VA ROBOTICS: Robotic Assisted Upper-Limb Neurorehabilitation in Stroke Patients

• Sponsor: US Department of Veterans Affairs (U.S. Fed)
• Overall Status: Completed
• CT.gov ID: NCT00372411
• Collaborator: Burke Medical Research Institute (Other)
• Enrollment: 127
• Locations: 5
• Arms: 3
• Duration (Months): 44
• Patients Per Site: 25.4
• Patients Per Site Per Month: 0.6

Study Details

Study Description

Brief Summary

This study will compare robotic training with usual care and intensive comparison therapy to attempt to improve upper extremity function.

Condition or Disease	Intervention/Treatment	Phase
Stroke	Device: Robot-Assisted Therapy - MIT-MANUS System; Other: Intensive Comparison Therapy; Other: Usual Care	Phase 2

Detailed Description

Primary Hypothesis: The primary hypothesis is that robotic training compared with usual care and intensive comparison therapy will lead to improved upper extremity function. Eligible patients will be randomized to one of three treatment arms: 1) usual care, 2) intensive comparison therapy, or 3) robotic training. Participants assigned to intensive comparison therapy or robot-assisted training will receive treatment for 12 weeks consisting of three one-hour sessions a week and then usual care for the remainder of follow-up. Those assigned to usual care will receive the usual stroke care as delivered at each participating medical center for 36 weeks and then will be offered, as compassionate care, their choice of either robot-assisted or intensive comparison therapy.

Secondary Hypothesis: The secondary hypotheses are that compared with usual care and intensive comparison therapy, robotic treatment will lead to improved quality of life and task performance involving proximal and distal control of the paretic arm. If the robotic arm is effective, two other secondary objects are to evaluate its early (less than 12 week) and late (36 week) effects on the primary and secondary outcomes.

Primary Outcome: The primary study outcome is the change in the Fugl-Meyer score of neurological impairment at 12 weeks relative to baseline.

Secondary Outcome: Secondary outcomes include the Stroke Impact Scale and Wolf Motor Function Test. A cost-effectiveness analysis is planned but only will be conducted if robotic training is more effective than usual care.

Study Abstract: CSP #558 will be a randomized, multi-center, outpatient phase II trial to assess the safety and efficacy of robot-assisted therapy for neurorehabilitation in stroke patients with moderate to severe upper extremity impairment.

The target sample is 158 patients: 26 usual care, 66 intensive comparison therapy and 66 robot training. This sample size will provide 90% power to detect a 5-point mean difference in the Fugl-Meyer scale between robot training and usual care and 3-point mean difference between robot training and intensive comparison therapy. There will be one interim analysis of the primary endpoint at 12 months for the purpose of sample size re-estimation using an adaptive design. The planned study duration is 33 months with 24 months of intake and nine months of follow-up.

IMPACT STATEMENT

If robotic training proves to be beneficial it will not only provide a functionally-oriented and neurophysiologically appropriate therapy, but also will make more widely available high-quality, evidence-based rehabilitative care at a time when there is a shortage of experienced therapists and a progressively growing rehabilitative need for veterans and all Americans.

Study Design

• Study Type: Interventional
• Actual Enrollment: 127 participants
• Allocation: Randomized
• Intervention Model: Parallel Assignment
• Masking: Single (Outcomes Assessor)
• Primary Purpose: Treatment
• Official Title: CSP #558 - Robotic Assisted Upper-Limb Neurorehabilitation in Stroke Patients
• Study Start Date: Dec 1, 2005
• Actual Primary Completion Date: Jul 1, 2009
• Actual Study Completion Date: Aug 1, 2009

Arms and Interventions

Arm	Intervention/Treatment
Experimental: Arm 1 Robot-Assisted Therapy - MIT-MANUS System	Device: Robot-Assisted Therapy - MIT-MANUS System; The MIT-MANUS robot consists of four modules to train the entire upper limb: module A: shoulder-elbow; module B: anti-gravity; module C: wrist, and module D: hand-unit. Training will be given for 12 weeks and is divided into 4 consecutive blocks, with 9 training sessions per block.
Active Comparator: Arm 2 Intensive Comparison Therapy	Other: Intensive Comparison Therapy; The intensive comparison therapy protocol being used in CSP#558 was developed and field-tested at the Burke Rehabilitation Hospital. It exposes the patient to the identical number of treatments, time, and intensity that robot treated patients will receive (12 weeks, 3 times per week). Therapy sessions can be conducted on back-to-back days if needed and on more than 3 days a week (with approval from the Study Chairman) over a treatment period of up to 14 weeks in order to complete the 36 treatment sessions. During each 1-hour session, subjects participate in therapy in four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques (NDT)/Bobath arm training (Bobath, 1979).; Other Names: ICT
Other: Arm 3 Usual Care	Other: Usual Care; The usual chronic stroke care as delivered at each participating medical center

Outcome Measures

Primary Outcome Measures

1. Fugl-Meyer Assessment for Motor Recovery (FM) Scale [6, 12, 24 and 36 weeks minus baseline]

   Fugl-Meyer (FM) is a standard instrument for the quantitative clinical assessment of motor impairment and function. In this study the upper extremity subsection of the FM was used. The FM assesses several impairment dimensions by using a 3 point ordinal scale: 0 = cannot perform, 1 = can perform partially and 2 = can perform fully. These measures are summed to an overall score is Scoring for upper extremity FM ranges from 0 (worst, completely plegic) to 66 (best, normal). Higher scores indicate better functioning. Outcome measure is the change in the FM score at 6, 12, 24 and 36 weeks relative to baseline.

Secondary Outcome Measures

1. Stroke Impact Scale [6, 12, 24 and 36 weeks minus baseline]

   The Stroke Impact Scale (SIS) is stroke specific, self-reported measure that evaluates function and quality of life in eight clinically relevant domains. The domains of hand function, activities of daily living, instrumental activities of daily living, mobility, and social participation were used; total score ranges from 0 to 100 with higher values indicating better functioning. Outcome is change at 6, 12, 24 and 36 weeks relative to baseline.

2. Wolf Motor Function Test [6, 12, 24 and 36 weeks minus baseline]

   The Wolf Motor Function Test (WMFT) is a functionally-based test designed to provide an objective measure of both proximal (during tasks such as lifting the hand from table to box top) and distal control (grasping pencil, bringing soda can to mouth) of the paretic arm for patients after stroke or traumatic brain injury. The WMFT consists of 17 items, of which 15 measure time to perform functional tasks. The tasks are averaged to produce a score in seconds that ranges from 0 to 120 seconds, with higher scores indicating worse functioning. Outcome measure is the change in the Wolf score at 6, 12, 24 and 36 weeks relative to baseline.

3. Change in the Numeric Rating Scale (NRS) at 12 Weeks Relative to Baseline [12 weeks minus baseline]

   The Numeric Rating Scale (NRS) for pain is a self report scale ranging from 0 (no Pain) to 10 (pain as bad as you can imagine).

4. Change in the Modified Ashworth Scale for Spasticity at 12 Weeks Relative to Baseline [12 weeks minus baseline]

   The Modified Ashworth Scale for spasticity is a measurement of spasticity across 9 muscle groups. Each muscle group is scored on a 0 to 5 scale with higher scores indicating worse functioning. The total score is the average score from the 9 muscle groups and ranges from 0 to 5 with higher scores indicating worse functioning.

Eligibility Criteria

Criteria

• Ages Eligible for Study: 18 Years and Older
• Sexes Eligible for Study: All
• Accepts Healthy Volunteers: No

Inclusion Criteria:

• A single new focal unilateral stroke with diagnosis verified by brain imaging (MRI) or CT scan that has occurred at least 6 months prior to enrollment.

• Age 18 years of age and older.

• Upper extremity Fugl-Meyer score >=7 or <=38 (max 66). (These thresholds include neither hemiplegic nor fully recovered motor function in 14 muscles of the shoulder, elbow, and wrist.)

• Cognitive function sufficient to understand the experiments and follow instructions (Mini-Mental Status Score of 22 and higher or interview for aphasic subjects).

• Willingness to avoid any new alternative arm rehabilitative approaches for the duration of the study.

• Written informed consent to participate in the study.

Exclusion Criteria:

• A fixed contracture in the affected limb that prevents movement along the entire dynamic range required by either robot-assisted or intensive comparison therapy.

• Joint pain that occurs within the range of motion required by the intervention protocols (intensive comparison therapy and robot-assisted therapy).

• Enrolled in or planning to enroll in another interventional research trial that involves the upper limbs using procedures proposed to enhance or limit the function of the upper extremity (such as adjuvant rehabilitation or Botox injections) during the 36 weeks of participation.

• Prior experience with robot-assisted therapy or intensive comparison therapy for the upper limb as described in this specific study.

• A complicating medical condition that would prevent completion of the trial.

• Unable to complete the nine-month study, e.g., extended planned vacation.

• Any other known pre-existing bone diseases that might increase the risk of bone fracture or other injury from intensive comparison therapy or robot-assisted therapy.

Contacts and Locations

Locations

	Site	City	State	Country	Postal Code
1	VA Connecticut Health Care System (West Haven)	West Haven	Connecticut	United States	06516
2	North Florida/South Georgia Veterans Health System	Gainesville	Florida	United States	32608
3	VA Maryland Health Care System, Baltimore	Baltimore	Maryland	United States	21201
4	Providence VA Medical Center	Providence	Rhode Island	United States	02908-4799
5	VA Puget Sound Health Care System, Seattle	Seattle	Washington	United States	98108

Sponsors and Collaborators

• US Department of Veterans Affairs
• Burke Medical Research Institute

Investigators

• Study Chair: Albert Lo, MD PhD, Providence VA Medical Center

Study Documents (Full-Text)

None provided.

More Information

Publications

None provided.

• Responsible Party: US Department of Veterans Affairs
• ClinicalTrials.gov Identifier: NCT00372411
• Other Study ID Numbers: 558
• First Posted: Sep 7, 2006
• Last Update Posted: Jan 13, 2014
• Last Verified: Dec 1, 2013

• Keywords provided by US Department of Veterans Affairs: Stroke; Rehabilitation; Single-blind; Clinical Trial; Randomized controlled trial; Upper extremity; Robotics
• Additional relevant MeSH terms: Stroke

Study Results

Participant Flow

Recruitment Details	Between 11-8-06 and 10-31-08, 200 were screened and 127 randomized from 4 VA medical centers: Gainesville, West Haven, Baltimore, and Seattle. Usual care enrollment was stopped after 15 months when target information was attained per protocol. Recruitment to the robot-assisted and intensive comparison groups continued for 24 months.

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
Period Title: Overall Study
STARTED	49	50	28
COMPLETED	44	42	25
NOT COMPLETED	5	8	3

Baseline Characteristics

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care	Total
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.	Total of all reporting groups
Overall Participants	49	50	28	127
Age (years) [Mean (Standard Deviation) ]
Mean (Standard Deviation) [years]	66 (11)	64 (11)	63 (12)	65 (11)
Sex: Female, Male (Count of Participants)
Female	2 4.1%	2 4%	1 3.6%	5 3.9%
Male	47 95.9%	48 96%	27 96.4%	122 96.1%
Ethnicity (NIH/OMB) (Count of Participants)
Hispanic or Latino	1 2%	2 4%	0 0%	3 2.4%
Not Hispanic or Latino	48 98%	48 96%	28 100%	124 97.6%
Unknown or Not Reported	0 0%	0 0%	0 0%	0 0%
Race (NIH/OMB) (Count of Participants)
American Indian or Alaska Native	0 0%	0 0%	0 0%	0 0%
Asian	0 0%	1 2%	0 0%	1 0.8%
Native Hawaiian or Other Pacific Islander	0 0%	0 0%	0 0%	0 0%
Black or African American	10 20.4%	12 24%	5 17.9%	27 21.3%
White	38 77.6%	35 70%	23 82.1%	96 75.6%
More than one race	1 2%	2 4%	0 0%	3 2.4%
Unknown or Not Reported	0 0%	0 0%	0 0%	0 0%
Region of Enrollment (participants) [Number]
United States	49 100%	50 100%	28 100%	127 100%
Index Stroke Type (participants) [Number]
Hemorrhagic	7 14.3%	6 12%	6 21.4%	19 15%
Ischemic	42 85.7%	44 88%	22 78.6%	108 85%
Index Stroke Location (Number) [Number]
Anterior circulation ≥1/3 of hemisphere	6 12.2%	14 28%	6 21.4%	26 20.5%
Anterior circulation <1/3 of hemisphere	17 34.7%	21 42%	10 35.7%	48 37.8%
Small deep infarct	17 34.7%	15 30%	6 21.4%	38 29.9%
Posterior circulation	9 18.4%	0 0%	6 21.4%	15 11.8%
Time from Index Stroke to Randomization (Years) [Mean (Standard Deviation) ]
Mean (Standard Deviation) [Years]	3.6 (4.0)	4.8 (4.0)	6.2 (5.0)	4.7 (4.2)
Fugl-Meyer Assessment (units on a scale) [Mean (Standard Deviation) ]
Mean (Standard Deviation) [units on a scale]	19.7 (10.7)	17.3 (8.4)	20.3 (9.0)	18.9 (9.5)
Comorbidity Disease Index (participants) [Number]
≤1 Domain	7 14.3%	7 14%	2 7.1%	16 12.6%
2 Domains	13 26.5%	24 48%	8 28.6%	45 35.4%
≥3 Domains	29 59.2%	19 38%	18 64.3%	66 52%

Outcome Measures

1. Primary Outcome

Title	Fugl-Meyer Assessment for Motor Recovery (FM) Scale
Description	Fugl-Meyer (FM) is a standard instrument for the quantitative clinical assessment of motor impairment and function. In this study the upper extremity subsection of the FM was used. The FM assesses several impairment dimensions by using a 3 point ordinal scale: 0 = cannot perform, 1 = can perform partially and 2 = can perform fully. These measures are summed to an overall score is Scoring for upper extremity FM ranges from 0 (worst, completely plegic) to 66 (best, normal). Higher scores indicate better functioning. Outcome measure is the change in the FM score at 6, 12, 24 and 36 weeks relative to baseline.
Time Frame	6, 12, 24 and 36 weeks minus baseline

Outcome Measure Data

Analysis Population Description

All participants with baseline and follow-up measures were included by intention-to-treat

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
Measure Participants	47	46	27
6 weeks minus baseline	3.01 (0.92)	3.19 (0.91)	-2.08 (1.08)
12 weeks minus baseline	3.87 (1.05)	4.01 (1.06)	-1.06 (1.00)
24 weeks minus baseline	4.22 (1.05)	2.75 (1.06)	-2.02 (1.14)
36 weeks minus baseline	5.07 (1.18)	2.81 (1.18)	-0.53 (1.19)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol, comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.08
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses
	Method	ANCOVA
	Comments	Analysis is of change at 12 weeks minus baseline adjusted for the study site as a fixed effect, the Comorbidity Disease Index, and baseline FM value.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	2.17
	Confidence Interval	(2-Sided) 95%; -0.23 to 4.58
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 2

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.02
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	Mixed Models Analysis
	Comments	Secondary analysis used longitudinal methods to assess the effect of treatment on outcomes over 36 weeks with visits at 6, 12, 24, and 36 weeks.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	2.88
	Confidence Interval	(2-Sided) 95%; 0.57 to 5.18
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 3

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.92
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses
	Method	ANCOVA
	Comments	Analysis is of change at 12 weeks minus baseline adjusted for the study site as a fixed effect, the Comorbidity Disease Index, and baseline FM value.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-0.14
	Confidence Interval	(2-Sided) 95%; -2.94 to 2.65
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 4

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.63
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	Mixed Models Analysis
	Comments	Secondary analysis used longitudinal methods to assess the effect of treatment on outcomes over 36 weeks with visits at 6, 12, 24, and 36 weeks.
Method of Estimation	Estimation Parameter	Median Difference (Net)
	Estimated Value	-0.58
	Confidence Interval	(2-Sided) 95%; -2.97 to 1.81
	Parameter Dispersion	Type: Value:
	Estimation Comments

2. Secondary Outcome

Title	Stroke Impact Scale
Description	The Stroke Impact Scale (SIS) is stroke specific, self-reported measure that evaluates function and quality of life in eight clinically relevant domains. The domains of hand function, activities of daily living, instrumental activities of daily living, mobility, and social participation were used; total score ranges from 0 to 100 with higher values indicating better functioning. Outcome is change at 6, 12, 24 and 36 weeks relative to baseline.
Time Frame	6, 12, 24 and 36 weeks minus baseline

Outcome Measure Data

Analysis Population Description

[Not Specified]

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
Measure Participants	47	46	27
6 weeks minus baseline	5.49 (1.61)	4.08 (1.57)	-3.79 (2.47)
12 weeks minus baseline	6.31 (1.68)	5.77 (1.67)	-3.03 (2.34)
24 weeks minus baseline	5.58 (1.68)	4.23 (1.69)	-0.26 (2.37)
36 weeks minus baseline	5.14 (1.89)	4.97 (1.88)	0.76 (2.49)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol, comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.009
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses
	Method	ANCOVA
	Comments	Analysis is change at 12 weeks minus baseline, adjusted for study site as a fixed effect, Comorbidity Disease Index, and the baseline SIS value.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	7.64
	Confidence Interval	(2-Sided) 95%; 2.03 to 13.24
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 2

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.04
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	Mixed Models Analysis
	Comments	Secondary analysis used longitudinal methods to assess the effect of treatment on outcomes over 36 weeks with visits at 6, 12, 24, and 36 weeks.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	5.95
	Confidence Interval	(2-Sided) 95%; 0.34 to 11.56
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 3

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.81
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis is change at 12 weeks minus baseline, adjusted for study site as a fixed effect, Comorbidity Disease Index, and the baseline SIS value.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	0.54
	Confidence Interval	(2-Sided) 95%; -3.87 to 4.94
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 4

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.55
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	Mixed Models Analysis
	Comments	Secondary analysis used longitudinal methods to assess the effect of treatment on outcomes over 36 weeks with visits at 6, 12, 24, and 36 weeks.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	1.19
	Confidence Interval	(2-Sided) 95%; -2.74 to 5.12
	Parameter Dispersion	Type: Value:
	Estimation Comments

3. Secondary Outcome

Title	Wolf Motor Function Test
Description	The Wolf Motor Function Test (WMFT) is a functionally-based test designed to provide an objective measure of both proximal (during tasks such as lifting the hand from table to box top) and distal control (grasping pencil, bringing soda can to mouth) of the paretic arm for patients after stroke or traumatic brain injury. The WMFT consists of 17 items, of which 15 measure time to perform functional tasks. The tasks are averaged to produce a score in seconds that ranges from 0 to 120 seconds, with higher scores indicating worse functioning. Outcome measure is the change in the Wolf score at 6, 12, 24 and 36 weeks relative to baseline.
Time Frame	6, 12, 24 and 36 weeks minus baseline

Outcome Measure Data

Analysis Population Description

[Not Specified]

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
Measure Participants	47	46	27
6 weeks minus baseline	-3.98 (2.72)	-3.24 (2.64)	7.38 (2.53)
12 week minus baseline	-3.96 (3.00)	-4.89 (3.00)	7.54 (2.97)
24 weeks minus baseline	-5.97 (2.77)	-3.08 (2.74)	8.59 (3.10)
36 weeks minus baseline	-6.23 (2.83)	-2.66 (2.79)	7.30 (2.44)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.22
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis is change at 12 weeks minus baseline adjusted for the study site as a fixed effect, the Comorbidity Disease Index, and baseline WMFT value.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-4.41
	Confidence Interval	(2-Sided) 95%; -11.52 to 2.70
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 2

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.005
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	Mixed Models Analysis
	Comments	Secondary analysis used longitudinal methods to assess the effect of treatment on outcomes over 36 weeks with visits at 6, 12, 24, and 36 weeks.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-8.10
	Confidence Interval	(2-Sided) 95%; -13.61 to -2.60
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 3

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.82
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis is change at 12 weeks minus baseline adjusted for the study site as a fixed effect, the Comorbidity Disease Index, and baseline WMFT value.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	0.93
	Confidence Interval	(2-Sided) 95%; -7.03 to 8.89
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 4

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.55
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	Mixed Models Analysis
	Comments	Secondary analysis used longitudinal methods to assess the effect of treatment on outcomes over 36 weeks with visits at 6, 12, 24, and 36 weeks.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-2.13
	Confidence Interval	(2-Sided) 95%; -9.20 to 4.93
	Parameter Dispersion	Type: Value:
	Estimation Comments

4. Secondary Outcome

Title	Change in the Numeric Rating Scale (NRS) at 12 Weeks Relative to Baseline
Description	The Numeric Rating Scale (NRS) for pain is a self report scale ranging from 0 (no Pain) to 10 (pain as bad as you can imagine).
Time Frame	12 weeks minus baseline

Outcome Measure Data

Analysis Population Description

[Not Specified]

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
Measure Participants	47	46	27
Least Squares Mean (Standard Error) [units on a scale]	-0.61 (0.29)	0.24 (0.30)	0 (0.38)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.08
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis of covariance at 12 weeks, adjusted for study site as a fixed effect, Comorbidity Disease Index, and the baseline value of the outcome.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-0.81
	Confidence Interval	(2-Sided) 95%; -1.73 to 0.11
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 2

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.03
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis of covariance at 12 weeks, adjusted for study site as a fixed effect, Comorbidity Disease Index, and the baseline value of the outcome.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-0.84
	Confidence Interval	(2-Sided) 95%; -1.62 to -0.06
	Parameter Dispersion	Type: Value:
	Estimation Comments

5. Secondary Outcome

Title	Change in the Modified Ashworth Scale for Spasticity at 12 Weeks Relative to Baseline
Description	The Modified Ashworth Scale for spasticity is a measurement of spasticity across 9 muscle groups. Each muscle group is scored on a 0 to 5 scale with higher scores indicating worse functioning. The total score is the average score from the 9 muscle groups and ranges from 0 to 5 with higher scores indicating worse functioning.
Time Frame	12 weeks minus baseline

Outcome Measure Data

Analysis Population Description

[Not Specified]

Arm/Group Title	Robot-assisted Therapy	Intensive Comparison Therapy	Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.	Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.	Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
Measure Participants	47	46	27
Least Squares Mean (Standard Error) [units on a scale]	-0.07 (0.09)	0.12 (0.09)	-0.04 (0.11)

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Usual Care
	Comments	Because randomization to usual care was stopped after 15 months as specified by the protocol comparisons of robot-assisted therapy with usual care included only patients who were recruited during this period.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.95
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis of covariance at 12 weeks, adjusted for study site as a fixed effect, Comorbidity Disease Index, and the baseline value of the outcome.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	0.01
	Confidence Interval	(2-Sided) 95%; -0.25 to 0.26
	Parameter Dispersion	Type: Value:
	Estimation Comments

Statistical Analysis 2

Statistical Analysis Overview	Comparison Group Selection	Robot-assisted Therapy, Intensive Comparison Therapy
	Comments	Comparisons between robot-assisted therapy and intensive comparison therapy included all patients who underwent randomization and were evaluated.
	Type of Statistical Test	Superiority or Other
	Comments
Statistical Test of Hypothesis	p-Value	0.10
	Comments	All p-values are 2-sided with a significance level of 0.022 to adjust for multiple comparisons and interim analyses.
	Method	ANCOVA
	Comments	Analysis of covariance at 12 weeks, adjusted for study site as a fixed effect, Comorbidity Disease Index, and the baseline value of the outcome.
Method of Estimation	Estimation Parameter	Mean Difference (Net)
	Estimated Value	-0.19
	Confidence Interval	(2-Sided) 95%; -0.42 to 0.04
	Parameter Dispersion	Type: Value:
	Estimation Comments

Adverse Events

Time Frame	Adverse Events were monitored and collected over the entire participant follow-up period of 36 weeks
Adverse Event Reporting Description	Site personnel inquired about adverse and serious adverse events at each study contact.
Arm/Group Title	Robot-assisted Therapy		Intensive Comparison Therapy		Usual Care
Arm/Group Description	Robot-Assisted Therapy uses the MIT-MANUS robot and consists of four modules, shoulder-elbow, anti-gravity, wrist, and hand-unit to train the entire upper limb. Training was given for 12 weeks and was divided into 4 consecutive blocks, with 9 training sessions per block.		Intensive comparison therapy consisted of the identical number of treatments, time, and intensity of Robot-assisted therapy (12 weeks, 3 times per week). Each 1-hour therapy session consisted of four successive stages: 1) warm-up and assisted stretching; 2) active arm treatments; 3) goal-directed planar reaching, and 4) functionally based Neurodevelopment Techniques/Bobath arm training.		Usual Care consisted of the usual chronic stroke care as delivered at each participating medical center.
All Cause Mortality
	Robot-assisted Therapy		Intensive Comparison Therapy		Usual Care
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	/ (NaN)		/ (NaN)		/ (NaN)
Serious Adverse Events
	Robot-assisted Therapy		Intensive Comparison Therapy		Usual Care
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	11/49 (22.4%)		18/50 (36%)		9/28 (32.1%)
Cardiac disorders
Cardiac disorder	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Cardiac failure	0/49 (0%)	0	1/50 (2%)	1	1/28 (3.6%)	1
Cardiac failure congestive	0/49 (0%)	0	1/50 (2%)	2	1/28 (3.6%)	1
Coronary artery occlusion	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Myocardial infarction	0/49 (0%)	0	0/50 (0%)	0	0/28 (0%)	0
Cardiac Disorders	1/49 (2%)	2	2/50 (4%)	3	3/28 (10.7%)	3
Gastrointestinal disorders
Pancreatitis	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Upper gastrointestinal haemorrhage	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Vomiting	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Gastrointestinal Disorders	0/49 (0%)	0	3/50 (6%)	3	0/28 (0%)	0
General disorders
Chest pain	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Death	0/49 (0%)	0	2/50 (4%)	2	0/28 (0%)	0
Drug interaction	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
General Disorders	1/49 (2%)	1	2/50 (4%)	2	1/28 (3.6%)	1
Hepatobiliary disorders
Cholecystitis	1/49 (2%)	2	0/50 (0%)	0	0/28 (0%)	0
Cholecystitis acute	2/49 (4.1%)	2	0/50 (0%)	0	0/28 (0%)	0
Hepatobiliary Disorders	2/49 (4.1%)	4	0/50 (0%)	0	0/28 (0%)	0
Infections and infestations
Cellulitis	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Localised infection	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Pneumonia	1/49 (2%)	2	1/50 (2%)	1	1/28 (3.6%)	1
Urinary tract infection	1/49 (2%)	1	1/50 (2%)	1	0/28 (0%)	0
Infections	3/49 (6.1%)	5	2/50 (4%)	2	1/28 (3.6%)	1
Injury, poisoning and procedural complications
Ankle fracture	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Fall	2/49 (4.1%)	2	2/50 (4%)	2	1/28 (3.6%)	1
Overdose	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Road traffic accident	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Injury	3/49 (6.1%)	3	4/50 (8%)	4	1/28 (3.6%)	1
Neoplasms benign, malignant and unspecified (incl cysts and polyps)
Prostate cancer	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Renal cancer	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Neoplasms	0/49 (0%)	0	2/50 (4%)	2	0/28 (0%)	0
Nervous system disorders
Cerebrovascular accident	1/49 (2%)	1	0/50 (0%)	0	1/28 (3.6%)	1
Convulsion	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Hemiparesis	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Transient ischaemic attack	1/49 (2%)	1	1/50 (2%)	1	0/28 (0%)	0
Nervous System Disorders	2/49 (4.1%)	3	2/50 (4%)	2	1/28 (3.6%)	1
Psychiatric disorders
Aggression	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Depression	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Psychiatric Disorders	0/49 (0%)	0	1/50 (2%)	1	1/28 (3.6%)	1
Renal and urinary disorders
Nephrolithiasis	1/49 (2%)	1	0/50 (0%)	0	0/28 (0%)	0
Neurogenic bladder	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Urinary tract obstruction	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Renal and Urinary Disorders	1/49 (2%)	1	0/50 (0%)	0	2/28 (7.1%)	2
Respiratory, thoracic and mediastinal disorders
Pulmonary oedema	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Sleep apnoea syndrome	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Respiratory Disorders	0/49 (0%)	0	1/50 (2%)	1	1/28 (3.6%)	1
Social circumstances
Respite care	1/49 (2%)	1	1/50 (2%)	2	1/28 (3.6%)	2
Social Circumstances	1/49 (2%)	1	1/50 (2%)	2	1/28 (3.6%)	2
Surgical and medical procedures
Cardiac pacemaker battery replacement	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Colon polypectomy	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Hernia repair	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Knee arthroplasty	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Surgical and Medical Procedures	0/49 (0%)	0	2/50 (4%)	2	2/28 (7.1%)	2
Vascular disorders
Hypotension	0/49 (0%)	0	0/50 (0%)	0	1/28 (3.6%)	1
Orthostatic hypotension	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Thrombosis	0/49 (0%)	0	1/50 (2%)	1	0/28 (0%)	0
Vascular Disorders	0/49 (0%)	0	2/50 (4%)	2	1/28 (3.6%)	1
Other (Not Including Serious) Adverse Events
	Robot-assisted Therapy		Intensive Comparison Therapy		Usual Care
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	12/49 (24.5%)		9/50 (18%)		0/28 (0%)
General disorders
Pain, Stiffness, or Soreness	10/49 (20.4%)	23	6/50 (12%)	7	0/28 (0%)	0
Fatigue	3/49 (6.1%)	6	0/50 (0%)	0	0/28 (0%)	0
Swelling or Bruising	1/49 (2%)	1	1/50 (2%)	3	0/28 (0%)	0
Cut, Scratch or Irritation	2/49 (4.1%)	2	2/50 (4%)	2	0/28 (0%)	0
Numbness	2/49 (4.1%)	2	0/50 (0%)	0	0/28 (0%)	0

Limitations/Caveats

[Not Specified]

More Information

Certain Agreements

All Principal Investigators ARE employed by the organization sponsoring the study.

There is NOT an agreement between Principal Investigators and the Sponsor (or its agents) that restricts the PI's rights to discuss or publish trial results after the trial is completed.