Effects of constraint-induced movement therapy for the lower extremity among individuals post-stroke: A randomized controlled clinical trial

1Introduction

Patients post-stroke commonly suffer from upper extremity (UE) and lower extremity (LE) dysfunction (Hendricks, van Limbeek, Geurts, & Zwarts, 2002). Four out of five stroke survivors initially present with UE and LE hemiparesis (Hendricks et al., 2002). Further, at six months post stroke over 80% of patients will demonstrate persistent functional deficits (e.g., impairment of hand manual dexterity, muscle weakness, loss of postural control and abnormal gait), especially those with significant impairment at onset (Kwakkel, Kollen, & Lindeman, 2004). Despite rehabilitation efforts, only 5% of stroke survivors regain full function of the affected UE and LE (Staines, McIlroy, & Brooks, 2009; Whitall, McCombe Waller, Silver, & Macko, 2000). These findings strongly support the need to develop more effective therapeutic interventions for the paretic UE and LE.

Constraint-induced movement therapy (CIMT) is a common intervention for the treatment of the post-stroke paretic upper extremity (UE) (Lindsay et al., 2008). The development of CIMT as a therapeutic intervention was derived from the work of Edward Taub and colleagues (Taub et al., 1994). This intervention was targeted to overcome the phenomenon known as ‘learned non-use’. CIMT is comprised of four components: (i) administering intense, daily therapeutic exercises over consecutive days; (ii) utilizing a ‘shaping’ technique for providing function-oriented, supervised exercises for the paretic limb; (iii) behavioral strategies, also called a transfer package that facilitate the transfer of the learned skills from this intervention into everyday activities; and (iv) strategies to facilitate the use of the paretic limb (e.g., restraint for the non-paretic limb in UE CIMT) (Morris, Taub, & Mark, 2006; Taub et al., 1994; Taub et al., 2006; Taub et al., 2013). Various clinical trials and systematic reviews have investigated the efficacy of CIMT for improving the function of the paretic UE post-stroke. The results of these studies indicate that CIMT yields positive effects as seen in the improved function of the paretic UE post-stroke (Chiu & Ada, 2016; Corbetta, Sirtori, Castellini, Moja, & Gatti, 2015; Etoom et al., 2016; Fleet et al., 2014; Peurala et al., 2012; Sirtori, Corbetta, Moja, & Gatti, 2009; Stevenson, Thalman, Christie, & Poluha, 2012; Wolf et al., 2010; Wolf et al., 2006; Wolf et al., 2008).

As CIMT has been shown to yield positive functional effects for the UE post-stroke, studies have attempted to assess the therapeutic benefits of a CIMT program for the LE impairments and locomotion deficits post-stroke (Abdullahi, Aliyu, et al., 2021; Aruin, Hanke, Chaudhuri, Harvey, & Rao, 2000; Bonnyaud et al., 2013; Choi, Shin, Bang, & Choi, 2017; Dos Anjos, Morris, & Taub, 2020; e Silva et al., 2017; Fritz, Pittman, Robinson, Orton, & Rivers, 2007; Hase, Suzuki, Matsumoto, Fujiwara, & Liu, 2011; Kallio, Nilsson-Wikmar, & Thorsen, 2014; Marklund & Klassbo, 2006; Numata, Murayama, Takasugi, & Oga, 2008; Regnaux et al., 2008; Rodriguez & Aruin, 2002; Vearrier, Langan, Shumway-Cook, & Woollacott, 2005; Zhu et al., 2016). These studies speculated that a CIMT program for the LE may yield positive changes in balance and ambulation for people post-stroke. However, the bulk of these studies are small scale case reports and lack robust experimental evidence to indicate if CIMT is effective for reducing LE impairments and functional deficits. Further, the majority of these studies did not include all the elements (e.g., shaping techniques and transfer package) recommended for a successful CIMT program. Moreover, a number of studies have utilized a restraint device for the LE, which maybe a safety concern and may induce further coordination abnormalities.

The positive effects of CIMT are not necessarily related to the use of a restraint. Ample evidence suggests that the restraint alone does not yield similar functional improvement in the UE post-stroke, as does a complete CIMT program (Brogardh, Vestling, & Sjolund, 2009; Corbetta et al., 2015; Kwakkel, Veerbeek, van Wegen, & Wolf, 2015; Uswatte, Taub, Morris, Barman, & Crago, 2006). Further, various studies have reported positive effects following a CIMT program that did not use a restraint device for the UE (Brogardh et al., 2009; Corbetta et al., 2015; Kwakkel et al., 2015; Uswatte et al., 2006). CIMT program for LE (CIMT-LE), shows great potential for improving balance and ambulation for people post-stroke and further studies are needed to ascertain its effectiveness for improving LE function among people post-stroke. Therefore, the aim of the current study is to investigate the effectiveness of a CIMT-LE program for improving balance and ambulation among people post-stroke.

2Methods

3Results

The number of individuals assessed for eligibility were 53, of those 15 were excluded either for not meeting the inclusion criteria or did not wish to participate in the study. The number of participants who were randomized and allocated in their respective groups were 38 individuals. All the included individuals received the intervention program, attended the follow-up session and were included in the final analysis of the study (Fig. 1).

Fig. 1

Flowchart of participant recruitment.

Summary of participants’ characteristics is shown in Table 1. Results of independent t-tests for baseline measures showed that both groups were not significantly different at the outset of the study across all four outcome measures (FMA-LE), BBS, 10MWT and 6MWT. The overall adherence to the program was good for both groups, with an average of 3.3 hours/session for the CIMT group and an average of 3.2 hours/session for the control group. Adverse events were uncommon with 5 adverse event recorded for the CIMT group, and 3 for the control group. There were no missing data for all outcome measures.

The results of the ANOVA analyses show that both therapeutic programs are effective as seen across all outcome measures. A summary of outcome measures results overtime according to group is presented in Table 2. As both therapeutic programs progressed, FMA-LE scores, BBS score, 10MWT and 6MWT increased. Furthermore, a significant difference was found between groups as CIMT program yielded better improvement (as seen in outcome measures post-intervention) compared to control group program. In addition, retention of therapeutic effect was maintained among groups as seen in three-months follow-up assessment results.

The ANOVA results for FMA-LE showed that as the therapeutic program progressed, the FMA-LE significantly increased F(2,72) = 89.9, p < 0.001, η_p² = 0.71 (Fig. 2). Test of between-subjects effects indicated that the CIMT group had significantly higher scores compared to the control group F(1,36) = 6.6, p = 0.015, η_p² = 0.15. Post-hoc analyses showed that FMA-LE was not significantly different between groups at baseline (p = 0.48); however, FMA-LE was significantly different between groups at the end of the therapeutic program (p = 0.007) and approached significance at three months follow-up (p = 0.056).

Fig. 2

Mean scores of the Fugl-Meyer assessment-lower extremity (FMA-LE) as a function of time for both groups: CIMT and control. The time points at which outcome measures were collected from participants were: baseline (pre-intervention), end (post-intervention) and follow-up (3 months following the program conclusion).

For BBS scores ANOVA results showed that as the therapeutic program progressed, the BBS scores significantly increased F(2,72) = 125.3, p < 0.001, η_p² = 0.78 (Fig. 3). Main effects for time of assessment were superseded by time of assessment by group interaction F(2,72) = 10.5, p < 0.001, η_p² = 0.23. Test of between-subjects effects indicated that the CIMT group had significantly higher scores compared to the control group F(1,36) = 8.5, p = 0.006, η_p² = 0.19. Post-hoc analyses showed that BBS was not significantly different between groups at baseline (p = 0.92); however, BBS was significantly different between groups at the end of the therapeutic program (p = 0.001) and at three months follow-up (p < 0.001).

Fig. 3

Mean scores of the Berg balance scale (BBS) as a function of time for both groups: CIMT and control. The time points at which outcome measures were collected from participants were: baseline (pre-intervention), end (post-intervention) and follow-up (3 months following the program conclusion).

For 10MWT, ANOVA results showed that as the therapeutic program progressed, the 10MWT scores significantly increased F(2,72) = 159.1, p < 0.001, η_p² = 0.82 (Fig. 4). Main effects for time of assessment were superseded by time of assessment by group interaction F(2,72) = 4.2, p = 0.02, η_p² = 0.12. Test of between-subjects effects indicated that the CIMT group had significantly higher scores compared to the control group F(1,36) = 4.7, p = 0.04, η_p² = 0.12. Post-hoc analyses showed that 10MWT was not significantly different between groups at baseline (p = 0.97); however, 10MWT was significantly different between groups at the end of the therapeutic program (p = 0.02) and at three months follow-up (p = 0.005).

Fig. 4

Mean scores of the ten meter walk test (10MWT) measured in m/s as a function of time for both groups: CIMT and control. The time points at which outcome measures were collected from participants were: baseline (pre-intervention), end (post-intervention) and follow-up (3 months following the program conclusion).

For 6MWT, ANOVA results showed that as the therapeutic program progressed, the 6MWT scores significantly increased F(2,72) = 54.4, p < 0.001, η_p² = 0.60 (Fig. 5). Main effects for time of assessment were superseded by time of assessment by group interaction F(2,72) = 6.5, p = 0.003, η_p² = 0.15. Test of between-subjects effects indicated no significant difference (p = 0.23) between groups. Post-hoc analyses showed that 6MWT was not significantly different between groups at baseline (p = 0.82); and at the end of the therapeutic program the p-value approached significance (p = 0.053) but was not significantly different between the two groups. Lastly, at three months follow-up there were no significant differences between groups (p = 0.14).

Fig. 5

Mean scores of the six minute walk test (6MWT) measured in meters as a function of time for both groups: CIMT and control. The time points at which outcome measures were collected from participants were: baseline (pre-intervention), end (post-intervention) and follow-up (3 months following the program conclusion).

4Discussion

The current study examined the effectiveness of a CIMT-LE program compared to conventional post-stroke rehabilitation. The results indicate that at the end of the therapeutic programs both interventions yielded beneficial functional effects as measured by the FMA-LE, BBS, 10MWT and 6MWT. However, the CIMT program was more effective than the dose-matched, conventional post-stroke rehabilitation. These findings are in congruent with previous studies that examined CIMT for the lower extremity among people post-stroke (Abdullahi, Aliyu, et al., 2021; Aruin et al., 2000; Bonnyaud et al., 2013; Choi et al., 2017; Dos Anjos et al., 2020; e Silva et al., 2017; Fritz et al., 2007; Hase et al., 2011; Kallio et al., 2014; Marklund & Klassbo, 2006; Numata et al., 2008; Regnaux et al., 2008; Rodriguez & Aruin, 2002; Vearrier et al., 2005; Zhu et al., 2016). Providing individuals post-stroke with effective rehabilitation is clinically important and warrants further research, especially with the increasing number of people who continue to suffer from post-stroke impairments.

Previous studies that have examined CIMT for the LE have primarily focused on providing intensive LE rehabilitation and did not include, or only included some of the elements of CIMT. For studies that involve the upper extremity, ample evidence suggests that the TP amplifies the beneficial effects of CIMT in comparison to programs that did not include TP (Taub et al., 2013; Uswatte & Taub, 2013). Moreover, in a recent case report, the authors reported that the use of TP for a CIMT-LE program led to added improvements in gait and mobility among people post-stroke (Dos Anjos et al., 2020). It is still unknown if the addition of TP yields superior improvements compared to a CIMT-LE program without TP. Previous studies have reported positive benefits of a CIMT-LE without the use of TP (Abdullahi, Aliyu, et al., 2021; Aruin et al., 2000; Choi et al., 2017; e Silva et al., 2017).

The intervention protocol for CIMT-LE in the current study omitted the use of a restraint for the non-hemiparetic LE. The rationale for omitting a restraint is two-fold. One, to avoid the safety concerns as a restraint may increase the risk of falling. Two, the use of a restraint may prompt an abnormal gait or postural control pattern for participants during the intervention. The results of the study indicated that even with the absence of a restraint device the CIMT-LE protocol was successful in yielding positive functional benefits. This findings corroborates previous studies that concluded there are no differences between a CIMT-LE intervention with or without a LE restraint (Dos Anjos et al., 2020; e Silva et al., 2017). Similarly, previous studies investigating CIMT for the upper extremity have reported positive effects following a CIMT program that did not use a restraint device for the upper extremity (Brogardh et al., 2009; Corbetta et al., 2015; Kwakkel et al., 2015; Uswatte et al., 2006).

In a recent systematic review and meta-analysis, the authors reviewed 16 studies with different designs that utilized CIMT-LE (Abdullahi, Truijen, et al., 2021). The authors reported that the majority (n = 10) of studies included in their review were rated as level II evidence. The authors concluded that CIMT-LE improves motor function, balance, functional mobility and gait among people post-stroke. Further, the authors indicated that CIMT-LE is superior to conventional therapy in improving quality of life for people post-stroke. In a recent RCT, the authors aimed to compare the difference between two CIMT-LE protocols, one that focused on duration of therapy time while the other focused on number of task practice repetitions. The results of the study showed that both means of intervention delivery were effective in improving motor function, balance and functional mobility (Abdullahi, Aliyu, et al., 2021). Similarly, a previous RCT examined the efficacy of a modified CIMT-LE for improving gait parameters and center of mass displacement. The results showed that a modified CIMT-LE intervention was effective in improving the center of mass displacement both in the sagittal and frontal planes and also led to gait improvements.

In the current study participants from both groups showed significant differences following their respective interventions. However, the changes seen among the CIMT-LE group were clinically significant as seen by the improvements in the most of the study’s outcome measures. For FMA-LE, participants in the CIMT-LE group showed a change of six points in their average scores following the interventions compared to four points for the control group (Table 2). The change in average scores on the FMA-LE for the CIMT group is equal to the reported minimal clinically important difference (MCID) of six points (Pandian, Arya, & Kumar, 2016). Similarly, for the BBS both groups showed a significant improvement in their scores following their respective interventions. However, the change of 7 points for the CIMT-LE group (compared to 4 points for the control group) surpasses the reported MCID for the BBS of 5 points (Tamura, Miyata, Kobayashi, Takeda, & Iwamoto, 2021). For the 10MWT, the improvement seen among the CIMT-LE group of (0.14 m/s) is a substantial meaningful change as reported in a previous study (Perera, Mody, Woodman, & Studenski, 2006). For the 6MWT, although the change following the intervention was significant, it did not reach clinical significance.

The follow-up results at three months show that both groups retained the functional improvements following the intervention. There were no statistical significant differences in all outcome measures between the end of the therapeutic program and at three months follow-up. The retention of the therapeutic effects could be explained by the fact that all participants were given home instructions and exercises to continue treatment at home. This finding is similar to previous research that have reported that the beneficial effects of CIMT-LE are retained at three and six months following the conclusion of therapeutic training (Aruin et al., 2000; Marklund & Klassbo, 2006). With the increasing number of people who suffer from post-stroke impairment, it is essential to administer therapeutic programs that are effective in improving the functional status of individuals and are retained long after the program ends. Thus, enhancing the rehabilitation services provided to people post-stroke, and help alleviate the high loads that rehabilitation specialists face (Clarke et al., 2018).

The current study is not without limitations. Blinding of the clinicians administering the interventions was not possible; however, bias was minimized by blinding participants to study hypothesis. Another limitation was that the participants included in the current study were people with chronic stroke. As such, the findings may not be generalizable to other individuals in different phases of post-stroke recovery. However, specifying individuals post-stroke in the chronic stage was intentional from the outset of the study. Any recorded improvement seen following the intervention would more likely be attributed to the therapeutic program and not due to spontaneous recovery (Cassidy & Cramer, 2017).

5Conclusion

An intensive CIMT-LE program compared to an intensity-matched conventional program yielded significant clinical improvements among people post-stroke. These improvements were seen in LE motor recovery, postural balance and gait speed. Furthermore, these improvements were retained after three months of the therapeutic program conclusion. Therefore, these findings provide further support to the application of CIMT-LE for facilitating the recovery of people post-stroke.

Conflict of interest

The author reports no conflict of interest.

Funding

The author gratefully acknowledges Qassim University, represented by the Deanship of Scientific Research, for the financial support for this research (number 20007-fcohsb-2020-1-1-W) during the academic year 1442AH/2021AD.