Hindawi

Neural Plasticity
Volume 2021, Article ID 7266263, 9 pages
https://doi.org/10.1155/2021/7266263

Research Article
Associated Mirror Therapy Enhances Motor Recovery of the
Upper Extremity and Daily Function after Stroke: A Randomized
Control Study

Jin-Yang Zhuang ,1 Li Ding ,1 Bei-Bei Shu,2 Dan Chen ,2 and Jie Jia 1,2,3

1
Department of Rehabilitation Medicine, Huashan Hospital, Fudan University, Shanghai, China
2
Department of Rehabilitation Medicine, Shanghai Jing’an District Central Hospital, Shanghai, China
3
National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, China

Correspondence should be addressed to Jie Jia; [email protected]

Received 13 April 2021; Accepted 31 August 2021; Published 29 September 2021

Academic Editor: Xu-Yun Hua

Copyright © 2021 Jin-Yang Zhuang et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

Bimanual cooperation plays a vital role in functions of the upper extremity and daily activities. Based on the principle of bilateral
movement, mirror therapy could provide bimanual cooperation training. However, conventional mirror therapy could not
achieve the isolation of the mirror. A novel paradigm mirror therapy called associated mirror therapy (AMT) was proposed to
achieve bimanual cooperation task-based mirror visual feedback isolating from the mirror. The study was aimed at exploring
the feasibility and effectiveness of AMT on stroke patients. We conducted a single-blind, randomized controlled trial. Thirty-
six eligible patients were equally assigned into the experimental group (EG) receiving AMT and the control group (CG)
receiving bimanual training without mirroring for five days/week, lasting four weeks. The Fugl-Meyer Assessment Upper Limb
subscale (FMA-UL) for upper extremity motor impairment was used as the primary outcome. The secondary outcomes were
the Box and Block Test (BBT) and Functional Independence Measure (FIM) for motor and daily function. All patients
participated in trials throughout without adverse events or side effects. The scores of FMA-UL and FIM improved significantly
in both groups following the intervention. Compared to CG, the scores of FMA-UL and FIM were improved more significantly
in EG after the intervention. The BBT scores were improved significantly for EG following the intervention, but no differences
were found in the BBT scores of CG after the intervention. However, no differences in BBT scores were observed between the
two groups. In summary, our study suggested that AMT was a feasible and practical approach to enhance the motor recovery
of paretic arms and daily function in stroke patients. Furthermore, AMT may improve manual dexterity for poststroke
rehabilitation.

1. Introduction lesional upper limb also suffered motor dysfunction in 3

months after the onset of stroke [5], which hinders physical
Stroke is a leading cause of mortality and long-term disability function and independent daily activities.
worldwide [1], which results in a global economic burden for Compared to the healthy population, stroke patients
health care [2, 3]. Currently, many advanced technologies tend to avoid bilateral motor patterns in daily activities [6].
have been worked out and used for stroke survivors. Never- Lots of daily activities are inseparable from bimanual coop-
theless, we still face many challenges for poststroke rehabili- eration, such as twisting the towel, driving the car, and get-
tation, for instance, the paretic upper extremity. After stroke, ting dressed. For this reason, bilateral task relearning is
about 80% of patients remain having upper extremity motor essential for stroke patients. However, most therapeutic
impairment [4]. Besides, researchers have found that the ipsi- methods for stroke are concentrated on improving the
2 Neural Plasticity

contralesional arm function, ignoring participation of the 2. Methods

less affected side [7–9]. It is remarkable that protocols of
bilateral treatment (BT) which involve bilateral training with 2.1. Study Design. This study was an assessor-blinded, pre-
rhythmic auditory cues, bilateral priming, and device-driven test-posttest, randomized controlled trial. A separate investi-
bilateral training have been used as clinical treatments for gator was responsible for the clinical assessments but
stroke rehabilitation [10–12]. Based on bilateral, repetitive, blinded to the allocation. Meanwhile, two occupation thera-
and symmetrical motor principles, most bilateral treatments pists who were responsible for the therapeutic regimens were
(BTs) are executed through two independent and paralleled trained by one researcher. All patients received assessments
actions, which ignore cooperation between the hands; for before the intervention, after 2-week and 4-week interven-
instance, Sainburg et al. proposed the symmetrical coopera- tion. The study was approved by the ethics committee insti-
tive tasks regarded as a bilateral synergy framework for post- tutions of Huashan Hospital (KY2017-230) and registered at
stroke rehabilitation [6]. the Chinese Clinical Trial Registry (ChiCTR1800018351).
In addition to conventional physical intervention
methods, mirror therapy (MT) which relies on visual illu- 2.2. Participants. Patients were recruited from the Depart-
sion is regarded as a bilateral treatment [13, 14]. Under the ment of Rehabilitation Medicine, Huashan Hospital Affili-
MT environment, a plane mirror is placed in the median ated Jing’an Branch. Patients who had a first-ever ischemic
sagittal plane between upper limbs to induce the visual illu- or hemorrhagic stroke, occurring three months to one year,
sion, and patients are asked to move both arms as far as pos- aged between 25 and 75 years without severe cognitive
sible. Contrary to viewing directly on both arms, MT can impairment
provide normal visual stimulation of bilateral movement, (Mini‐Mental State Examination ðMMSEÞ score > 24), were
which has been proven to promote better the activation included. All patients were within the Brunnstrom stage of
and functional connectivity in the somatosensory system of hand over III and with modified Ashworth scale ≤ 2. Patients
the brain [15]. In addition, better than most protocols of who met any of the following conditions were excluded: (i)
BT, MT may have a priming effect on motor recovery the condition deteriorated during the intervention; that is,
through mirror illusion [14, 16]. Hence, compared to con- the stroke relapsed or a new infarction occurred; (ii) psychi-
ventional BTs, MT may be a superior approach for bilateral atric disorder or other serious illness that interfered with the
task relearning for stroke patients. Following the types of patients’ ability to obey the therapists; (iii) and having expe-
action, the protocols of MT contain manipulation of objects, rienced other central intervention methods, for instance,
manipulation without objects, and both in combination [17]. transcranial direct current stimulation (tDCS). The enrolled
However, relying entirely on a plane mirror or “mirror box,” patients were given written informed consent before the
the conventional protocol of MT cannot achieve it for isolat- study.
ing two hands from the mirror and only provides unilateral We speculated that the primary outcome (FMA-UL) had
visual feedback. Due to the limitations, manipulation of a group × time interaction. Based on the previous camera
objects under MT cannot attain bimanual cooperation and technique-based MVF studies [21, 25, 26], an effect size of
may affect the priming activation of mirror visual feedback 0.27 to 0.45 was expected to detect the differences in the
(MVF) [18–20]. Besides, the poor posture in the conven- improvements between groups. Given the reliability and
tional MT procedure can easily cause pressure on the spine safety margin, an effect size of 0.27 was anticipated for
and impede effective bimanual cooperation relearning [21]. repeated analysis of variance (ANOVA). Then, we estimated
To overcome the limitations of the traditional MT, a total sample size of 30 which was needed for providing
researchers have proposed novel mirror setups. Camera 80% power to detect the differences between groups on
technique-based MVF, which offered bilateral visual feed- FMA-UL with a type I error of 0.05 and a dropout rate of
back, was one of those, and previous researches have been 20%. In addition, we reviewed all published clinical trials
verified that camera technique-based MVF can promote on MT, and the sample size of most studies ranged from
the functional recovery of stroke [22–24]. 10 to 20 patients in each group. Therefore, we planned to
We previously put forward a novel camera technique- recruit 18 patients in each group. The process of recruiting
based MVF with an operable mirror environment [21]. patients is shown in Figure 1.
Patients can achieve synchronous movement of both upper
extremities isolated from the mirror in such an environment. 2.3. Randomization and Allocation. Eligible patients were
Previous studies have proven its clinical feasibility and effec- randomly assigned to the control group (CG) and experi-
tiveness for stroke rehabilitation [21, 25]. Based on the setup, mental group (EG). An independent researcher executed
we developed a novel MT paradigm, in which both upper the randomization procedure, generated through a random
extremities were associated with one object, and patients data generator on the computer. A sealed envelope was used
were asked to complete the same tasks to realize the associ- to confirm the group of each patient who was satisfied with
ation of both sides. In the paradigm of MT, we named it recruitment criteria. When receiving an envelope, therapists
associated mirror therapy (AMT). We conducted a random- were informed to perform the patient assignment.
ized controlled trial to certificate AMT’s feasibility and
clinical efficacy, and we hypothesized that AMT could pro- 2.4. Intervention. All enrolled patients received the conven-
mote the recovery of the paretic upper extremity and daily tional stroke rehabilitation program for four weeks, five day-
function for stroke patients. s/week, and around four hours/day. The conventional stroke
Neural Plasticity 3

Accessed for eligibility

Enrollment
(n = 210) Excluded (n = 24)
not meeting inclusion
criteria (n = 150)
Baseline clinical assessment (n = 36)

Allocation
Randomize (n = 36)

EG (n = 18) CG (n = 18)

Evaluation after 2-week Evaluation after 2-week

Intervention

treatment (n = 18) treatment (n = 18)

Evaluation after 4-week Evaluation after 4-week

treatment (n = 18) treatment (n = 18)

Figure 1: The flowchart of recruiting patients.

program consisted of physiotherapy, occupation therapy, In addition to the conventional rehabilitation, patients in
speech therapy, and respiratory management. EG received half an hour of AMT firstly. Based on the
patient’s condition, therapists selected 2 to 3 kinds of bilat-
2.4.1. Experimental Group eral cooperative tasks. Subsequently, another half-hour
upper limb training was applied, including stretching, relax-
(1) Setup. The setup (1200 mm × 940 mm × 702 mm) was ing, and functional activities.
mounted with a 23.8-inch light-emitting diode screen of
30° tilt, fixed on the mirror setup to present the mirror 2.4.2. Control Group. Patients in CG received the same dose
image, and blocked the direct view of both hands [21]. of training as EG. However, the only difference was that CG
Two cameras were mounted on the top of the mirror setup received bimanual cooperation training without camera
to capture the movements of the hands. In the mirrored technique-based MVF, where patients had a direct view of
environment, patients were allowed to put both hands on both arms. To assure the performance of bilateral coopera-
the bottom of the “mirror box”, of which one side opening tive tasks, therapists also provided necessary assistance to
was beneath the screen. The therapist could assist the help patients (see Figures 2(c) and 2(d)).
patients on the other side. Patients could sit in a suitable
and comfortable position by adjusting the setup height dur- 2.5. Assessments. The basic information, including age, sex,
ing the treatment. lesion side of the brain, stroke type, and duration after stroke
onset, was recorded. The clinical outcomes were concerned
(2) Associated Mirror Therapy (AMT). Based on the above with motor impairment, motor function, and daily function.
setup, we created a novel clinical paradigm of MT, in which The outcome measures were evaluated repeatedly before the
patients not only could see the regular bimanual cooperation intervention, after 2-week and 4-week intervention to verify
but also could attain the bilateral cooperative tasks with the clinical efficacy. The specific evaluation details of outcome
assistance of therapists. In the paradigm, both upper extrem- measures were as follows.
ities were associated with the identical object and completed The primary outcome was the change of motor impair-
synchronously the same task, e.g., holding a ball, grabbing ment measured through the Fugl-Meyer Assessment Upper
and rolling a cylinder, stacking of towels, lifting a stick, Limb subscale (FMA-UL). The FMA-UL subscale with good
and pushing a sanding board (see Figures 2(a) and 2(b)). psychometric properties indicated high reliability and valid-
Patients were required to focus on the screen and imagined ity for motor impairment [27]. The FMA-UL subscale
doing cooperative tasks with both arms. Meanwhile, patients included 33-item upper limb activities. Each item was rated
were asked to perform the same training synchronously by on a 0 to 2 ordinal scale. The maximum score of the FMA-
the affected side as much as possible. During the trial, ther- UL subscale was 66.
apists offered essential directions to make patients concen- Secondary outcomes were the performances of motor
trate on the screen and immerse themselves in mirror and daily function. The Box and Block Test (BBT) with sat-
illusion. Another role of therapists was to supervise and isfactory reliability and validity was used to assess motor
ensure the completion of actual bilateral cooperative tasks function for manual dexterity in stroke patients [28]. The
of patients. Conforming to the motion of the less affected BBT contains 150 colored wooden cube blocks (1 inch, 2:5
arm of the patient, the therapist could provide active, cm × 2:5 cm × 2:5 cm). The participants were told to move
assisted, or passive movement for the affected side alterna- one-by-one blocks as many as possible from a rectangular
tively. We named the novel paradigm “associated mirror box container to the other of equal size within 60 seconds.
therapy” (AMT) for achieving a practical bimanual interac- Both hands’ scores of the BBT were calculated, respectively,
tion under camera technique-based MVF. by the number of blocks transferred. The Functional
4 Neural Plasticity

(a) (b)

(c) (d)

Figure 2: AMT and conventional bimanual training for stroke rehabilitation: (a, b) AMT: grabbing and rolling a cylinder/holding a ball; (c,
d) conventional bimanual training: grabbing and rolling a cylinder/holding a ball.

Independent Measurement (FIM) was widely applied to each group. All patients completed the trial without side
evaluate participation after stroke [29]. FIM involved six effects, and adverse events occurred during the trial.
aspects of daily function: self-care, sphincter control, trans- The clinical characteristics of the two groups of patients
fer, locomotion, communication, and social cognition abil- were demonstrated in Table 1. The median age (QR) was
ity. It was made of 18 items, and each item was graded on 54.0 (24.00) and 58.0 (22.75) years for EG and CG, respec-
a 1 to 7 ordinal scale. The total score ranged from 7 to 126. tively (P = 0:350), with no difference in sex between groups.
No significant difference was found in the course of stroke
2.6. Statistical Analyses. The data were analyzed by using (P = 0:198). There were no differences in the type and
SPSS version 20.0 for Windows (SPSS Inc., Chicago, IL, location of stroke between EG and CG (P = 0:725, P =
USA). We used Shapiro-Wilk’s test to check the underlying 0:738). No differences were found between the two groups
model assumptions for normality of distribution entirely. in the Brunnstrom stages of the proximal and distal areas
None of evaluation indicators satisfied the normal distribu- of the affected upper extremity (P = 0:464, P = 0:876).
tion. Baseline characteristics of the patients between groups
were compared by using chi-square tests or Fisher’s exact
3.2. Treatment Effects on Clinical Outcomes. The improve-
test (including gender, side of paralysis, stroke types, and
ments of paretic arm impairment and daily function were
Brunnstrom stage) when appropriate. The Mann–Whitney
observed in both groups. Manual dexterity had a significant
U test was used to examine the baseline data of continuous
change in EG after the intervention, whereas the improve-
variables between groups (including age and course of the
ment did not occur in CG. Significant group-by-time inter-
disease). The generalized estimating equation (GEE) model
action effects were found in FAM-UL scores (Waldχ2 =
based on a binary outcome with first-order autoregressive
correlation structure (AR (1)) was used to explore multitime 174:434, P < 0:001), BBT scores (Waldχ2 = 18:594, P =
repeated measurement analysis [30, 31], including three out- 0:002), and FIM scores (Waldχ2 = 100:165, P < 0:001) after
come indicators (FMA-UL, FIM, and BBT). The main effects the intervention; therefore, the single group or time effect
of group, time, and group-by-time interaction were analyzed estimate was not applicable during the study (see Table 2).
in the GEE model. A value of P < 0:05 was considered The treatment effects of clinical outcomes are shown in
significant. Tables 3–5. The detailed comparisons between the two
groups were reported below.
3. Results After the 4-week trial, FMA-UL scores in both groups
were significantly higher than before (P < 0:001 and P <
3.1. Baseline Characteristics. From October 2018 to August 0:001, respectively). Both EG and CG had a continuous
2019, 36 stroke patients were recruited, with 18 patients in improvement in FMA-UL scores over time, including the
Neural Plasticity 5

Table 1: Characteristics of study participants (n = 36).

Variable EG (n = 18) CG (n = 18) P value

Age (years), M (QR) 54.0 (24.00) 58.0 (22.75) 0.350
Sex, N
Male/female 12/6 12/6 1.000
Lesion side, N
Left/right 9/9 10/8 0.738
Stroke type, N
Hemorrhagic/schemic 13/5 11/7 0.725
Months after stroke onset, M (QR) 4.0 (5.25) 5.0 (7.25) 0.198
Brunnstrom (3/4/5/6), N
Distal 12/2/2/2 14/1/1/2 0.876
Proximal 12/2/2/2 16/1/0/1 0.464
EG: experimental group; CG: conventional group.

Table 2: Description for group effect, time effect, and group × time effect on motor impairment, motor function, and daily function.

Group Time Group × time

Outcomes
Waldχ2 P Waldχ2 P Waldχ2 P
FAM-UL 4.858 0.028 141.058 <0.001 174.434 <0.001
FIM 3.893 0.048 58.687 <0.001 100.165 <0.001
BBT 0.192 0.662 17.310 <0.001 18.594 0.002
FAM-UL: Fugl-Meyer Assessment Upper Limb subscale; BBT: Box and Block Test; FIM: Functional Independence Measure.

Table 3: Description and comparison between groups for statistical outcomes on motor impairment, motor function, and daily function.

Pretest After 2 weeks After 4 weeks

Outcomes
EG CG P EG CG P EG CG P
FMA-UL 32.5 (25.50) 28.0 (11.00) 0.290 41.5 (13.25) 30.0 (11.75) 0.018 45.0 (22.50) 30.5 (13.50) 0.001
FIM 108.0 (8.00) 104.5 (15.00) 0.287 111.(8.50) 106.0 (15.50) 0.041 113.5 (8.50) 107.0 (14.50) 0.003
BBT 0.5 (12.00) 0.0 (3.00) 0.780 2.0 (21.50) 0.0 (3.25) 0.569 3.0 (24.00) 0.0 (6.25) 0.377
EG: experimental group; CG: conventional group; FAM-UL: Fugl-Meyer Assessment Upper Limb subscale; BBT: Box and Block Test; FIM: Functional
Independence Measure.

Table 4: Description for motor impairment, motor function, and daily function in EG.

Outcomes Pretest After 2 weeks After 4 weeks P

a b
FMA-UL 32.5 (25.50) 41.5 (23.25) 45.0 (22.50) <0.001
FIM 108.0 (8.00) 111.0 (8.50) 113.5 (8.50) <0.001
BBT 0.5 (12.00) 2.0 (21.50) 3.0 (24.00) <0.001
a
Comparison between pretest and after 2-week intervention. PFMA‐UL < 0:001, PFIM < 0:001, and PBBT = 0:002. bComparison between after 2-week intervention
and after 4-week intervention. PFMA‐UL < 0:001, PFIM < 0:001, and PBBT < 0:001. EG: experimental group; FAM-UL: Fugl-Meyer Assessment Upper Limb
subscale; BBT: Box and Block Test; FIM: Functional Independence Measure.

Table 5: Description for motor impairment, motor function, and daily function in CG.

Outcomes Pretest After 2 weeks After 4 weeks P

FMA-UL 28.0 (11.00) 30.0 (11.75)c 30.5 (13.50)d <0.001
FIM 104.5 (15.00) 106.0 (15.50) 107.0 (14.50) <0.001
BBT 0.0 (3.00) 0.0 (3.25) 0.0 (6.25) 0.107
c
Comparison between pretest and after 2-week intervention. PFMA‐UL < 0:001, PFIM < 0:001, and PBBT = 1. dComparison between after 2-week intervention and
after 4-week intervention. PFMA‐UL < 0:001, PFIM = 0:006, and PBBT = 0:043. CG: conventional group; FAM-UL: Fugl-Meyer Assessment Upper Limb subscale;
BBT: Box and Block Test; FIM: Functional Independence Measure.
6 Neural Plasticity

first two weeks (P < 0:001 and P < 0:001, respectively) and proving the feasibility of combining bilateral symmetrical
the last two weeks of intervention (P < 0:001 and P < 0:001 tasks with MT, patients were asked to concentrate on the
, respectively). Post hoc analyses indicated no difference in reflection side of the mirror and could not ensure the partic-
FAM-UL between EG and CG before the trial (P = 0:290). ipation of the paretic side. New setups for MT conquered the
Moreover, the scores of FMA-UL in the EG were signifi- limitations of the conventional mirror, for instance, the cam-
cantly higher than those in the CG after 2 and 4 weeks era technique-based MT proposed by Lee et al., which real-
(P = 0:018 and P = 0:001, respectively). ized MVF effect delay and bilateral movements [22, 24].
Significant improvements of BBT scores in the first and We previously put forward novel camera technique-based
the last two weeks were observed in EG (P = 0:002 and P < MVF [21]. To achieve bimanual coordination control under
0:001). After the intervention, BBT scores in the EG were the mirror, we designed bimanual cooperation tasks in
significantly improved (P < 0:001). However, in the CG, only which both arms were associated with one object and com-
in the last two weeks, the BBT scores were significantly pleted the same tasks synergistically.
improved (P = 0:043). After the 4-week intervention, no dif- Compared to usual care, camera technique-based MVF
ference in BBT scores was observed in the CG than before was proven to enhance the motor impairment of the upper
(P = 0:107). By comparing EG with CG, no differences in extremity after stroke [22, 25]. In line with the results of pre-
BBT scores were observed before the trial, after 2-week and vious studies, the improvements in motor impairment mea-
4-week intervention (P = 0:780, P = 0:569, and P = 0:377, sured by FAM-UL were observed in both groups after the
respectively). Although the difference in manual dexterity intervention, and patients in the EG were improved more
measured by BBT scores was not significant between both significantly than the CG. Furthermore, compared with
groups, a clinical improvement is in favor of AMT. bilateral arm training, researchers also found a more signif-
After the 4-week trial, FIM scores in both groups were icant improvement for the distal arm which was in favor of
significantly higher than before (P < 0:001 and P < 0:001, MT [39]. Our results were similar to the above study. Previ-
respectively). A significant improvement of FIM scores in ous studies revealed that MVF might have the potential to
the first and the last two weeks was observed in the EG promote motor learning by activating neural areas related
(P < 0:001 and P < 0:001, respectively). FIM scores in the to spatial attention, which was beneficial to enhance the per-
CG were also improved in the first and the last two weeks ception of the paretic arm [40, 41]. Therefore, compared to
(P < 0:001 and P = 0:006, respectively). When compared conventional BT, the result might be interpreted that MVF
between groups, no difference in FIM scores was observed activated the related sensorimotor brain area through visual
before the trial (P = 0:287). However, improvement of FIM illusion. Previously, Rodrigues et al. put forward adding an
scores in the EG was better than the CG after 2 and 4 weeks object to the plane mirror to realize bilateral symmetrical
(P = 0:041 and P = 0:003, respectively). training under MT [38]. Researchers discovered that no dif-
ferences were found between bilateral symmetrical tasks
4. Discussion with or without MT. Our results were different from it.
Noticeably, the MT paradigm in our study was different
In the present study, we proposed a novel paradigm of MT, from the one Rodrigues et al. proposed. The main difference
called AMT, which achieved bimanual cooperation under was that we used camera technique-based MVF rather than
camera technique-based MVF. Besides, we testified to the a plane mirror. In addition, in the present study, patients
feasibility and effectiveness of AMT. All patients participated could accomplish practical bimanual tasks under the thera-
in trials throughout without adverse events or side effects, pist’s assistance, but the paretic upper extremity’s quality
proving that the AMT was safe and feasible. The study dem- could not be guaranteed in the above conventional MT
onstrated that using AMT as an auxiliary therapy to usual paradigm.
care could decrease the motor impairment of the paretic When it came to motor function, one previous study
upper extremity and enhance daily function for stroke pointed out that MT could promote the manual dexterity
patients. In addition, AMT may increase manual dexterity of stroke patients evaluated by BBT [42]. Our finding was
after stroke. similar to the study. In the comparison before and after
The coordinated control was also regarded as an essen- the intervention, there was no difference of manual dexterity
tial function of the standard upper extremities, especially in CG. However, a significant effect of gross hand dexterity
for instrumental activities of daily living (IADL) [32–34]. measured by BBT was found in AMT after the intervention.
After stroke, patients lacked the participation of the paretic Although the difference in motor function improvements
upper extremity in daily tasks [35]. Previous studies com- between both groups was not statistically significant, the per-
bined MT with daily functional activities, demonstrating formance was better in AMT. Besides, interestingly, a sus-
that the MT paradigms could enhance the motor recovery tained motor function improvement of the upper extremity
of the paretic upper extremity in stroke patients [36, 37]. changed with time in EG, but not in CT. This phenomenon
Although many protocols of MT were proposed, few of might be caused where AMT had a sustained regular visual
those achieved practical bimanual tasks to associate both input of movement, which may better promote central brain
upper extremities. Rodrigues et al. developed the bilateral remodelling [15]. Our previous studies revealed that camera
task-based MT, which related both arms to one object under technique-based MVF could activate motor preparation and
the mirror environment [38]. As far as we know, this was the brain network segregation by inducing mirror illusion, which
first study to propose bilateral tasks based on MT. Despite might promote motor execution for stroke patients [21, 25].
Neural Plasticity 7

In this study, we used bilateral cooperation tasks with or agreed to its submission. Jin-Yang Zhuang and Li Ding con-
without camera technique-based MVF for stroke patients. tributed equally to this work.
All patients had gained significant improvement in daily
function after the trial, in line with previous studies of BT Acknowledgments
or MT [43, 44]. In addition, previous studies showed that
MT was more effective than conventional methods in We would like to thank all the patients who participated in
improving the daily function of stroke patients [36, 45]. the study. This study was funded by the National Key
Our result was similar to the above conclusion. A more R&D Program of China (Grant Nos. 2018YFC2002300 and
significant improvement in daily function was observed in 2018YFC2002301), National Natural Science Foundation
AMT when comparing the differences between both groups. for Innovative Research Group Project of China (Grant
It might be related to the more significant improvement of No. 82021002), National Natural Science Foundation of
motor impairment of paretic arms in EG after the interven- China Major Research Program Integration Project (Grant
tion. The daily function is related closely to the upper No. 9194830003), National Natural Science Foundation of
extremity function, for instance, self-care. However, in the China (Grant No. 82002385), and Shanghai Sailing Program
present study, we did not compare the improvements (Grant No. 20YF1403400).
between both groups in different aspects of daily function
based on FIM. Then, the improvements of specific daily References
functions in AMT were unknown.
The study has some limitations. Firstly, the novel para- [1] C. O. Johnson, M. Nguyen, and G. A. Roth, “Global, regional,
digm relied on camera technique-based MVF which is and national burden of stroke, 1990-2016: a systematic analy-
labour intensive. Secondly, though we have estimated the sis for the Global Burden of Disease Study 2016,” The Lancet
sample size, the results should be considered cautiously Neurology, vol. 18, no. 5, pp. 439–458, 2019.
because of the small sample size. Thirdly, we only conduct [2] S. I. Hay, A. A. Abajobir, and K. H. Abate, “Global, regional,
the 4-week intervention without follow-up, and the long- and national disability-adjusted life-years (DALYs) for 333
diseases and injuries and healthy life expectancy (HALE) for
term and sustained intervention effects for stroke patients
195 countries and territories, 1990-2016: a systematic analysis
are unknown. Hence, we will search for a more convenient for the Global Burden of Disease Study 2016,” The Lancet,
and economical method of AMT by equipment upgrade. vol. 390, no. 10100, pp. 1260–1344, 2017.
In addition, a larger RCT should be conducted to certify fur- [3] M. Katan and A. Luft, Global burden of stroke. Seminars in
ther the long-term and sustained effects of AMT on stroke neurology, Thieme Medical Publishers, 2018.
patients. [4] P. Langhorne, F. Coupar, and A. Pollock, “Motor recovery
after stroke: a systematic review,” The Lancet Neurology,
5. Conclusions vol. 8, no. 8, pp. 741–754, 2009.
[5] J. Metrot, J. Froger, I. Hauret, D. Mottet, L. van Dokkum, and
In summary, this is the first study to propose a novel and I. Laffont, “Motor recovery of the ipsilesional upper limb in
advantageous MT paradigm achieving bimanual cooperation subacute stroke,” Archives of Physical Medicine and Rehabili-
under camera technique-based MVF. The present study tation, vol. 94, no. 11, pp. 2283–2290, 2013.
demonstrates that AMT is a feasible and effective method [6] R. Sainburg, D. Good, and A. Przybyla, “Bilateral synergy: a
to improve motor impairment of the paretic arm, enhance framework for post-stroke rehabilitation,” Journal of neurology
daily function, and may increase the ability of manual dexter- & translational neuroscience, vol. 1, no. 3, 2013.
ity after stroke. [7] C. Albert, M. D. Lo, P. D. Guarino, and L. G. Richards, “Robot-
assisted therapy for long-term upper-limb impairment after
stroke,” Th e new engl and journal o f medicine, vol. 362,
Data Availability pp. 1772–1783, 2010.
We have utilized an unauthorized translation of the Chinese [8] S. L. Wolf, C. J. Winstein, J. P. Miller et al., “Effect of
MMSE in the research, and we have received the permission constraint-induced movement therapy on upper extremity
function 3 to 9 months after stroke: the EXCITE randomized
of Psychological Assessment Resources (PAR). The study
clinical trial,” Jama, vol. 296, no. 17, pp. 2095–2104, 2006.
data are available from the corresponding author.
[9] A. Pollock, S. E. Farmer, M. C. Brady et al., “Interventions for
improving upper limb function after stroke,” The Cochrane
Conflicts of Interest Database of Systematic Reviews, vol. 11, 2014.
[10] C. M. Stinear, M. A. Petoe, S. Anwar, P. A. Barber, and W. D.
The authors declare no conflicts of interest. Byblow, “Bilateral priming accelerates recovery of upper limb
function after stroke: a randomized controlled trial,” Stroke,
Authors’ Contributions vol. 45, no. 1, pp. 205–210, 2014.
[11] A. Wolf, R. Scheiderer, N. Napolitan, C. Belden, L. Shaub, and
Jie Jia conceived the study design and provided the research M. Whitford, “Efficacy and task structure of bimanual training
theories to support the study. Jin-Yang Zhuang and Li Ding post stroke: a systematic review,” Topics in Stroke Rehabilita-
collected the data, wrote the manuscript, and conducted the tion, vol. 21, no. 3, pp. 181–196, 2014.
experiment. Bei-Bei Shu and Dan Chen edited the manu- [12] A. L. E. Q. van Delden, C. L. E. Peper, G. Kwakkel, and P. J.
script. All the authors have reviewed the manuscript and Beek, “A Systematic Review of Bilateral Upper Limb Training
8 Neural Plasticity

Devices for Poststroke Rehabilitation,” Stroke Research and [27] J. Sanford, J. Moreland, L. R. Swanson, P. W. Stratford, and
Treatment, vol. 2012, 17 pages, 2012. C. Gowland, “Reliability of the Fugl-Meyer assessment for test-
[13] M. A. R. Y. E. L. L. E. N. S. T. O. Y. K. O. V. DMC, “A review of ing motor performance in patients following stroke,” Physical
bilateral training for upper extremity hemiparesis,” Occupa- Therapy, vol. 73, no. 7, pp. 447–454, 1993.
tional Therapy International, vol. 16, no. 3-4, 2009. [28] V. Mathiowetz, G. Volland, N. Kashman, and K. Weber,
[14] Y.-c. Li, C.-y. Wu, Y.-w. Hsieh et al., “The priming effects of “Adult norms for the Box and Block Test of manual dexterity,”
mirror visual feedback on bilateral task practice: a randomized American Journal of Occupational Therapy, vol. 39, no. 6,
controlled study,” Occupational Therapy International, pp. 386–391, 1985.
vol. 2019, 9 pages, 2019. [29] D. Chumney, K. Nollinger, K. Shesko, K. Skop, M. Spencer,
[15] C. H. Cheng, “Effects of observing normal and abnormal goal- and R. A. Newton, “Ability of Functional Independence Mea-
directed hand movements on somatosensory cortical activa- sure to accurately predict functional outcome of stroke-
tion,” European Journal of Neuroscience, vol. 47, no. 1, specific population: systematic review,” Journal of rehabilita-
pp. 48–57, 2018. tion research & development, vol. 47, no. 1, p. 17, 2010.
[16] M. E. Stoykov and S. Madhavan, “Motor priming in neuroreh- [30] S. L. Zeger and K.-Y. Liang, “Longitudinal data analysis for dis-
abilitation,” Journal of Neurologic Physical Therapy, vol. 39, crete and continuous outcomes,” Biometrics, vol. 42, no. 1,
no. 1, pp. 33–42, 2015. pp. 121–130, 1986.
[17] N. Morkisch, H. Thieme, and C. Dohle, “How to perform mir- [31] Y. Ma, M. Mazumdar, and S. G. Memtsoudis, “Beyond
ror therapy after stroke? Evidence from a meta-analysis,” Repeated-Measures analysis of Variance,” Regional Anesthesia
Restorative Neurology and Neuroscience, vol. 37, no. 5, and Pain Medicine, vol. 37, no. 1, pp. 99–105, 2012.
pp. 421–435, 2019. [32] V. Krishnan and S. Jaric, “Effects of task complexity on coordi-
[18] K.-B. Lim, H.-J. Lee, J. Yoo, H.-J. Yun, and H.-J. Hwang, “Effi- nation of inter-limb and within-limb forces in static bimanual
cacy of mirror therapy containing functional tasks in post- manipulation,” Motor Control, vol. 14, no. 4, pp. 528–544,
stroke patients,” Annals of Rehabilitation Medicine, vol. 40, 2010.
no. 4, pp. 629–636, 2016. [33] J. H. Cauraugh and S. Kim, “Two coupled motor recovery pro-
[19] K. N. Arya, S. Pandian, D. Kumar, and V. Puri, “Task-based tocols are better than One,” Stroke, vol. 33, no. 6, pp. 1589–
mirror therapy augmenting motor recovery in poststroke 1594, 2002.
hemiparesis: a randomized controlled trial,” Journal of Stroke [34] J. Jia, “Implication of left-right coordination and counterbal-
and Cerebrovascular Diseases, vol. 24, no. 8, pp. 1738–1748, ance for rehabilitation of upper limbs and hand function post
2015. stroke (review),” Chinese Journal of Rehabilitation Theory and
[20] R. W. Selles, M. E. Michielsen, J. B. J. Bussmann et al., “Effects Practice, vol. 24, no. 12, pp. 1365–1369, 2018.
of a mirror-induced visual illusion on a reaching task in stroke [35] L. E. H. van Dokkum, E. le Bars, D. Mottet, A. Bonafé, N. M. de
patients: implications for mirror therapy training,” Neuroreh- Champfleur, and I. Laffont, “Modified brain activations of the
abilitation and Neural Repair, vol. 28, no. 7, pp. 652–659, 2014. nondamaged hemisphere during ipsilesional upper-limb
[21] L. Ding, X. Wang, X. Guo et al., “Camera-based mirror visual movement in persons with initial severe motor deficits post-
feedback: potential to improve motor preparation in stroke stroke,” Neurorehabilitation and Neural Repair, vol. 32,
patients,” IEEE Transactions on Neural Systems and Rehabili- no. 1, pp. 34–45, 2018.
tation Engineering, vol. 26, no. 9, pp. 1897–1905, 2018. [36] K. Kim, S. Lee, D. Kim, K. Lee, and Y. Kim, “Effects of mirror
[22] C.-S. Chang, Y.-Y. Lo, C.-L. Chen, H.-M. Lee, W.-C. Chiang, therapy combined with motor tasks on upper extremity func-
and P.-C. Li, “Alternative motor task-based pattern training tion and activities daily living of stroke patients,” Journal of
with a digital mirror therapy system enhances sensorimotor Physical Therapy Science, vol. 28, no. 2, pp. 483–487, 2016.
signal rhythms post-stroke,” Frontiers in Neurology, vol. 10, [37] K. N. Arya, S. Pandian, and D. Kumar, “Task-based mirror
2019. therapy enhances ipsilesional motor functions in stroke: a pilot
[23] S. Hoermann, L. F. dos Santos, N. Morkisch et al., “Compu- study,” Journal of Bodywork and Movement Therapies, vol. 21,
terised mirror therapy with augmented reflection technology no. 2, pp. 334–341, 2017.
for early stroke rehabilitation: clinical feasibility and integra- [38] L. C. Rodrigues, N. C. Farias, R. P. Gomes, and S. M. Michael-
tion as an adjunct therapy,” Disability and Rehabilitation, sen, “Feasibility and effectiveness of adding object-related
vol. 39, no. 15, pp. 1503–1514, 2017. bilateral symmetrical training to mirror therapy in chronic
[24] H.-M. Lee, P.-C. Li, and S.-C. Fan, “Delayed mirror visual stroke: a randomized controlled pilot study,” Physiotherapy
feedback presented using a novel mirror therapy system Theory and Practice, vol. 32, no. 2, pp. 83–91, 2016.
enhances cortical activation in healthy adults,” Journal of Neu- [39] A. Samaie and K. M. Amoozadeh, “A comparetive study on the
roengineering and Rehabilitation, vol. 12, no. 1, 2015. effects of mirror therapy and bilateral arm training on hand
[25] L. Ding, X. Wang, S. Chen et al., “Camera-based mirror visual function of chronichemiparetic patients,” Koomesh, vol. 17,
input for priming promotes motor recovery, daily function, no. 3, pp. 589–595, 2016.
and brain network segregation in subacute stroke patients,” [40] F. J. A. Deconinck, A. R. P. Smorenburg, A. Benham,
Neurorehabilitation and Neural Repair, vol. 33, no. 4, A. Ledebt, M. G. Feltham, and G. J. P. Savelsbergh, “Reflections
pp. 307–318, 2019. on mirror therapy: a systematic review of the effect of mirror
[26] L. Ding, X. Wang, X. Guo et al., “Effects of camera-based visual feedback on the brain,” Neurorehabilitation and Neural
mirror visual feedback therapy for patients who had a stroke Repair, vol. 29, no. 4, pp. 349–361, 2015.
and the neural mechanisms involved: protocol of a multicentre [41] C.-H. Cheng, S.-H. Lin, C.-Y. Wu, Y.-H. Liao, K.-C. Chang,
randomised control study,” BMJ Open, vol. 9, no. 3, and Y.-W. Hsieh, “Mirror illusion modulates M1 activities
p. e022828, 2019. and functional connectivity patterns of perceptual-attention
Neural Plasticity 9

circuits during bimanual movements: a magnetoencephalog-

raphy study,” Frontiers in Neuroscience, vol. 13, 2020.
[42] S. Samuelkamaleshkumar, S. Reethajanetsureka,
P. Pauljebaraj, B. Benshamir, S. M. Padankatti, and J. A. David,
“Mirror therapy enhances motor performance in the paretic
upper limb after stroke: a pilot randomized controlled trial,”
Archives of Physical Medicine and Rehabilitation, vol. 95,
no. 11, pp. 2000–2005, 2014.
[43] J.-H. L. Min-Jae Lee, H.-M. Koo, and S.-M. Lee, “Effectiveness
of bilateral arm training for improving extremity function and
activities of daily living performance in hemiplegic patients,”
Journal of Stroke and Cerebrovascular Diseases, vol. 26, 2017.
[44] G. Yavuzer, R. Selles, N. Sezer et al., “Mirror therapy improves
hand function in subacute stroke: a randomized controlled
trial,” Archives of Physical Medicine and Rehabilitation,
vol. 89, no. 3, pp. 393–398, 2008.
[45] Y. Park, M. Chang, K.-M. Kim, and D.-H. An, “The effects of
mirror therapy with tasks on upper extremity function and
self-care in stroke patients,” Journal of Physical Therapy Sci-
ence, vol. 27, no. 5, pp. 1499–1501, 2015.