INTRODUCTION
Gait disorders are observed to prevail in a great proportion of stroke patients and limit their ambulatory ability in the community (Keenan et al., 1984). Hemiparetic patients predominantly support their body weight with the nonparetic lower limb after the occurrence of a stroke (Bohannon and Larkin, 1985). Interlimb symmetry of weight support is an important goal of rehabilitation (Sackley, 1991). Commonly, ground reaction force (GRF) has been used for the assessment of the lower limb support function, as it has simple applicability and high accuracy (Hesse et al., 1994). GRF is an important indicator of the intensity of the force on a body during ground contact (McClay et al., 1994). To our knowledge, previous studies have investigated the relationship between walking and the GRF patterns of stroke patients (Chen et al., 2007; Hsiao et al., 2016; Kim and Eng, 2003). The evaluation of gait in hemiparetic patients by the analysis of the paretic side is commonly used as the basis for gait treatments. However, gait analysis focuses on the rehabilitation of the paretic side suggesting that that there is a lack of evaluation of the nonparetic side.
It is suggested that severe stroke results in primary motor control problems in both the paretic side and the nonparetic side because they are compensating for the disturbed control of the contralateral side (Shiavi et al., 1987). The nonparetic limb adaptations may be the result of multiple compensatory mechanisms which would cause hemiparetic patients to adapt to their walking strategy (Chen et al., 2003; Gaviria et al., 1996). Additionally, their sedentary life contributes to the muscle atrophy on the nonparetic side (Tokuno and Eng, 2006). Indeed, a previous study has shown that considerable muscle weakness is present in the nonparetic side of hemiparetic patients (Gerrits et al., 2009). It is indicated that the muscle weakness may cause secondary complications, as a result of corticospinal tract damage (Madhavan et al., 2011). In fact, the estimated percentage of uncrossed tracts is about 10%–20% (Chollet et al., 1991). Thus, researchers have suggested that they could be activated with bilateral movement (Mudie and Matyas, 2000).
Despite its clinical importance, there has been little to no research on the difference between the nonparetic side in hemiparetic patients and the healthy subjects. To completely understand how persons after stroke overcome disabilities, one must understand the movement strategies used in the paretic side and the nonparetic side. It is proposed to us that the nonparetic limb function should not be neglected during exercise rehabilitation. Therefore, we conducted this study to compare the nonparetic side in hemiparetic patients with the dominant side in healthy subjects. The purpose of this study was to compare the GRF during gait of the nonparetic side in hemiparetic patients and the dominant side in healthy subjects.
METERIALS AND METHODS
Participants
A total of 40 participants (20 healthy subjects and 20 hemiparetic patients after stroke) were recruited in the study. The inclusion criteria were as follows: (a) age<70 years; (b) diagnosis of hemiplegia after stroke; (c) <24 months after the occurrence of stroke; (d) ability to follow verbal commands; (e) ability to walk to 50 m independently without any gait aids; (f) had Modified Ashworth Scale grade lesser than 2. The exclusion criteria were as follows: (a) lack of visual dysfunction; (b) unstable blood pressure or cardiovascular condition as determined by the physician; (c) strokes involving the brainstem or cerebellum; (d) orthopaedic problems from conditions other than stroke. Only healthy subjects with no neurological or musculoskeletal disorders affecting the lower limbs, and no history of surgery of the spine or lower limbs were considered. A total of 40 participants were randomly selected from people who responded to flyers that were placed throughout the hospital and to word of mouth. The subject characteristics are presented in Table 1. This study was approved by the Ethics Committee of Yongin University (approval number: 2-1040966-AB-N-01-20), and all subjects provided written informed consent.
10-m walk test
Neurological patients are the most regularly studied group of patients in studies applying walking assessments as an outcome measure (Graham et al., 2008). In this test, the individual independently walked without an assistive device for a distance of 10 m. The time taken for the intermediate 6 m is measured to allow for acceleration and deceleration. It can be either tested at the preferred walking speed or the maximum walking speed. The average value of three repetitions was calculated.
Vertical ground reaction force
The subjects were requested to walk at their most natural speed along a pathway over force plates used to record the GRF. The GRF data were collected using two force plates (464×508×82.5 mm, model OR 6-7-2000, Advanced Mechanical Technologies Inc., Watertown, MA, USA) set at a sampling rate of 1,000 Hz. Three appropriate trials were collected for each limb.
Range of motion in ankle joint
The digital goniometer has appropriate concurrent criterion-related validity as a tool for the measurement of joint ROM and equivalent inter- and intrarater reliability when compared to the general goniometer (Carey et al., 2010). The ankle ROM was measured with the participant placed in a supine position on a treatment table. A low cushion was placed below the knee joint to avoid a hyperextended knee. The digital goniometer axis was aligned approximately 1.5 cm inferior to the lateral malleolus. The stationary arm was aligned parallel to the longitudinal axis of the fibula, lining up with the head of the fibula. The participant completed three trials per limb, with the average of each side used as the representative value.
Statistical analysis
Statistical analyses were performed using IBM SPSS ver. 18.0 (IBM Co., Armonk, NY, USA). Data were expressed as means± standard deviations. The differences between the paretic side and the nonparetic side in hemiparetic patients were assessed by performing the Wilcoxon signed-rank test. The Mann-Whitney test was used to determine the differences between the nonparetic side in hemiparetic patients and the dominant side in healthy subjects. The Pearson correlation coefficients were used to elucidate the relationship between the GRF and ankle ROM. A P-value of <0.05 was considered as statistically significant.
RESULTS
10-m walk test
The self-selected and fast-velocity walking speeds were reduced in hemiparetic patients compared with those in healthy subjects (P<0.05) (Table 2).
Vertical ground reaction force
There was no statistically significant difference in the loading response force (LRF) and loading response time (LRT) values between the paretic side and nonparetic side of hemiparetic patients (P>0.05). The loading response impulse (LRI) value from the nonparetic side was significantly higher than from the paretic side of hemiparetic patients (P<0.05). All mid stance force (MSF), mid stance time (MST), and mid stance impulse (MSI) values from the nonparetic side were observed to be significantly higher than from the paretic side of hemiparetic patients (P<0.05). There was no statistically significant difference in the terminal stance force (TSF) and terminal stance time (TST) values between the paretic side and the nonparetic side sides of hemiparetic patients (P>0.05). The terminal stance impulse (TSI) value from the nonparetic side was observed to be significantly higher than from the paretic side of hemiparetic patients (P<0.05) (Table 3). The LRF value from the dominant side of healthy subjects was observed to be significantly higher than from the nonparetic side of hemiparetic patients (P<0.05). The LRT and LRI values from the nonparetic side of hemiparetic patients were significantly higher than from the dominant side of healthy subjects (P<0.05). The MTF and MSI values from the nonparetic side of hemiparetic patients were significantly higher than from the dominant side of healthy subjects (P<0.05). However, there was no statistically significant difference in the MST value between the nonparetic side of hemiparetic patients and the dominant side of healthy subjects (P>0.05). The TSF and the TST values from the dominant side of healthy subjects were observed to be significantly higher than from the nonparetic side of hemiparetic patients (P<0.05). The TSI value from the nonparetic side of hemiparetic patients was significantly higher than from the dominant side of healthy subjects (P<0.05) (Table 4).
Range of motion in ankle joint
There was a statistically significant difference in ankle dorsiflexion ROM value between the paretic side and nonparetic side of hemiparetic patients (P<0.05) (Table 5). Additionally, there was a statistically significant difference in the dorsiflexion ROM value between the dominant side of healthy subjects and the nonparetic side of hemiparetic patients (P<0.05) (Table 6).
Correlations between the ground reaction force and ankle range of motion
There was a significant correlation between the LRF value and the ankle dorsiflexion ROM (r=0.512, P<0.05). There was a significant correlation between the LRT value and the ankle dorsiflexion ROM (r=−0.342, P<0.05). There was a significant correlation between the LRI value and the ankle dorsiflexion ROM (r=0.391, P<0.05).
There was no significant correlation between MSF value and ankle dorsiflexion ROM (r=0.222, P>0.05). There was a significant correlation between the MST value and the ankle dorsiflexion ROM (r=0.302, P<0.05). There was no significant correlation between the MSI value and the ankle dorsiflexion ROM (r= −0.156, P>0.05). There was a strong and significant correlation between the TSF value and the ankle dorsiflexion ROM (r=0.530, P<0.05). There was a significant correlation between the TST value and the ankle dorsiflexion ROM (r=0.539, P<0.05). There was a significant correlation between the TSI value and the ankle dorsiflexion ROM (r=−0.405, P<0.05) (Table 7).
DISCUSSION
In this study, the vertical GRF values were evaluated and compared to analyse the differences between the hemiparetic patients and healthy subjects. A common GRF was a bimodal M form because the vertical force applied during the initial contact and push-off in the stance phase exceeds body weight, while those applied during the midstance were lesser than the body weight (Murray et al., 1969). Several hemiparetic patients may lose the heel-strike and push-off mechanism, changing the GRF form from the bimodal M form to an irregular form (Chen et al., 2007).
Our results show that the LRF and TSF values were different between hemiparetic patients and healthy subjects as the shock absorption on the heels in the initial contact and push-off in the phase does not work properly in hemiparetic patients. Individuals with poststroke hemiparesis often show reduced muscle activation of the paretic limb during walking (Burridge et al., 2001). There is sufficient evidence that the muscle weakness is reflected by the inability of stroke patients to generate normal muscle force (Bourbonnais and Vanden Noven, 1989).
In the present study, the RLT value indicated the later time both the paretic side and the nonparetic side in hemiparetic patients compared with healthy subjects, whereas the TST value indicated the earlier time. Based on these results, the hemiparetic patients cannot apply sufficient weight shift on both the paretic side and nonparetic side. Thus, they require additional time in the stance phase than the healthy subjects. The impulse values of both the paretic side and nonparetic side of hemiparetic patients were greater than those of the dominant side of healthy subjects; they were significantly greater on the nonparetic side than on the paretic side. These findings were consistent with earlier results, in which the nonparetic side showed greater impulse values than the paretic side (Horvath et al., 2001). This may explain why a proper weight was not loaded on the paretic side earlier. In addition, the reason is that additional time was needed to forward the weakened paretic limb during the swing phase. This creates difficulties such as the loading of overweight on the nonparetic side. There are differences between the performance of a nonparetic limb and a healthy limb (Olney et al., 1991; Olney et al., 1994; Parvataneni et al., 2007). For example, muscle activity is increased on the nonparetic limb during gait. It may not have a positive muscle activity on walking and might propose impairments (Raja et al., 2012). In addition, stroke patients often show changes in spatiotemporal properties during walking (Patterson et al., 2008). Such spatiotemporal asymmetries may be due to the compensatory strategies used by either the paretic limb, or the nonparetic limb (Chen et al., 2005). This can negatively affect walking ability (Olney and Richards, 1996). Thus, to improve walking ability, it is essential to not only understand the movement of the paretic limb, but to also understand the movement of the nonparetic limb. This study is clinically significant since we compared the nonparetic side in hemiparetic patients with the dominant side in healthy subjects.
Many previous studies have analyzed the relationship between the ankle and gait in stroke patients (Kitatani et al., 2016; Lamontagne et al., 2000; Lamontagne et al., 2002). Ankle dorsiflexion ROM is a crucial factor for the gait cycle (Dobkin et al., 2004). Therefore, we measured ROM in ankle dorsiflexion. The results showed that the paretic ankle dorsiflexion ROM was smaller than the nonparetic ankle dorsiflexion ROM in hemiparetic patients. It was consistent with the results of previous studies (Chung et al., 2004; Lamontagne et al., 2000; Lin et al., 2006). Decreased ankle dorsiflexion ROM of the paretic side can be caused by many factors, such as passive stiffness and spasticity of the plantar flexors (Dietz et al., 1981). Notably, the ankle dorsiflexion ROM of the nonparetic side was significantly smaller than that of the dominant side of healthy subjects in this study. It has been proposed that the ipsilaterally mediated effects from the neurological lesion may contribute to the ample muscle coactivation. The coactivation of the nonparetic side mainly during the stance phase had been indicated (Shiavi et al., 1987). It is likely that the increased muscle coactivation levels reported on the nonparetic side support postural stability during gait (Lamontagne et al., 2002). Though beneficial for postural stability, the increased coactivation reduces the effectiveness of movement and may produce part of the higher energy cost of locomotion reported after stroke (Zamparo et al., 1995). Limited dorsiflexion can change the foot positioning in weight bearing and decreased capacity to shift the center of gravity during standing or gait (Lin et al., 2006). In particular, the major compensations for limited dorsiflexion would include the hyperextension of the knee joint and flexion of the hip joint. Subsequently, the GRF vector would be in a backward direction of the hip joint. Therefore, these patients have an impaired dynamic balance during standing or gait (An and Jo, 2017). In the standard gait, the plantarflexor passive component controls the ankle dorsiflexion and assists the forward propulsion of the body at push-off (Hof et al., 1983).
Even though our results are not sufficient to generalize the correlation because of our small sample size, they demonstrate a strong correlation between the GRF during gait and ankle ROM. Ankle dorsiflexion ROM and plantarflexor peak torque are related and appear to be significant factors that contribute to ankle plantarflexor power and moments during walking. This suggests that increasing platarflexor peak torque and ankle dorsiflexion ROM may help increase gait ability and speed (Mueller et al., 1995). In addition, earlier studies showed some significant correlation between ankle plantarflexor and GRF (Chen et al., 2007; Lamontagne et al., 2000; Turns et al., 2007). We surmise that lesser ankle dorsiflexion ROM will make it difficult for appropriate power on the plantarflexor during the push-off. Maintaining proper conditions for ankle mobility may promote the dorsiflexion during gait, which aids the ankle push-off at the end of the stance phase of the gait.
We believe that despite the fact that the sample size was too small for generalization, the results provide a significant starting point for the future measurement of the relationship between the GRF and ankle ROM. Therefore, the results of this study must be confirmed with a larger sample size. This study also had other limitations to consider when interpreting the results. The position and the size of the force plate were not suitable for individuals with a step length.
In conclusion, there was a mostly significant difference in the nonparetic side of hemiparetic patients and the dominant side of healthy subjects. The results from our study can be used to aid exercise rehabilitation programs that solve the problems of the nonparetic side as well as the paretic side. These assessments can be identified based on a patient’s status, thus helping therapists to choose the best exercise programs for the rehabilitation of stroke patients.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
