The purpose of this project is to examine and compare the immediate and long-term effects of combined Botulinum toxin type A (BoNT-A) injection between condensed and distributed robot-assisted training (RT) programs in patients with spastic hemiplegic stroke. Spasticity, a common impairment after stroke, has a profound impact on activity and participation for patients. According to the result of the investigator's ongoing study, the investigators found BoNT-A injection combined with robot-assisted training is recommended to enhance functional recovery for patients with spastic hemiplegic stroke. However, the optimal program as considering the RT frequency is unknown.

The aims of this study are to determine and compare the immediate and longer-term effects between condensed and distributed programs of RT following BoNT-A injection in subjects with spastic hemiplegic stroke .

Participants with chronic spastic hemiplegic stroke will be recruited and randomly assigned to either condensed or distributed RT groups post BoNT-A injection. Each training session included 40 minutes RT, followed by 40-minute functional training. The condensed group will receive 4 sessions per week, for 6 weeks, the distributed group 2 sessions per week, for 12 weeks. Body function and structures outcome measures include Fugl-Meyer Assessment, Modified Ashworth Scale. Activity and participation measures include Wolf Motor Function Test, Motor Activity Log, and Canadian Occupational Performance Measure. In addition, to directly reflect a patient's unique needs and goals, Goal Attainment Scaling will be assessed. Evaluators will be blind to group allocation. The outcome will be measured at pre-treatment, post-treatment, and 6-week follow-up. The investigators will also use the movement time of robot without powered assistance and surface EMG to determine the motor learning processes of patients receiving the two practice frequencies of RT.

This comparative efficacy study will be the first to examine and compare the motor learning processes and immediate and long-term effects between condensed and distributed RT post BoNT injection. The results may provide clinicians with the appropriate methods to scheduling RT following BoNT-A injection to improve upper limb functions for patients with hemiplegic spasticity stroke.

Patients with spasticity are thought to have greater levels of disability, poorer quality of life, and more caregiver burden than patients without spasticity. Thus, the treatment of spasticity, particularly of the upper limb, is one of the important issues in post-stroke rehabilitation. Botulinum toxin A (BoNT-A) injection becomes the most widely used for managing focal spasticity.(de Paiva, Meunier, Molgo, Aoki, & Dolly, 1999) There was a large body of evidence supporting the efficacy of BoNT-A on post-stroke spasticity (Bhakta, Cozens, Chamberlain, & Bamford, 2000; Elia, Filippini, Calandrella, & Albanese, 2009). However, the effect of BoNT on active function remains uncertain (Foley et al., 2013; Sheean, 2001). Patients with spasticity might use abnormal motor program to perform their movements. Once the spasticity is reduced by BoNT-A, the established abnormal program could no longer direct the movement execution, resulting in difficulty in performing movement or task. Evidence-based behavioral approaches with accurate sensorimotor input post BoNT-A injection programs might help patients relearn motor control and reestablish correct motor program, and, thus, optimize spasticity treatment outcomes. Robotic therapy (RT) has been proved effective in patients with moderate and severe arm weakness after stroke. RT is a novel approach to stroke rehabilitation that uses intensive, repetitive, interactive, and individualized practice as an optimal strategy to enhance motor learning. (Mirelman A, Bonato P, Deutsch JE. Effects of training with a robot virtual reality system compared with a robot alone on the gait of individuals after stroke. Stroke. 2009;40:169 -174. 5. Brewer BR, McDowell SK, Worthen-Chaudhari LC. Poststroke upper extremity rehabilitation: a review of robotic systems and clinical results. Top Stroke Rehabil. 2007;14:22- 44). An increasing number of efficacy studies have shown that RT had positive impacts on upper limbs motor function and strength in stroke survivors (Hsieh et al., 2016; Hung et al., 2016; Kwakkel, Kollen, & Krebs, 2008; Mehrholz, Platz, Kugler, & Pohl, 2009; Prange, Jannink, Groothuis-Oudshoorn, Hermens, & IJzerman, 2006). One study that investigated the effectiveness of BoNT-A injection combined with RT (Pennati, Da Re, Messineo, & Bonaiuti, 2015) showed that the hybrid approach reduces spasticity and improved motor performance in patients with chronic stroke. The key element leading to effective intervention for RT is the high-intensity, repetitive, task-specific practices. As well as treatment intensity, the scheduling of training is also shown to influence skills acquisition. Massed practice refers to a practice schedule that allows no periods of rest in between practice, whereas distributed practice refers to a schedule where periods of practice are interspersed with periods of rest (Verdaasdonk, Stassen, Van Wijk, & Dankelman, 2007). Distributed practice has been demonstrated to benefit complex psychomotor skills acquisition in motor learning,(Mackay, Morgan, Datta, Chang, & Darzi, 2002) including skills transfer. In order to develop the optimal RT schedule following BoNT-A injection, we should consider the duration of BoNT-A effect. It was known that the antispastic effect of one BoNT-A injection usually lasted for 3-4 months, and the effect tended to decrease since 2 months post-injection. When provided in a fixed dosage, we did not know should we condense the training program within 2 months post-injection, when there was great effect of BoNT-A, or distribute the training program within 4 months to across the whole effective period of BoNT-A. Therefore, the aims of this study are to determine and compare the immediate / maintained effects between condensed and distributed training programs of RT following BoNT-A injection in subjects with spastic hemiplegic stroke.

Participants with stroke will be recruited from a rehabilitation department of tertiary referral hospital. Patients who have spastic hemiplegic stroke are screened for eligibility to join the study. The inclusion criteria are: (1) clinical and imagine diagnosis of a first or recurrent unilateral stroke ≥ 6 months; (2) upper limb spasticity (modified Ashworth scale of ≥ 1+ for elbow flexor and/or forearm pronator and/or finger flexor and/or wrist flexor muscles (Bohannon & Smith, 1987); (3) initial motor part of UE of FMA score ranging from 13 to 56, indicating moderate to severe movement impairment (Duncan, Goldstein, Matchar, Divine, & Feussner, 1992; Fugl Meyer, Jaasko, & Leyman, 1975; Park, Wolf, Blanton, Winstein, & Nichols-Larsen, 2008); (4) no serious cognitive impairment (i.e., Mini Mental State Exam score > 20) (Teng & Chui, 1987); (5) age ≥ 18 years ; and (6) willing to provide written informed consent. Participants are excluded if they are pregnant, or with bilateral hemispheric or cerebellar lesions, sever aphasia, significant visual field deficits or hemineglect, contraindication for BoNT-A injection, treatment with BoNT-A within 4 months before recruitment, any fixed joint contracture of the affected upper limb, or a history of orthopedic or other neurological diseases and/or medical conditions that would prevent adherence to the rehabilitation protocol. To minimize possible confounding effects of upper limb motor ability, we will stratify participants into groups based on upper limb motor function (FMA UE score: 13-40 or a FMA UE score: 41-56) (Luft et al., 2004; Thielman, Kaminski, & Gentile, 2008). Intervention BoNT-A injections Injections are administered at baseline by 1 of 2 senior rehabilitation physicians who have more than 10 years of experience of BoNT-A injection. Doses and muscles selected for BoNT-A injection are individualized on the basis of the spasticity patterns, severity of spasticity, and treatment goals. Botox brand BoNT-A Purified Neurotoxin Complex, (Allergan Pharmaceuticals, Irvine, CA) is used in this study. Location of the targeted muscle is confirmed by using echo guide. After injection with BoNT-A, patients will receive different frequency schedule of treatment according to their group allocation. Participants in the condensed group receive 4 sessions per week for 6 weeks and participants in the distributed practice group attende 2 sessions per week for 12 weeks. During each session, participants receive 45 minutes of repetitive RT with the InMotion 2.0 robot (Interactive Motion Technologies Inc., Watertown, MA), followed by 30 minutes of transition-to-task practice. The outcome measures were in accordance with the International Classification of Functioning, Disability and Health (ICF) framework published by the World Health Organization World Health. Body function and structures outcome measures include Fugl-Meyer Assessment (FMA), Modified Ashworth Scale(Masiero et al., 2006). Activity and participation measures include Wolf Motor Function Test (WMFT), Motor Activity Log (MAL), Canadian Occupational Performance Measure (COPM) and Goal attainment scale (GAS). Evaluators are blind to group allocation. The outcomes are measured at 4 time points: before intervention, mid- intervention, post intervention, and 6 weeks after intervention, except the MAL which is measured before intervention, post intervention, and 6 weeks after intervention. Body function and structures measures Fugl-Meyer Assessment (FMA) The upper-extremity (UE) subscale of the FMA is used to assess neuromusculoskeletal and movement related functions. It consists of 33 upper extremity items for the reflexes and movement of shoulder, elbow, forearm, wrist, hand, and coordination/speed. They are scored on a 3-point ordinal scale (0-cannot perform, 1-performs partially, 2-performs fully)(Fugl Meyer et al., 1975). Higher score indicates better motor function of UE, and the maximum score is 66. Satisfactory psychometric properties of the FMA have been demonstrated(Platz et al., 2005) Modified Ashworth Scale (MAS) Spasticity of skeletal muscle in upper extremity is evaluated by using the MAS scale (Bohannon & Smith, 1987). It uses a 6-point scale to score the resistance to passive movement for each join with higher score indicating higher spasticity. The MAS has shown good reliability and validity (Pandyan et al., 1999). Activity and participation measures Wolf Motor Function Test (WMFT) The WMFT is a quantitative measure of upper extremity motor ability through timed and functional tasks (Uswatte, Taub, Stuss, Winocur, & Robertson, 1999). The WMFT includes 17 tasks (15 function-based and 2 strength-based). Performances were timed and rated by using a 6-point ordinal scale. The WMFT has good interrater reliability and criterion validity in patients with UE hemiparesis (Morris, Uswatte, Crago, Cook, & Taub, 2001) . Motor Activity Log (MAL) The MAL is a semi-structured interview to rate how well [quality of movement scale (QOM)] and how much [amount of use scale (AOU)] they use their affected upper extremity in 30 daily activities using a 6-point scale (Uswatte, Taub, Morris, Light, & Thompson, 2006). Higher scores represent better performance. The MAL has established reliability, validity, and responsiveness in patients with stroke(Uswatte et al., 2006; Uswatte, Taub, Morris, Vignolo, & McCulloch, 2005; Van der Lee, Beckerman, Knol, De Vet, & Bouter, 2004). Canadian Occupational Performance Measure (COPM) The COPM, a semi-structured interview, will be used to assess a participant's perception of the performance in the areas of self-care, productivity and leisure (Dedding, Cardol, Eyssen, & Beelen, 2004). The COPM can capture information about activities that the participant wants, needs or is expected to perform. The participant is asked to rate five most important activities on a 10-point scale for performance, ranging from 1 (not at all able) to 10 (able to perform extremely well), and for satisfaction, ranging from 1 (not at all satisfied) to 10 (extremely satisfied)(Dedding et al., 2004). The COPM is a reliable, valid, and responsive outcome measure for practitioners and researchers (Carswell et al., 2004). Goal attainment scale (GAS) The GAS is a useful measure of participants' individual goals that could be achieved in the course of intervention (L. Turner-Stokes, 2009). In the first session, the therapist will discuss with the participants and their families to define the goals. Each goal is rated on a 5-point scale ranging from -2 to +2 in the last session. The GAS enables the data to be placed on a quantitative measurement scale and can be helpful for qualitative interpretation (L. Turner-Stokes, 2009). Evidence of validity and responsiveness for the GAS was reported in previous studies(Gordon, Powell, & Rockwood, 1999; Rockwood, Stolee, & FoxP, 1993) Data Analysis Data are analyzed with PASW statistics 18 (SPSS Inc, Chicago, IL, USA). To examine the baseline differences among three groups, the chi-square test is used for categorical data and student t tests for continuous variables. We use mixed ANOVA to test the within subject effect of time (4 measurement time points: pre-test, mid-test, post-test, follow-up) and to test the between subject effect of group (two treatment groups: condensed treatment and distributed treatment) and to test interaction effect between time and group. Pair wise comparison with Bonferroni adjustments is used to examine the differences between measurement time points.

Inclusion Criteria:

1. clinical and imagine diagnosis of a first or recurrent unilateral stroke ≥ 3 months
2. upper limb spasticity (modified Ashworth scale of ≥ 1+ for elbow flexor and/or forearm pronator and/or finger flexor muscles and/or wrist flexor muscles (Bohannon & Smith, 1987)
3. initial motor part of UE of FMA score ranging from 17 to 56, indicating moderate to severe movement impairment (Duncan, Goldstein, Matchar, Divine, & Feussner, 1992; Fugl Meyer, Jaasko, & Leyman, 1975; Park, Wolf, Blanton, Winstein, & Nichols-Larsen, 2008)
4. no serious cognitive impairment (i.e., Mini Mental State Exam score > 20) (Teng & Chui, 1987)
5. age ≥ 18 years
6. willing to provide written informed consent

Exclusion Criteria:

1. pregnant
2. with bilateral hemispheric or cerebellar lesions
3. sever aphasia
4. significant visual field deficits or hemineglect
5. contraindication for BoNT-A injection
6. treatment with BoNT-A within 4 months before recruitment
7. any fixed joint contracture of the affected upper limb
8. a history of orthopedic or other neurological diseases and/or medical conditions that would prevent adherence to the rehabilitation protocol.

All-Cause Mortality, Serious, and Other [Not Including Serious] Adverse Events were not monitored/assessed

All-Cause Mortality, Serious, and Other [Not Including Serious] Adverse Events were not monitored/assessed