Phys. Ther. Korea 2019; 26(2): 49-60
Published online May 31, 2019
https://doi.org/10.12674/ptk.2019.26.2.049
© Korean Research Society of Physical Therapy
Chang-man An1, and Jung-suk Roh2
1Dept. of Physical Therapy, The Graduate School, Hanseo University,
2Dept. of Physical Therapy, Division of Health Science, Hanseo University
Correspondence to: Jung-suk Roh
After stroke, in order to improve gait function, it is necessary to increase the muscle strength and to enhance the propriocetive function of the lower extremity. This study aimed to compare the effects of open kinetic chain (OKC) versus closed kinetic chain (CKC) isokinetic exercise of the hemiparetic knee using the isokinetic equipment on lower extremity sensorimotor function and gait ability in patients with chronic stroke. Thirty participants with chronic hemiplegia (> 6 months post-stroke) were randomly divided into 2 equal groups: CKC group and OKC group. Patients from both groups attended conventional physiotherapy sessions 3 times a week for 6 weeks. Additionally, subjects from the CKC group performed isokinetic exercise using the CKC attachment, while those from the OKC group performed isokinetic exercise using the OKC attachment. The isokinetic knee and ankle muscles strength, position sense of the knee joint, and spatiotemporal gait parameters were measured before and after interventions. The knee muscles peak torque/body weight (PT/BW) and hamstring/quadriceps (H/Q) ratio significantly increased in both groups ( CKC isokinetic exercise can be an effective therapeutic intervention for the improvement of sensorimotor function of the lower extremity and gait functions, such as gait velocity and symmetry. CKC position in isokinetic strength training is effective to improve functional ability in patients with chronic stroke.Background:
Objects:
Methods:
Results:
Conclusion:
Keywords: Closed kinetic chain, Isokinetic exercise, Knee, Open kinetic chain, Stroke
After stroke, the sensorimotor dysfunction is a prominent clinical features and is known to be one of the main factors in slowing the recovery of functions (Cramp et al, 2006;,Lin, 2005). Especially, lower extremity muscle weakness and impaired proprioception can lead to markedly reduced functional abilities, such as normal gait function, in hemiplegic patients (Lin, 2005;,Patten et al, 2004). They may have increased spatiotemporal gait asymmetry and decreased gait velocity and energy efficiency (Hus et al, 2003;,Patterson et al, 2010;,Wist et al, 2016). In clinical practice, physical therapy was intended to increase muscle strength and enhance proprioceptive function (Bloem et al, 2000;,Wist et al, 2016).
Progressive resistive training (PRT) has been known as an effective intervention for muscle strengthening in various musculoskeletal and neuromuscular diseases (Jan et al, 2009;,Wist et al, 2016). PRT is generally classified into open kinetic chain (OKC) exercise and closed kinetic chain (CKC) exercise (Stensdotter et al, 2003). OKC exercise are single joint movements that are performed in non-weight bearing positions with a free distal extremity. In contrast, CKC exercise are multi-joint movements performed in weight bearing or simulated weight bearing postures with a fixed distal extremity (Stensdotter et al, 2003). OKC exercise is a more frequently applied treatment option, because it implies the absence of weight loading, and therefore, is associated with reduced exercise tolerance and risk of falls (Irish et al, 2010). Nevertheless, CKC exercise is known to better improve the muscle coordination pattern and functional status compared to OKC exercise, because of the recruitment of more muscle groups and requirement of additional skeletal stabilization (Mansfield et al, 2010;,Protas et al, 2005;,Stensdotter et al, 2003). In addition, CKC exercise may provide more sensory feedback through multiple joint proprioceptive reactions (Bloem et al, 2000;,Irish et al, 2010).
The isokinetic muscle strengthening represents a potentially interesting physical therapy intervention for patients with stroke. Because of the real-time visual and auditory feedback provided by different devices and the easy monitoring of subjects’ performance, it can increase the adherence to the physical therapy intervention (Hammami et al, 2012). Previously, maximal isokinetic strength training in OKC position of the knee musculature has been found to significantly improve gait velocity without increasing muscle tone in patients with stroke (Engardt et al, 1995;,Sharp and Brouwer, 1997). These authors concluded that the use of isokinetic exercises in hemiparetic patients was beneficial not only for muscle strengthening, but also for functional ability improvement. However, Kim et al (2001) pointed out the lack of a control group in previous studies that reported on improvement of functional ability through isokinetic exercise. Rather, previous research on isokinetic strength exercise in OKC position performed during 6 weeks in hemiplegic patients showed that although the muscle strength improved, no significant effect on gait function was detected (Kim et al, 2001). It was suggested that isokinetic strength training may need to be performed in conjunction with practical tasks to be of functional benefit.
Many previous studies have compared the effectiveness of the OKC and CKC exercise, and the effects of muscle strengthening exercise using the isokinetic equipment in stroke patient was also confirmed (Irish et al, 2010;,Sharp and Brouwer, 1997;,Stensdotter et al, 2003). However, although there are two types (OKC and CKC) of isokinetic equipment, most previous studies have used OKC attachment to the isokinetic equipment for isokinetic strengthening exercises in patients with stroke, while CKC attachment was rarely used (Killington et al, 2010;,Kim et al, 2001). When isokinetic strength training is performed in the CKC position to improve function after stroke, we expect it will be more effective in improving function not only by providing muscle strengthening, but also by allowing muscular control and sensory stimulation of the lower extremity. It is necessary to study the effects of CKC exercise using the isokinetic equipment. Therefore, the purpose of this study was to compare the effects of OKC versus CKC exercise of the knee joint using the isokinetic equipment on the lower extremity sensorimotor function and gait ability in patients with chronic stroke.
This study was carried out prospectively and the convenient sampling method was used to select the samples. Community dwelling stroke survivors were recruited on a volunteer basis through the rehabilitation center of a university hospital. All subjects were randomly divided equally into 2 groups: The CKC isokinetic exercise group and OKC isokinetic exercise group. A conventional rehabilitation program was applied to all subjects and maximal concentric isokinetic exercise was additionally applied to the hemiparetic knee in CKC or OKC position.
Thirty community-dwelling individuals who had experienced a stroke and had residual unilateral weakness were included. The inclusion criteria were as follows: (1) a history of a single stroke at least 6 months before participating in the study, (2) ability to walk independently for a minimum of 10 meters without assistive device, (3) an activity tolerance of 45 minutes with resting time, and (4) nonparticipation in any formal training program or similar interventions. Participants were excluded if they had (1) visual impairment and comprehensive aphasia, (2) an unstable medical condition (i.e., uncontrolled hypertension, arrhythmia, congestive heart failure, or unstable cardiovascular status), (3) a significant musculoskeletal problem (i.e., fracture, unstable joint conditions). Each subject received an explanation of the study’s purpose and methods prior to participation and provided informed consent according to the ethical principles of the Declaration of Helsinki.
Isokinetic exercise of the hemiparetic knee was performed using the Biodex System 3 PRO dynamometer (Biodex Medical Systems, Inc., Shirley, NY, USA). The non-paretic lower limb was not trained. Verbal encouragement and visual feedback on a monitor were provided during training sessions to motivate maximal effort (McNair et al, 1996). All participants received a total of 45 minutes of CKC or OKC isokinetic exercise per session, 3 times per week for 6 weeks. Each training session included a warm-up of 5 minutes using a stationary bicycle and 10 minutes of passive stretching to cool-down of the knee extensors and flexors after the isokinetic exercise.
The OKC isokinetic exercise was conducted using the OKC knee attachment. Each subject was seated in a comfortable upright position with a 110˚ hip flexion and 90˚ knee flexion. The trunk, pelvis, and thigh were fixed to the dynamometer chair with Velcro straps to minimize body movement and then the knee attachment was fixed 1 ㎝ above the medial malleolus. The OKC isokinetic exercise consisted of 15 sets of maximal knee extension and flexion performed at 3 angular velocities, including low speed (at 90˚/sec, 5 repetitions, 5 sets), moderate speed (at 120˚/sec, 8 repetitions, 5 sets), and high speed (at 150˚/sec, 10 repetitions, 5 sets), separated by 10 seconds of rest after every set and 2 minutes of rest before each angular velocity change (Büyükvural et al, 2015) (Figure 1).
Isokinetic exercise (A: open kinetic chain, B: closed kinetic chain).
The CKC isokinetic exercise was conducted using a “leg press” knee attachment. Participants were positioned in the seat of the isokinetic dynamometer with 90˚ hip flexion and 45˚ knee flexion. Similar to the OKC position, the participant’s trunk, pelvis, and thigh were fixed to the dynamometer chair with Velcro straps to minimize body movement. Then, to stabilize the subject’s foot and footplate, a padded Velcro brand hook and loop fastener foot strap was threaded through the middle strap guides and secured over the forefoot. The heel strap was then threaded through the appropriate heel strap slots and secured tightly against the patient’s heel. The footplate could be used in a fixed position of 5-10˚ plantarflexion. The CKC isokinetic exercise consisted of 15 sets of maximal knee extension and flexion performed at 3 angular velocities, including low speed (at 90˚/sec, 5 repetitions, 5 sets), moderate speed (at 120˚/sec, 8 repetitions, 5 sets), and high speed (at 150˚/sec, 10 repetitions, 5 sets) separated by 10 seconds of rest after every set and 2 minutes of rest before each angular velocity change (Figure 1).
Measurements of the knee and ankle isokinetic strength, knee joint position sense, and gait performance were assessed 2 to 3 days before and 2 to 3 days after the intervention.
Maximal concentric contraction of the knee and ankle muscles was measured by the Biodex System 3 PRO. This equipment has shown good-to-excellent test-retest reliability of the muscle strength measurements (Flansbjer et al, 2005). Prior to the test, all participants performed a 5-minute warm-up using a stationary cycle and test procedures using sub-maximal practice trials to familiarize with the equipment. After a 10-minute rest time, to measure the knee isokinetic muscle strength, the subject was seated with the lateral femoral epicondyle of the knee joint axis aligned with the mechanical axis of the dynamometer. Then, the knee attachment was fixed 1 ㎝ above the medial malleolus. Each participant was required to fold the arms across the chest, and their trunk, pelvis, and thigh were fixed to the dynamometer chair with a Velcro strap in order to prevent compensation by the upper limbs, minimize body movement, and ensure optimal movement of the knee joint of the paretic limb during the test (Pincivero et al, 2003). Further, to promote maximal concentric contraction, visual feedback from a computer monitor and verbal instruction of push and pull as hard as possible were provided (McNair et al. 1996). Five maximum concentric contractions were performed at 60˚/sec, and the peak torque generated over 5 repetitions was recorded and normalized to the body weight (MacIntyre et al, 2010).
Then, maximal concentric contraction was measured for the dorsiflexors and plantarflexors. Participants were seated with the ankle joint axis aligned with the mechanical axis of the dynamometer. After performing a practice trial, participants were instructed to push and pull the attachment as hard and as fast as possible. Five maximum concentric contractions were performed at 30˚ /sec, and the peak torque generated over 5 repetitions was recorded and normalized to the body weight (MacIntyre et al, 2010).
In addition, the hamstring/quadriceps (H/Q) ratio was calculated to determine the movement quality of the knee joint, as this index may be used as a potential biomarker to probe the therapeutic effective ness of muscle strengthening. The formula for calculating H/Q ratio was as follows (Hong et al, 2012):
H/Q ratio (%) = (isokinetic peak torque of hamstring ÷ isokinetic peak torque of quadriceps) × 100
Biodex System 3 PRO was used to assess the position sense of the knee joint. Before the test, one practice trail was performed with eyes open to familiarize the tester with the explanation and equipment. After 10 minutes of rest, the participants wore an eye patch and earplugs to prevent the input of visual and auditory information (Peixoto et al, 2011). To measure the position sense of the knee joint, the participant was placed on the equipment and the starting angle of the test was set at 90˚ knee flexion. The target joint angles were 30˚ and 60˚ knee flexion; participants were instructed in advance to hold for 10 seconds and perceive the target joint angle position. During passive extension of the knee joint at an angular velocity of 2˚/sec, each participant was instructed to press a stop button when he or she recognized the target angle (Rombaut et al, 2010). The knee joint reposition error was defined as the absolute difference between the reproduction angle and target angle. A total of 3 repetitions were performed and 10 seconds of rest were provided for each measurement. The mean value of the error angles was used for the analysis and was measured before the strength test to minimize the effect of muscle fatigue (Hiemstra et al, 2001). Sekir et al (2008) showed that the position sense measurement using isokinetic device has high intra-tester reliability (intraclass correlation coefficient=. 90-.94).
Spatiotemporal gait parameters were measured using the GAITRite mat (CIR Systems Inc., Clifton, NJ, USA). The GAITRite mat contains 6 sensor pads with 13,824 sensors encapsulated in a roll-up carpet with an active area of 3.66 m length and .61 m width. Data were sampled at 30 ㎐. Participants performed 3 trials of walking at a self-selected speed across a level 10 m walkway with a pressure sensitive mat in the middle (Patterson et al, 2010). Prior to data collection, participants performed one practice trial to familiarize themselves with the procedure. Three trials were recorded for each participant and the average values were used for subsequent analysis. The following variables were obtained: gait velocity, step length, and swing time. Spatial (step length) and temporal (swing time) gait symmetry indices were calculated as shown below (Hendrickson et al, 2014):
Symmetry index = non-paretic limb value ÷ (paretic limb value + non-paretic limb value)
The gait symmetry index (SI) ranges from 0 to 1. A gait SI of .5 indicates equal values of the paretic and non-paretic limbs (i.e., perfect symmetry). A SI <.5 indicates that the paretic limb has a greater value than the non-paretic limb, while SI >.5 indicates that the non-paretic limb has a greater value than the paretic limb for a particular variable (Hendrickson et al, 2014).
Data were statistically analyzed using PASW Statistics for Windows, ver. 18 (SPSS Inc., Chicago, IL, USA). Continuous variables, such as age, disease duration, height, body weight, Korean version of the Modified Barthel Index, isokinetic PT/BW, H/Q ratio, reposition error, and gait spatiotemporal parameters were presented as mean±standard deviation. Demographic and clinical characteristics of subjects were compared between two groups by the independent t-test (for continuous variables). The χ2 test was used to compare the categorical variables, such as gender, hemiplegic side, and type of stroke, between two groups. A paired t-test was used to analyze the training effects within group differences (pre and post intervention), and the independent t-test for differences between the groups. Statistical significance for α was set at .05.
There were 15 patients (8 males, 7 females) with a mean age of 54.2±12.4 years in the CKC group and 15 patients (9 males, 6 females) with the mean age of 53.0±8.7 years in the OKC group. Demographic and clinical features of the participants are listed in Table 1. No significant difference in baseline characteristics and physical parameters were found between the groups (p>.05).
Table 1 . Demographics and clinical characteristics of the participants (N=30).
![]() |
After 6 weeks of isokinetic exercise, the PT/BW of the knee muscles significantly increased in both groups compared to pre-intervention PT/BW (p<.01). However, no statistically significant difference was found between the 2 groups in training effects (p>.05). The mean change in the knee extensor PT/BW were 26.29 Nm/㎏ in the CKC group and 32.32 Nm/kg in the OKC group, while those for the knee flexor PT/BW were 12.01 Nm/㎏ in the CKC group and 16.96 Nm/㎏ in the OKC group. The PT/BW of the ankle plantarflexor PT/BW significantly increased in the CKC group compared to that in the OKC group (p<.01). The mean change values in the ankle plantarflexor PT/BW were 9.98 Nm/㎏ in the CKC group and .90 Nm/㎏ in the OKC group. However, the PT/BW of the ankle dorsiflexor was not significantly different between two groups (p>.05) (Table 2).
Table 2 . Comparison of the isokinetic strength and position sense of the paretic lower extremity pre and post intervention.
![]() |
The H/Q ratio was not significantly different between the groups (p>.05). However, it significantly increased in both groups compared to pre-intervention H/Qratio (p<.01). The mean change values in the H/Q ratio were 9.16 % in the CKC group and 16.55 % in the OKC group. The reposition error was significantly decreased in the CKC group compared to the OKC group after intervention (p<.05). The mean change values of the knee joint reposition error were -2.82˚ in the CKC group and -.63˚ in the OKC group (Table 2).
Gait velocity significantly increased in both groups after intervention (p<.01). Inparticular, statistical analysis of the mean changes showed significant improvements in the CKC group compared to the OKC group (p<.01). The mean change values in the gait velocity were .14 ㎧ in the CKC group and .10 ㎧ in the OKC group. The spatial gait SI was significantly improved in the CKC group compared to the OKC group (p<.01). However, the temporal gait SI was not significantly different between the groups (p<.05) (Table 3)
Table 3 . Comparison of the spatiotemporal gait parameters pre and post intervention (N=30).
![]() |
Patients with stroke experience muscle weakness and impaired proprioceptive function in the lower extremity (Cramp et al, 2006;,Lin, 2005;,Patten et al, 2004). Many researchers have concluded that the knee muscle strength and proprioceptive function are closely related to gait function (Hsu et al, 2003;,Sharp and Brouwer 1997). Therefore, intervention programs emphasizing muscle strengthening and enhancing proprioceptive function are indicated for hemiplegic stroke patients without injury. In the current study, we aimed to analyze the effect of isokinetic PRT performed in CKC or OKC positions on the lower extremity sensorimotor function and gait performance in patients with chronic stroke. For this purpose, isokinetic equipment was used to perform PRT of the hemiparetic knee joint in CKC or OKC positions for 6 weeks. As a result, the OKC isokinetic exercise showed a significant improvement in the gait velocity, as well as in the muscle strengthening and H/Q ratio of the knee compared with pre-intervention. On the contrary, the CKC isokinetic exercise significantly improved muscle strengthening in the ankle plantarflexor, as well as muscle strength and H/Q ratio of the knee. In addition, it was confirmed that the reposition error of the knee joint decreased, while the gait velocity and spatial gait symmetry improved more significantly in the CKC group.
In general, isokinetic devices are known to be safe for evaluation and strength training without the risk of injury or increased muscle tone in persons with stroke (Engart et al, 1995;,Flansbjer et al, 2005;,Kim et al, 2001;,Sharp and Brouwer, 1997). They are very effective in inducing maximal concentric contraction of target muscles; in addition, they produce a faster rate of strength gain and reduce muscle tenderness better than isotonic exercise (Coudeyre et al, 2016;,Pontes et al, 2018). In the present study, both methods of isokinetic PRT were found very effective in improving the strength of target muscles. The OKC isokinetic exercise is very useful to repeatedly and constantly elicit maximal concentric contractions of the knee extensors and flexors. However, the effect of OKC isokinetic exercise was restricted to the target muscle, while CKC isokinetic exercise had a positive effect on the improvement of the ankle strength adjacent to the target muscle. These results may be related to the fact that the CKC isokinetic exercise effectively induces the contraction of more muscles around the target muscle by multi-joint movements (Augustsson et al, 1998;,Jan et al, 2009;,Stendotter et al, 2003). In other words, the CKC isokinetic exercise used the ankle muscles as well as the knee muscles, while the OKC isokinetic exercise mostly used the isolated knee extensor and flexors.
After interventions, with both CKC and OKC group showing improvement in gait velocity compared to baseline status. Previous studies similar results reported that isokinetic peak torque of the affected knee and ankle muscles predicted the gait velocity in patients with stroke (Flansbjer et al, 2005;,Hsu et al, 2003;,Olney et al, 1991). In addition, isokinetic strength training program of the hemiparetic knee and ankle muscles resulted in significant improvement in muscle strength and gait velocity (Büyükvural et al, 2015;,Killington et al, 2010;,Patterson et al, 2010). In our study, the increased strength of the knee muscles of the subjects after the intervention had a direct effect on the improvement of the gait velocity (Hyun et al, 2015).
Noteworthy, the comparison between the groups showed that the gait velocity was significantly higher in the CKC group than in the OKC group. The minimum clinically important difference value for gait velocity was set at .10-.20 ㎧ (Bohannon and Glenney, 2014). In this study, the gait velocity of the OKC group improved by .10 ㎧ and that of the CKC group improved by .14 ㎧. Therefore, the CKC isokinetic exercise was more clinically meaningful in improving gait velocity. The reason for this result could be the fact that CKC isokinetic exercise improved the strength of the ankle muscles as well as of the knee muscles. The ankle plantarflexor contributes to propelling the body forward by creating strong ground reaction forces in the late stance to early pre-swing phase of the gait cycle with comfortable speed (Allen et al, 2011;,Neptune et al, 2001;,Olney er al, 1991). Perhaps the reason for the significant improvement of the plantarflexors seems to be the result of a statistically higher gait velocity in the CKC group.
Hemiplegic patients typically had a limitation of weight shifting toward the paretic side due to muscle weakness and impaired proprioception of the lower limb, resulting in an asymmetric gait pattern (Hsu et al, 2003;,Lin, 2005;,Patterson, 2010). The subjects who participated in this study had a temporal SI of .43 and a spatial SI of .45. This means that the step time and step length of the non-paretic lower limb were shortened during gait. Perhaps the instability of the paretic lower limb was the main reason during the stance phase (Hendrickson et al, 2014). In this study, an improvement of spatial gait symmetry was noted in the CKC group compared to the OKC group. Although the CKC isokinetic exercise has not progressed at the weight bearing position, the knee extension and flexion with the distal extremity firmly fixed was similar to the actual gait pattern. Lee et al (2013) reported that the CKC exercise showed significant changes in contact area and peak contact pressure of the hind foot compared to the OKC exercise. These results would imply that CKC exercise is beneficial for weight bearing ability of the affected leg. The pushing of the CKC isokinetic exercise was accompanied by the hip extension, knee extension, and ankle plantarflexion, and pulling was accompanied by the hip flexion, knee flexion, and ankle dorsiflexion (Stensdotter et al, 2003). This repetitive motion of the lower extremity was similar to the actual gait pattern and daily routines compared with OKC exercise. As the CKC approach is based on the functional movement such as normal gait pattern performed during the exercise, this might be more easily accepted by patients during exercise (Krawczyk et al, 2014).
Moreover, a greater improvement in reposition sense of the knee joint was found in the CKC exercise group. Previous study has reported similar findings, an 8 week leg press exercise (i.e., CKC) improved reposition sense by 2.8°, however, leg curl exercise (i.e., OKC) improved reposition sense by .6˚ in patients with knee osteoarthritis (Jan et al, 2009). The CKC exercise has been shown to enhance proprioceptive performance by stimulation of mechanoreceptors in the intra-articular and firing muscle spindles in patients with the knee disorders (Hurd et al, 2006;,Jan et al, 2009;,Shields et al, 2005). As a result, the functional movement of the CKC exercise and the improvement of the knee joint position sense contributed to the stability of the stance phase of the affected leg. Maybe this point was the reason for the longer step length of the unaffected lower limb and improved spatial gait symmetry (An and Jo, 2017).
Finally, both OKC and CKC isokinetic exercises were effective in improving the H/Q ratio. The H/Q ratio is a parameter used to describe the muscular control of the knee joint (Aagaard et al, 1998). Decreased H/Q ratio has been suggested to be associated with increased risk of the knee joint pain and ligament injury (Aagaard et al, 1998;,Hewett et al, 2008). Our result showed that the H/Q ratio was significantly increased in both groups compared to baseline measurement values. Basically, it seems that both interventions were able to improve the balance of muscular strength between the agonist and antagonist muscles of the knee. Further, this result suggested that both interventions can be used to prevent the knee from secondary damages due to improved muscular control of the knee joint.
The present study has some limitations that require consideration when interpreting the results. First, the follow-up measurements were not performed; accordingly, the carryover effect of the 2 types of isokinetic exercise could not be determined. Second, the intervention applied in this study was expected to have a sufficient influence on the hip muscles. Nonetheless, the lack of measuring the isokinetic strength of the hip muscles also limits the interpretation of the results. Finally, this study included a small sample, and therefore its statistical results are difficult to generalize. Therefore, future studies should prove the long-term effects of the 2 types isokinetic exercises on the lower extremities sensorimotor function, including on the hip joint and on functional ability.
This study was conducted to determine the effect of isokinetic exercise on the hemiparetic knee at the CKC and OKC position in patients with chronic stroke. As a result, both interventions were effective in improving the strength of the knee muscles. In addition, the CKC isokinetic exercise improved strength of the ankle plantarflexors and knee joint position sense and positive effected gait velocity and spatial gait symmetry. Therefore, in order to improve the functional ability though isokinetic exercise in patients with stroke, functional and task-oriented muscle strengthening (i.e., CKC position) may be more effective than simple muscle strengthening task (i.e., OKC position). In addition, it may be necessary to develop various interventions that can be performed in CKC position.
Phys. Ther. Korea 2019; 26(2): 49-60
Published online May 31, 2019 https://doi.org/10.12674/ptk.2019.26.2.049
Copyright © Korean Research Society of Physical Therapy.
Chang-man An1, and Jung-suk Roh2
1Dept. of Physical Therapy, The Graduate School, Hanseo University,
2Dept. of Physical Therapy, Division of Health Science, Hanseo University
Correspondence to:Jung-suk Roh
After stroke, in order to improve gait function, it is necessary to increase the muscle strength and to enhance the propriocetive function of the lower extremity. This study aimed to compare the effects of open kinetic chain (OKC) versus closed kinetic chain (CKC) isokinetic exercise of the hemiparetic knee using the isokinetic equipment on lower extremity sensorimotor function and gait ability in patients with chronic stroke. Thirty participants with chronic hemiplegia (> 6 months post-stroke) were randomly divided into 2 equal groups: CKC group and OKC group. Patients from both groups attended conventional physiotherapy sessions 3 times a week for 6 weeks. Additionally, subjects from the CKC group performed isokinetic exercise using the CKC attachment, while those from the OKC group performed isokinetic exercise using the OKC attachment. The isokinetic knee and ankle muscles strength, position sense of the knee joint, and spatiotemporal gait parameters were measured before and after interventions. The knee muscles peak torque/body weight (PT/BW) and hamstring/quadriceps (H/Q) ratio significantly increased in both groups ( CKC isokinetic exercise can be an effective therapeutic intervention for the improvement of sensorimotor function of the lower extremity and gait functions, such as gait velocity and symmetry. CKC position in isokinetic strength training is effective to improve functional ability in patients with chronic stroke.Background:
Objects:
Methods:
Results:
Conclusion:
Keywords: Closed kinetic chain, Isokinetic exercise, Knee, Open kinetic chain, Stroke
After stroke, the sensorimotor dysfunction is a prominent clinical features and is known to be one of the main factors in slowing the recovery of functions (Cramp et al, 2006;,Lin, 2005). Especially, lower extremity muscle weakness and impaired proprioception can lead to markedly reduced functional abilities, such as normal gait function, in hemiplegic patients (Lin, 2005;,Patten et al, 2004). They may have increased spatiotemporal gait asymmetry and decreased gait velocity and energy efficiency (Hus et al, 2003;,Patterson et al, 2010;,Wist et al, 2016). In clinical practice, physical therapy was intended to increase muscle strength and enhance proprioceptive function (Bloem et al, 2000;,Wist et al, 2016).
Progressive resistive training (PRT) has been known as an effective intervention for muscle strengthening in various musculoskeletal and neuromuscular diseases (Jan et al, 2009;,Wist et al, 2016). PRT is generally classified into open kinetic chain (OKC) exercise and closed kinetic chain (CKC) exercise (Stensdotter et al, 2003). OKC exercise are single joint movements that are performed in non-weight bearing positions with a free distal extremity. In contrast, CKC exercise are multi-joint movements performed in weight bearing or simulated weight bearing postures with a fixed distal extremity (Stensdotter et al, 2003). OKC exercise is a more frequently applied treatment option, because it implies the absence of weight loading, and therefore, is associated with reduced exercise tolerance and risk of falls (Irish et al, 2010). Nevertheless, CKC exercise is known to better improve the muscle coordination pattern and functional status compared to OKC exercise, because of the recruitment of more muscle groups and requirement of additional skeletal stabilization (Mansfield et al, 2010;,Protas et al, 2005;,Stensdotter et al, 2003). In addition, CKC exercise may provide more sensory feedback through multiple joint proprioceptive reactions (Bloem et al, 2000;,Irish et al, 2010).
The isokinetic muscle strengthening represents a potentially interesting physical therapy intervention for patients with stroke. Because of the real-time visual and auditory feedback provided by different devices and the easy monitoring of subjects’ performance, it can increase the adherence to the physical therapy intervention (Hammami et al, 2012). Previously, maximal isokinetic strength training in OKC position of the knee musculature has been found to significantly improve gait velocity without increasing muscle tone in patients with stroke (Engardt et al, 1995;,Sharp and Brouwer, 1997). These authors concluded that the use of isokinetic exercises in hemiparetic patients was beneficial not only for muscle strengthening, but also for functional ability improvement. However, Kim et al (2001) pointed out the lack of a control group in previous studies that reported on improvement of functional ability through isokinetic exercise. Rather, previous research on isokinetic strength exercise in OKC position performed during 6 weeks in hemiplegic patients showed that although the muscle strength improved, no significant effect on gait function was detected (Kim et al, 2001). It was suggested that isokinetic strength training may need to be performed in conjunction with practical tasks to be of functional benefit.
Many previous studies have compared the effectiveness of the OKC and CKC exercise, and the effects of muscle strengthening exercise using the isokinetic equipment in stroke patient was also confirmed (Irish et al, 2010;,Sharp and Brouwer, 1997;,Stensdotter et al, 2003). However, although there are two types (OKC and CKC) of isokinetic equipment, most previous studies have used OKC attachment to the isokinetic equipment for isokinetic strengthening exercises in patients with stroke, while CKC attachment was rarely used (Killington et al, 2010;,Kim et al, 2001). When isokinetic strength training is performed in the CKC position to improve function after stroke, we expect it will be more effective in improving function not only by providing muscle strengthening, but also by allowing muscular control and sensory stimulation of the lower extremity. It is necessary to study the effects of CKC exercise using the isokinetic equipment. Therefore, the purpose of this study was to compare the effects of OKC versus CKC exercise of the knee joint using the isokinetic equipment on the lower extremity sensorimotor function and gait ability in patients with chronic stroke.
This study was carried out prospectively and the convenient sampling method was used to select the samples. Community dwelling stroke survivors were recruited on a volunteer basis through the rehabilitation center of a university hospital. All subjects were randomly divided equally into 2 groups: The CKC isokinetic exercise group and OKC isokinetic exercise group. A conventional rehabilitation program was applied to all subjects and maximal concentric isokinetic exercise was additionally applied to the hemiparetic knee in CKC or OKC position.
Thirty community-dwelling individuals who had experienced a stroke and had residual unilateral weakness were included. The inclusion criteria were as follows: (1) a history of a single stroke at least 6 months before participating in the study, (2) ability to walk independently for a minimum of 10 meters without assistive device, (3) an activity tolerance of 45 minutes with resting time, and (4) nonparticipation in any formal training program or similar interventions. Participants were excluded if they had (1) visual impairment and comprehensive aphasia, (2) an unstable medical condition (i.e., uncontrolled hypertension, arrhythmia, congestive heart failure, or unstable cardiovascular status), (3) a significant musculoskeletal problem (i.e., fracture, unstable joint conditions). Each subject received an explanation of the study’s purpose and methods prior to participation and provided informed consent according to the ethical principles of the Declaration of Helsinki.
Isokinetic exercise of the hemiparetic knee was performed using the Biodex System 3 PRO dynamometer (Biodex Medical Systems, Inc., Shirley, NY, USA). The non-paretic lower limb was not trained. Verbal encouragement and visual feedback on a monitor were provided during training sessions to motivate maximal effort (McNair et al, 1996). All participants received a total of 45 minutes of CKC or OKC isokinetic exercise per session, 3 times per week for 6 weeks. Each training session included a warm-up of 5 minutes using a stationary bicycle and 10 minutes of passive stretching to cool-down of the knee extensors and flexors after the isokinetic exercise.
The OKC isokinetic exercise was conducted using the OKC knee attachment. Each subject was seated in a comfortable upright position with a 110˚ hip flexion and 90˚ knee flexion. The trunk, pelvis, and thigh were fixed to the dynamometer chair with Velcro straps to minimize body movement and then the knee attachment was fixed 1 ㎝ above the medial malleolus. The OKC isokinetic exercise consisted of 15 sets of maximal knee extension and flexion performed at 3 angular velocities, including low speed (at 90˚/sec, 5 repetitions, 5 sets), moderate speed (at 120˚/sec, 8 repetitions, 5 sets), and high speed (at 150˚/sec, 10 repetitions, 5 sets), separated by 10 seconds of rest after every set and 2 minutes of rest before each angular velocity change (Büyükvural et al, 2015) (Figure 1).
Isokinetic exercise (A: open kinetic chain, B: closed kinetic chain).
The CKC isokinetic exercise was conducted using a “leg press” knee attachment. Participants were positioned in the seat of the isokinetic dynamometer with 90˚ hip flexion and 45˚ knee flexion. Similar to the OKC position, the participant’s trunk, pelvis, and thigh were fixed to the dynamometer chair with Velcro straps to minimize body movement. Then, to stabilize the subject’s foot and footplate, a padded Velcro brand hook and loop fastener foot strap was threaded through the middle strap guides and secured over the forefoot. The heel strap was then threaded through the appropriate heel strap slots and secured tightly against the patient’s heel. The footplate could be used in a fixed position of 5-10˚ plantarflexion. The CKC isokinetic exercise consisted of 15 sets of maximal knee extension and flexion performed at 3 angular velocities, including low speed (at 90˚/sec, 5 repetitions, 5 sets), moderate speed (at 120˚/sec, 8 repetitions, 5 sets), and high speed (at 150˚/sec, 10 repetitions, 5 sets) separated by 10 seconds of rest after every set and 2 minutes of rest before each angular velocity change (Figure 1).
Measurements of the knee and ankle isokinetic strength, knee joint position sense, and gait performance were assessed 2 to 3 days before and 2 to 3 days after the intervention.
Maximal concentric contraction of the knee and ankle muscles was measured by the Biodex System 3 PRO. This equipment has shown good-to-excellent test-retest reliability of the muscle strength measurements (Flansbjer et al, 2005). Prior to the test, all participants performed a 5-minute warm-up using a stationary cycle and test procedures using sub-maximal practice trials to familiarize with the equipment. After a 10-minute rest time, to measure the knee isokinetic muscle strength, the subject was seated with the lateral femoral epicondyle of the knee joint axis aligned with the mechanical axis of the dynamometer. Then, the knee attachment was fixed 1 ㎝ above the medial malleolus. Each participant was required to fold the arms across the chest, and their trunk, pelvis, and thigh were fixed to the dynamometer chair with a Velcro strap in order to prevent compensation by the upper limbs, minimize body movement, and ensure optimal movement of the knee joint of the paretic limb during the test (Pincivero et al, 2003). Further, to promote maximal concentric contraction, visual feedback from a computer monitor and verbal instruction of push and pull as hard as possible were provided (McNair et al. 1996). Five maximum concentric contractions were performed at 60˚/sec, and the peak torque generated over 5 repetitions was recorded and normalized to the body weight (MacIntyre et al, 2010).
Then, maximal concentric contraction was measured for the dorsiflexors and plantarflexors. Participants were seated with the ankle joint axis aligned with the mechanical axis of the dynamometer. After performing a practice trial, participants were instructed to push and pull the attachment as hard and as fast as possible. Five maximum concentric contractions were performed at 30˚ /sec, and the peak torque generated over 5 repetitions was recorded and normalized to the body weight (MacIntyre et al, 2010).
In addition, the hamstring/quadriceps (H/Q) ratio was calculated to determine the movement quality of the knee joint, as this index may be used as a potential biomarker to probe the therapeutic effective ness of muscle strengthening. The formula for calculating H/Q ratio was as follows (Hong et al, 2012):
H/Q ratio (%) = (isokinetic peak torque of hamstring ÷ isokinetic peak torque of quadriceps) × 100
Biodex System 3 PRO was used to assess the position sense of the knee joint. Before the test, one practice trail was performed with eyes open to familiarize the tester with the explanation and equipment. After 10 minutes of rest, the participants wore an eye patch and earplugs to prevent the input of visual and auditory information (Peixoto et al, 2011). To measure the position sense of the knee joint, the participant was placed on the equipment and the starting angle of the test was set at 90˚ knee flexion. The target joint angles were 30˚ and 60˚ knee flexion; participants were instructed in advance to hold for 10 seconds and perceive the target joint angle position. During passive extension of the knee joint at an angular velocity of 2˚/sec, each participant was instructed to press a stop button when he or she recognized the target angle (Rombaut et al, 2010). The knee joint reposition error was defined as the absolute difference between the reproduction angle and target angle. A total of 3 repetitions were performed and 10 seconds of rest were provided for each measurement. The mean value of the error angles was used for the analysis and was measured before the strength test to minimize the effect of muscle fatigue (Hiemstra et al, 2001). Sekir et al (2008) showed that the position sense measurement using isokinetic device has high intra-tester reliability (intraclass correlation coefficient=. 90-.94).
Spatiotemporal gait parameters were measured using the GAITRite mat (CIR Systems Inc., Clifton, NJ, USA). The GAITRite mat contains 6 sensor pads with 13,824 sensors encapsulated in a roll-up carpet with an active area of 3.66 m length and .61 m width. Data were sampled at 30 ㎐. Participants performed 3 trials of walking at a self-selected speed across a level 10 m walkway with a pressure sensitive mat in the middle (Patterson et al, 2010). Prior to data collection, participants performed one practice trial to familiarize themselves with the procedure. Three trials were recorded for each participant and the average values were used for subsequent analysis. The following variables were obtained: gait velocity, step length, and swing time. Spatial (step length) and temporal (swing time) gait symmetry indices were calculated as shown below (Hendrickson et al, 2014):
Symmetry index = non-paretic limb value ÷ (paretic limb value + non-paretic limb value)
The gait symmetry index (SI) ranges from 0 to 1. A gait SI of .5 indicates equal values of the paretic and non-paretic limbs (i.e., perfect symmetry). A SI <.5 indicates that the paretic limb has a greater value than the non-paretic limb, while SI >.5 indicates that the non-paretic limb has a greater value than the paretic limb for a particular variable (Hendrickson et al, 2014).
Data were statistically analyzed using PASW Statistics for Windows, ver. 18 (SPSS Inc., Chicago, IL, USA). Continuous variables, such as age, disease duration, height, body weight, Korean version of the Modified Barthel Index, isokinetic PT/BW, H/Q ratio, reposition error, and gait spatiotemporal parameters were presented as mean±standard deviation. Demographic and clinical characteristics of subjects were compared between two groups by the independent t-test (for continuous variables). The χ2 test was used to compare the categorical variables, such as gender, hemiplegic side, and type of stroke, between two groups. A paired t-test was used to analyze the training effects within group differences (pre and post intervention), and the independent t-test for differences between the groups. Statistical significance for α was set at .05.
There were 15 patients (8 males, 7 females) with a mean age of 54.2±12.4 years in the CKC group and 15 patients (9 males, 6 females) with the mean age of 53.0±8.7 years in the OKC group. Demographic and clinical features of the participants are listed in Table 1. No significant difference in baseline characteristics and physical parameters were found between the groups (p>.05).
Table 1 . Demographics and clinical characteristics of the participants (N=30).
![]() |
After 6 weeks of isokinetic exercise, the PT/BW of the knee muscles significantly increased in both groups compared to pre-intervention PT/BW (p<.01). However, no statistically significant difference was found between the 2 groups in training effects (p>.05). The mean change in the knee extensor PT/BW were 26.29 Nm/㎏ in the CKC group and 32.32 Nm/kg in the OKC group, while those for the knee flexor PT/BW were 12.01 Nm/㎏ in the CKC group and 16.96 Nm/㎏ in the OKC group. The PT/BW of the ankle plantarflexor PT/BW significantly increased in the CKC group compared to that in the OKC group (p<.01). The mean change values in the ankle plantarflexor PT/BW were 9.98 Nm/㎏ in the CKC group and .90 Nm/㎏ in the OKC group. However, the PT/BW of the ankle dorsiflexor was not significantly different between two groups (p>.05) (Table 2).
Table 2 . Comparison of the isokinetic strength and position sense of the paretic lower extremity pre and post intervention.
![]() |
The H/Q ratio was not significantly different between the groups (p>.05). However, it significantly increased in both groups compared to pre-intervention H/Qratio (p<.01). The mean change values in the H/Q ratio were 9.16 % in the CKC group and 16.55 % in the OKC group. The reposition error was significantly decreased in the CKC group compared to the OKC group after intervention (p<.05). The mean change values of the knee joint reposition error were -2.82˚ in the CKC group and -.63˚ in the OKC group (Table 2).
Gait velocity significantly increased in both groups after intervention (p<.01). Inparticular, statistical analysis of the mean changes showed significant improvements in the CKC group compared to the OKC group (p<.01). The mean change values in the gait velocity were .14 ㎧ in the CKC group and .10 ㎧ in the OKC group. The spatial gait SI was significantly improved in the CKC group compared to the OKC group (p<.01). However, the temporal gait SI was not significantly different between the groups (p<.05) (Table 3)
Table 3 . Comparison of the spatiotemporal gait parameters pre and post intervention (N=30).
![]() |
Patients with stroke experience muscle weakness and impaired proprioceptive function in the lower extremity (Cramp et al, 2006;,Lin, 2005;,Patten et al, 2004). Many researchers have concluded that the knee muscle strength and proprioceptive function are closely related to gait function (Hsu et al, 2003;,Sharp and Brouwer 1997). Therefore, intervention programs emphasizing muscle strengthening and enhancing proprioceptive function are indicated for hemiplegic stroke patients without injury. In the current study, we aimed to analyze the effect of isokinetic PRT performed in CKC or OKC positions on the lower extremity sensorimotor function and gait performance in patients with chronic stroke. For this purpose, isokinetic equipment was used to perform PRT of the hemiparetic knee joint in CKC or OKC positions for 6 weeks. As a result, the OKC isokinetic exercise showed a significant improvement in the gait velocity, as well as in the muscle strengthening and H/Q ratio of the knee compared with pre-intervention. On the contrary, the CKC isokinetic exercise significantly improved muscle strengthening in the ankle plantarflexor, as well as muscle strength and H/Q ratio of the knee. In addition, it was confirmed that the reposition error of the knee joint decreased, while the gait velocity and spatial gait symmetry improved more significantly in the CKC group.
In general, isokinetic devices are known to be safe for evaluation and strength training without the risk of injury or increased muscle tone in persons with stroke (Engart et al, 1995;,Flansbjer et al, 2005;,Kim et al, 2001;,Sharp and Brouwer, 1997). They are very effective in inducing maximal concentric contraction of target muscles; in addition, they produce a faster rate of strength gain and reduce muscle tenderness better than isotonic exercise (Coudeyre et al, 2016;,Pontes et al, 2018). In the present study, both methods of isokinetic PRT were found very effective in improving the strength of target muscles. The OKC isokinetic exercise is very useful to repeatedly and constantly elicit maximal concentric contractions of the knee extensors and flexors. However, the effect of OKC isokinetic exercise was restricted to the target muscle, while CKC isokinetic exercise had a positive effect on the improvement of the ankle strength adjacent to the target muscle. These results may be related to the fact that the CKC isokinetic exercise effectively induces the contraction of more muscles around the target muscle by multi-joint movements (Augustsson et al, 1998;,Jan et al, 2009;,Stendotter et al, 2003). In other words, the CKC isokinetic exercise used the ankle muscles as well as the knee muscles, while the OKC isokinetic exercise mostly used the isolated knee extensor and flexors.
After interventions, with both CKC and OKC group showing improvement in gait velocity compared to baseline status. Previous studies similar results reported that isokinetic peak torque of the affected knee and ankle muscles predicted the gait velocity in patients with stroke (Flansbjer et al, 2005;,Hsu et al, 2003;,Olney et al, 1991). In addition, isokinetic strength training program of the hemiparetic knee and ankle muscles resulted in significant improvement in muscle strength and gait velocity (Büyükvural et al, 2015;,Killington et al, 2010;,Patterson et al, 2010). In our study, the increased strength of the knee muscles of the subjects after the intervention had a direct effect on the improvement of the gait velocity (Hyun et al, 2015).
Noteworthy, the comparison between the groups showed that the gait velocity was significantly higher in the CKC group than in the OKC group. The minimum clinically important difference value for gait velocity was set at .10-.20 ㎧ (Bohannon and Glenney, 2014). In this study, the gait velocity of the OKC group improved by .10 ㎧ and that of the CKC group improved by .14 ㎧. Therefore, the CKC isokinetic exercise was more clinically meaningful in improving gait velocity. The reason for this result could be the fact that CKC isokinetic exercise improved the strength of the ankle muscles as well as of the knee muscles. The ankle plantarflexor contributes to propelling the body forward by creating strong ground reaction forces in the late stance to early pre-swing phase of the gait cycle with comfortable speed (Allen et al, 2011;,Neptune et al, 2001;,Olney er al, 1991). Perhaps the reason for the significant improvement of the plantarflexors seems to be the result of a statistically higher gait velocity in the CKC group.
Hemiplegic patients typically had a limitation of weight shifting toward the paretic side due to muscle weakness and impaired proprioception of the lower limb, resulting in an asymmetric gait pattern (Hsu et al, 2003;,Lin, 2005;,Patterson, 2010). The subjects who participated in this study had a temporal SI of .43 and a spatial SI of .45. This means that the step time and step length of the non-paretic lower limb were shortened during gait. Perhaps the instability of the paretic lower limb was the main reason during the stance phase (Hendrickson et al, 2014). In this study, an improvement of spatial gait symmetry was noted in the CKC group compared to the OKC group. Although the CKC isokinetic exercise has not progressed at the weight bearing position, the knee extension and flexion with the distal extremity firmly fixed was similar to the actual gait pattern. Lee et al (2013) reported that the CKC exercise showed significant changes in contact area and peak contact pressure of the hind foot compared to the OKC exercise. These results would imply that CKC exercise is beneficial for weight bearing ability of the affected leg. The pushing of the CKC isokinetic exercise was accompanied by the hip extension, knee extension, and ankle plantarflexion, and pulling was accompanied by the hip flexion, knee flexion, and ankle dorsiflexion (Stensdotter et al, 2003). This repetitive motion of the lower extremity was similar to the actual gait pattern and daily routines compared with OKC exercise. As the CKC approach is based on the functional movement such as normal gait pattern performed during the exercise, this might be more easily accepted by patients during exercise (Krawczyk et al, 2014).
Moreover, a greater improvement in reposition sense of the knee joint was found in the CKC exercise group. Previous study has reported similar findings, an 8 week leg press exercise (i.e., CKC) improved reposition sense by 2.8°, however, leg curl exercise (i.e., OKC) improved reposition sense by .6˚ in patients with knee osteoarthritis (Jan et al, 2009). The CKC exercise has been shown to enhance proprioceptive performance by stimulation of mechanoreceptors in the intra-articular and firing muscle spindles in patients with the knee disorders (Hurd et al, 2006;,Jan et al, 2009;,Shields et al, 2005). As a result, the functional movement of the CKC exercise and the improvement of the knee joint position sense contributed to the stability of the stance phase of the affected leg. Maybe this point was the reason for the longer step length of the unaffected lower limb and improved spatial gait symmetry (An and Jo, 2017).
Finally, both OKC and CKC isokinetic exercises were effective in improving the H/Q ratio. The H/Q ratio is a parameter used to describe the muscular control of the knee joint (Aagaard et al, 1998). Decreased H/Q ratio has been suggested to be associated with increased risk of the knee joint pain and ligament injury (Aagaard et al, 1998;,Hewett et al, 2008). Our result showed that the H/Q ratio was significantly increased in both groups compared to baseline measurement values. Basically, it seems that both interventions were able to improve the balance of muscular strength between the agonist and antagonist muscles of the knee. Further, this result suggested that both interventions can be used to prevent the knee from secondary damages due to improved muscular control of the knee joint.
The present study has some limitations that require consideration when interpreting the results. First, the follow-up measurements were not performed; accordingly, the carryover effect of the 2 types of isokinetic exercise could not be determined. Second, the intervention applied in this study was expected to have a sufficient influence on the hip muscles. Nonetheless, the lack of measuring the isokinetic strength of the hip muscles also limits the interpretation of the results. Finally, this study included a small sample, and therefore its statistical results are difficult to generalize. Therefore, future studies should prove the long-term effects of the 2 types isokinetic exercises on the lower extremities sensorimotor function, including on the hip joint and on functional ability.
This study was conducted to determine the effect of isokinetic exercise on the hemiparetic knee at the CKC and OKC position in patients with chronic stroke. As a result, both interventions were effective in improving the strength of the knee muscles. In addition, the CKC isokinetic exercise improved strength of the ankle plantarflexors and knee joint position sense and positive effected gait velocity and spatial gait symmetry. Therefore, in order to improve the functional ability though isokinetic exercise in patients with stroke, functional and task-oriented muscle strengthening (i.e., CKC position) may be more effective than simple muscle strengthening task (i.e., OKC position). In addition, it may be necessary to develop various interventions that can be performed in CKC position.
Isokinetic exercise (A: open kinetic chain, B: closed kinetic chain).
Table 1 . Demographics and clinical characteristics of the participants (N=30).
![]() |
Table 2 . Comparison of the isokinetic strength and position sense of the paretic lower extremity pre and post intervention.
![]() |
Table 3 . Comparison of the spatiotemporal gait parameters pre and post intervention (N=30).
![]() |