty healthy participants.

Table 1. Participant demographics and clinical characteristics.

Parameters	Preoperative TKA (n = 20)	Control (n = 20)	p-Value
Demographic
Age (year)	71.65 (5.90)	70.40 (3.69)	0.427
Height (cm)	150.30 (6.16)	153.33 (4.44)	0.083
Body weight (kg)	61.43 (10.69)	54.34 (8.08)	0.023 *
Body mass index (kg/m2)	27.09 (3.63)	22.99 (2.80)	0.001 *
Skeletal muscle mass (kg)	18.91 (2.54)	18.92 (2.06)	0.995
The knee characteristics
K/L grade, number (%)
Grade 3	14 (70%)	-
Grade 4	6 (30%)	-
Knee OA localization, number (%)
Right	13 (65%)	-
Left	7 (35%)	-
Knee alignment, number (%)
Varus	15 (75%)	-
Valgus	5 (25%)	-

Values are reported as mean ± SD, or n (%). * Significance value set at p < 0.05.

shows the characteristics of all participants. There were no significant differences in age, height, and skeletal muscle mass. However, body weight and body mass index had differences as the preoperative TKA was usually heavier. According to the severity of knee OA, participants with KL grades 3 and 4 represented 70% and 30% of all participants, respectively (Table 1). Thirteen patients were concerned with the right knee, whereas the remaining seven with the left knee. Of those with radiographic knee OA, 75% exhibited a varus knee and 25% a valgus knee.

Table 2. Clinical and gait parameters obtained from TKA patients at the level walk.

Parameters	TKA Patients			CG (n = 20)	Group Comparisons (p-Value)
	T1 (n = 20)	T2 (n = 19)	T3 (n = 17)		T1-CG	T3-CG	T1-T2-T3
KOOS (range 0–100)
Symptoms	75.35 (13.11)	85.79 (10.31)	91.18 (5.64)	-	-	-	T1-T2 = 0.001 *
							T1-T3 = 0.001 *
Pain	71.20 (13.95)	93.16 (6.92)	96.18 (4.93)	-	-	-	T1-T2 = 0.001 *
							T1-T3 = 0.001 *
							T2-T3 = 0.001 *
ADL	66.60 (14.14)	83.26 (10.25)	87.82 (6.72)	-	-	-	T1-T2 = 0.001 *
							T1-T3 = 0.001 *
Sports/recreation	21.00 (13.63)	31.05 (11.62)	25.59 (8.64)	-	-	-	0.065
Quality of life	41.35 (16.33)	83.31 (13.05)	82.47 (12.11)	-	-	-	T1-T2 = 0.001 *
							T1-T3 = 0.001 *
Spatiotemporal-affected side
Gait speed (m/s)	0.61 (0.13)	0.61 (0.12)	0.73 (0.12)	1.03 (0.16)	0.001 *	0.001 *	T2-T3 = 0.005 *
Stride width (m)	0.13 (0.05)	0.12 (0.03)	0.16 (0.14)	0.09 (0.02)	0.001 *	0.028 *	0.267
Stride length (m)	0.79 (0.23)	0.82 (0.13)	0.90 (0.10)	1.11 (0.11)	0.001 *	0.001 *	0.176
Step length (m)	0.40 (0.09)	0.42 (0.08)	0.44 (0.08)	0.56 (0.06)	0.001 *	0.001 *	0.495
TC of the swing phase (cm)	2.50 (0.71)	2.77 (1.03)	2.43 (0.72)	2.22 (0.67)	0.199	0.347	0.187
Joint angle in sagittal plane at the time of TC (degree)
Hip flexion	18.81 (7.93)	14.20 (8.27)	13.21 (6.95)	23.53 (7.07)	0.001 *	0.001 *	T1-T2 = 0.030 *
Knee flexion	39.10 (5.06)	35.85 (4.62)	38.37 (6.93)	48.65 (7.30)	0.001 *	0.001 *	0.464
Ankle dorsiflexion	4.01 (5.06)	3.34 (4.62)	2.25 (5.70)	2.77 (3.74)	0.682	0.692	0.386
Knee active joint position sense (degree), affected side
Absolute error at 15°	6.49 (2.47)	5.09 (2.41)	5.24 (2.55)	4.94 (3.15)	0.093	0.756	0.267
Lower limb strength (Kgf/BW), affected side
Knee extensor strength	0.26 (0.09)	0.20 (0.07)	0.23 (0.08)	0.50 (0.13)	0.000 *	0.000 *	0.50
Knee flexor strength	0.12 (0.04)	0.11 (0.03)	0.11 (0.03)	0.20 (0.05)	0.000 *	0.000 *	0.690

Values are reported as mean ± SD. * Significance value set at p < 0.05. T1—preoperative; T2—at 3 months; T3—at 6 months; CG—control group. Kgf/Bw = Kilogram-force/bodyweight.

shows clinical and gait parameters at the level walk. At six months postoperatively, significant differences were mostly found in the five subscales of the KOOS (symptoms, pain, activity daily living, sport and recreation function and QOL) among the patient groups (p < 0.05). The global satisfaction with five subscales was improved especially the quality of life.

For level walking, the spatiotemporal parameters of TKA at six months had a significant difference compared to the control group. However, no difference was found among groups of TKA patients in terms of all parameters, except the gait speed of TKA at six months. This parameter indicates that TKA at six months walked faster than TKA at three months (0.73 ± 0.12 m/s vs. 0.61 ± 0.12 m/s; p < 0.05). In addition, the six months of TKA exhibited the same vertical position as the control group (2.43 ± 0.72 cm vs. 2.22 ± 0.67 cm, p = 0.347), which occurred at around 50% of the swing phase (Table 2, Figure 3a). At the time of TC, the preoperative TKA exhibited significantly greater hip flexion than the three-month postoperative TKA (p < 0.05). The knee flexion and ankle dorsiflexion were not different among the TKA patients (p = 0.464, p = 0.386). In addition, the hip flexion and the knee flexion of the TKA at six months were found to be significantly lower than that of the controls (p < 0.05) and ankle dorsiflexion was not significantly different between the groups (p = 0.692).

Table 2 presents the active joint position sense of the knee. The AE values did not differ significantly within pre-TKA and post-TKA groups. Similar results were demonstrated in AE between the TKA groups and control groups. Regarding the lower limb strength, there was no significant difference in quadriceps muscle strength between pre-TKA and post-TKA groups. The TKA groups had considerably less knee extensor strength than the control group. In the hamstring muscle, a comparable result was obtained. The flexor muscle strength of the TKA groups was considerably lower than that of the control group. There was no significant difference in hamstring muscle strength within pre-TKA and post-TKA groups.

3.2. Toe Clearance Height and Toe Trajectory of Crossing Obstacle with Different Heights

The TC height of the TKA at six months while crossing obstacles at 5 and 10 cm was lower than the controls at the end of the initial swing phase (p = 0.004, p = 0.001). No significant differences were found in the TC height of all TKA groups at the end of initial swing phase while crossing obstacles at the same levels (p = 0.587, p = 0.474) (Figure 3c,d). There were no significant differences between the TKA at six months and control groups in terms of the 2.5 cm TC height at the end of initial swing phase (p = 0.261). In addition, the analysis did not reveal any significant differences in the same TC height and phase within the TKA groups (p = 0.212) (Figure 3b). The toe rising of the 5 cm and 10 cm obstacles crossing at the end of initial swing phase were significantly difference between the TKA at 6 months and the control groups (0.38 ± 0.17 vs. 0.54 ± 0.24, p = 0.032; 0.54 ± 0.20 vs. 0.78 ± 0.33, p = 0.016, respectively) (Figure 3c,d), except with the slope of the 2.5 cm obstacle (0.40 ± 0.19 vs. 0.35 ± 0.13, p = 0.403) (Figure 3b). This suggests that the slopes of the TKA at six months were lower than those of the control group while crossing higher obstacles. In addition, the outcome of toe trajectory within the TKA groups presented that there were significant differences in the toe trajectory between preoperative and the TKA at three-month groups while crossing 5 cm obstacle (0.40 ± 0.15 vs. 0.52 ± 0.23, p = 0.034).

3.3. Step Characteristics of Obstacle Crossing

When walking across the obstacles, the participants elevated their toes at different time points of the initial swing phase before crossing them (Figure 4). Compared to the control groups, TKA at six months had increased percentages of initial swing phase at 5 and 10 cm obstacle heights (p = 0.003, p = 0.001) but not at 2.5 cm (p = 0.110). It implies that the TKA at six months lifted the toe slower than the control group (Figure 5).

3.4. Joint Motion of Crossing Obstacles with Different Heights

Crossing the obstacle at 2.5, 5, and 10 cm, the TKA at six months displayed significantly decreased hip flexion at the end of initial swing phase compared to the control group (p = 0.002, p = 0.001, p = 0.001, respectively) (Figure 6a,d,g). Additionally, the TKA at six months showed decreased knee flexion at the end of initial swing phase compared to the control group (p = 0.001, p = 0.001, p = 0.001, respectively) (Figure 6b,e,h). Moreover, they displayed no significant difference in ankle dorsiflexion compared to the control group at the same phase (p = 0.247, p = 0.074, p = 0.100, respectively) (Figure 6c,f,i).

4. Discussion

This study showed that patients following TKA for six months exhibited improvement in spatiotemporal gait parameters, i.e., gait speed, stride length, and step length at the level walk. The enhancement of theses parameters achieved after surgery was associated with knee pain release and collected knee alignment [38,39]. However, these results were not retrieved compared to those of the control group. Although the TKA patients were usually heavier than the controls, the body weight and BMI did not affect functional gait restoration. Accordingly, the findings of a previous study examined the relation between BMI and gait velocity during walking. It was found that TKA patients with the highest BMI improved their gait velocity even though they did not accomplish the same levels of gait velocity as healthy participants [40]. At level walking, the TKA at six months decreased in TC height which is close to the control group (2.43 ± 0.72 cm vs. 2.22 ± 0.67 cm, p = 0.347) (Table 1). TKA patients presented decreased hip flexion and knee flexion during swing phase. A similar phenomenon was observed during level walking. The TC height of knee OA was similar to asymptomatic controls; to reach the same TC height, the TKA patients exhibited increased hip abduction and knee flexion. [41]. In general, knee OA patients who underwent conservative treatment or TKA changed their subsequent lower limb movement to allow for adequate toe clearance and stability throughout swing phase. As reported, the evidence we found indicates that knee OA patients underwent neuromuscular changes that negatively affected their knee proprioception [42]. Likewise, the TKA patients underwent resection of anterior and posterior cruciate ligaments, leading to an alteration in proprioception [43]. These groups also exhibited changes in the gait movement pattern and were prone to tripping due to the impairment of sensory information in knee receptors.

Following TKA for six months, there was no significant difference in the TC height at 2.5 cm obstacle compared to the control group. Such height is as low as ground level so that participants can reach this TC height. Although a previous study on toe clearance during obstacle crossing to prevent the risk of trip-related falls in TKA patients (unilateral posterior cruciate-retaining knee replacement) revealed that surgical limb toe clearance at different obstacle heights (6 and 18 cm) was higher than that of the control group [26], we found that TC height of the TKA at six months was lower than that of the control group at the end of the initial swing phase at the obstacle height (5 and 10 cm). The distinct changes in TC height between these studies involved differences in marker attachment, toe displacement, type of surgery, and implant materials. The marker was at the great toe in the previous study while it was placed at the tip of the great toe in this study. Researchers defined the toe displacement in different circumstances. The former research described a vertical displacement between the toe and the top of the obstacle. Meanwhile, this study explained it as the vertical displacement between the toe tip and the ground surface. Moreover, the other factor is the type of surgery. The previous study performed unilateral TKA that retained some part of the knee receptors while, the present study had TKA patients whose cruciates were resected. In addition, these two studies used different types of implant materials. These can lead to alteration in proprioception and also changes in gait pattern of lower limb. Individuals may adapt to altered knee function when crossing over an obstacle. Even though the patients underwent treatment in different ways, it was found that all groups could cross the obstacles. However, the patient groups with TKA in this study may have had an increased propensity to step on an obstacle because the TKA at six months displayed a decrease in the slope of toe trajectory and a slow ability to elevate their toes compared to the control group at the 5 and 10 cm obstacles. In addition, related research focused on the kinematic analysis of the gait over the obstacles. Individuals who had undergone knee replacement were examined while crossing over the obstacles. The patients displayed an increase in hip hiking and a decrease in knee flexion to maintain toe clearance at the control level [26]. There is evidence to support that the lower extremity joint kinematics in swing limb including the hip, knee, and ankle joint angles while crossing the three different obstacles (2.5, 5, and 10 cm) affected the TC of surgical limb. In this research, the TKA at six months also displayed a significant decrease in hip flexion and knee flexion compared to the control group, although the angle of ankle dorsiflexion when crossing the obstacles was the same for both groups. The results of kinematic studies focusing on the affected swing limb showed a significant decrease in knee flexion. These circumstances may be related to the fact that the pathology is osteoarthritis of the knee. Patients with knee OA displayed common deformities in the lower extremity that affected the function of the lower limb, a change in the kinematics of the lower limb, an alteration in muscle tension, an increase in the force of gravity on the ligaments of the knee, a change in the tension of the outward knee ligament, and an alteration in the signals sent out from their mechanical receptors to the central nervous system [43]. These various changes in knee OA were related to gait mechanics, specifically to neuromuscular changes, which had a negative impact on proprioception [42]. Six months following surgery, the patients had altered sensory input and their quadriceps and hamstrings muscle strength decreased but remained within the normal range for the elderly; however, it was still lower than in the control group. It can be suggested that increasing lower limb strength may enhance leg lift and raise TC. The sensory information from the visual field can be considered as well. Earlier studies used visual information to view the situation before crossing the obstacles as a feed-forward. These characteristics can help individuals adapt their toe clearance while stepping before crossing [44,45]. As far as we know, these findings could well be probably responsible for trip-related falls during the initial swing phase of crossing higher obstacles because the reduction in slope of toe trajectory and delayed capacity to elevate legs would imply that the toe moves close to the obstacles.

Normally, knee joint proprioception is important to neuromotor control, as it receives the afferent information from mechanoreceptors in the muscles, ligaments, capsule, menisci, and skin. After that, the afferent input conveys the information to integral motor learning and the proceeding program of complex movements at supraspinal centers [46,47]. Locomotion results from complex interactions between the central nervous system, the sensory system and the musculoskeletal system. During locomotion, the central pattern generator (CPG), which is a neural circuit in the spinal cord, initiates the basic rhythm and neural activation pattern underlying locomotion. The movement patterns of walking or crossing over obstacles are regulated by the multisensory afferent input and the resultant force produced by coordinated activation of agonist and antagonist muscle groups. In the control of leg muscles activity, extensor muscles are primarily determined by proprioceptive feedback, and the flexor muscles are principally under central control [48]. Similarly, in the current study, knee OA patients who underwent TKA with resected cruciate ligaments displayed kinematic alterations and a decrease in the hip and knee flexion. The altered joints motion could be related to the altered afferent feedback of quadriceps muscles following TKA, contributing to the regulation of quadriceps alpha motor neuron excitability [49]. Another reason for the gait pattern control could be related to the afferent input, which is initiated from the ankle while crossing over the obstacles to the higher center, decreasing the gain of some reflex pathways because of the lack of knee proprioceptive ability. Thus, the interpretative signals from the ankle, knee, and hip joint receptors were not available. It can be reasonably assumed that knee OA pathology and the effects of TKA contribute to the alteration in the gait pattern of the lower limb and lead to tripping due to the impairment of sensory information in the knee receptors. As the focus of the study was on TC height and toe trajectory with a self-selected walking speed during obstacle crossing, it could not have been applicable to other work if the focus had been on these parameters with fast-walking or running.

5. Conclusions

In summary, the TKA group at six months showed an improvement in spatiotemporal parameters and were able to cross over the obstacles throughout the gait cycle. They were successful in crossing lower obstacles. However, they exhibited a decrease in TC height and slope of toe trajectory at the end of the initial swing phase and delay in stepping before crossing over the higher obstacles. Surprisingly, a decrease in hip flexion and knee flexion was found when achieving gait function while crossing higher obstacles after six months. This study only investigated the sagittal plane kinematics and this limitation could have influenced the results. Moreover, the TKA at six months may likely have been accustomed to abnormal gait movement for a long time to maintain toe clearance and this may have brought about changes in lower joint kinematics. The evidence from this study suggests that the increased risk of tripping during higher obstacle crossing owing to toe-obstacle contact may occur when the affected limb was crossing due to the toe trajectory close to the ground surface. When the patients encounter many activities of daily living that are surrounded by different ground levels, sidewalk curbs, stair climbing, etc., and they must be careful to avoid the possibility of tripping. Although the TKA patients could not return to normal gait patterns, their treatment for knee arthritis was deemed successful, delivering relatively high satisfaction in terms of pain relief, functional recovery, and required improvement in the quality of life. Therefore, rehabilitation programs for the lower extremities, such as strengthening the muscles of the lower limbs, may be an essential component of postoperative rehabilitation programs to improve muscle strength and the stepping control. We believe that our research will serve as a basic for future studies on trip-related fallings when crossing over obstacle in TKA patients.