Application of constraint-induced movement therapy for an individual with severe chronic upper-extremity hemiplegia. (Case Report).

Stroke is the leading cause of disability in the United States. (1) More than 750,000 people are estimated to have strokes each year in the United States, and over 4 million people in the United States are living with a disability as a result of stroke. (1,2) Interventions that address the sensorimotor impairments resulting from stroke, therefore, are important.

The concept of learned nonuse has been used to explain the motor deficits, that often occur following stroke. (3-7) Learned nonuse develops as a result of an upper motor neuron lesion that depresses the central nervous system and motor activity. When people who have sustained a stroke or brain injury initially attempt to move or use the affected body part or parts, they may experience failure and frustration. The individual learns how to compensate for this lack of movement by using the uninvolved or less involved extremities or the trunk, or both. The compensatory strategies become habit, (4) and eventually the individual does not attempt to move the affected extremity even when it is possible neurologically. The individual has learned to not use the affected extremity. Few, if any, rehabilitation techniques are proven to restore function or overcome learned nonuse in the affected upper extremity following a stroke. The demands of today's society and health care environment often necessitate the attainment of the highest functional level possible in a brief amount of time. For this reason, the therapeutic focus or a client's choice is often on compensating for lost movement by relying primarily on the side not affected by the stroke for activities of daily living (ADL). (3,8-10) Performing ADL tasks with one arm may still leave the individual with limited abilities. (11,12) Four years after stroke, many people report loss of arm function as a major problem, even though 96% of those studied reached independence in ADL according to their Barthel Index scores. (12) Persistent reliance on one side of the body also may result in consequences such as overuse syndromes, pain, frustration, and embarrassment. (11,13,14)

Constraint-induced movement therapy (CIMT) is an intervention that has research support for improving motor ability following a stroke or brain injury. Several investigations in the past 2 decades have demonstrated the effectiveness of CIMT with individuals who have residual upper-extremity weakness as the result of an upper motor neuron lesion. (3-5,15-20) The basic components of CIMT involve restraint of the unaffected arm for 90% of waking hours for a 2- to 3-week period in conjunction with repetitive training of the more affected upper extremity. (3,15-19,21) The less affected upper extremity is restrained with a mitt, sling, or glove. Clients typically participate in 6 to 7 hours of therapy a day plus home activities and ADL. A behavioral agreement is used to detail activities to be performed at home while wearing the restraint. This component of the program is intended to promote clients' adherence. The client also keeps a treatment diary to track use of the affected arm when away from the clinic.

Since 1999, the effects of CIMT have been investigated using neuroimaging techniques with people who had a stroke more than 6 months previously. These studies included, imaging via electroencephalogram (22,23) and focal transcranial magnetic stimulation. (16,17) These imaging techniques provide evidence of neuroplasticity following CIMT. The cortical changes seen with neuroimaging correspond to the functional and laboratory improvements demonstrated with motor assessments. The clients in these neuroimaging studies had typical CIMT (restraint 90% of waking hours, 6 hours of training for 10 out of 14 days, and a daily diary), with the exception of 2 studies in which participants received treatment for 8 out of 12 program days. (17,22)

Subject criteria for most published CIMT research primarily included the amount of movement a client must be able to perform with the more affected upper extremity. (3-5,16-20,24,25) All movement criteria included the ability to start from a resting position of forearm pronation and wrist flexion and actively extend each metacarpophalangeal (MCP) and interphalangeal (IP) joint at least 10 degrees and extend the wrist at least 20 degrees. (26) Individuals participating in these studies demonstrated improvements in the amount of use and quality of movement in the more involved upper extremity as well as carryover of skills from the clinic to the real world. (3-5,15-20,24,25)

Constraint-induced movement therapy, however, has not been found to benefit all individuals with hemiparesis. The majority of participants in CIMT investigations have met the inclusion criteria for movement. The ability to actively extend the more involved wrist 20 degrees and extend each MCP and IP joint at least 10 degrees, however, is not characteristic of many people with chronic hemiparesis.

Preliminary work investigating CIMT with clients with less functional ability of the hemiparetic arm also has had favorable results. Taub and Morris (27) reported improvement in 11 of 11 individuals who had at least 10 degrees of active wrist extension, 10 degrees of active thumb abduction, and 10 degrees of active extension of any other 2 digits, or roughly enough finger extension to release a tennis ball. The investigators also reported successful outcomes with 15 out of 16 individuals from an even lower functioning group. The minimum motor criteria for these subjects was the ability to "lift a wash rag off a tabletop using any type of prehension they could manage, and then release the rag." (27(p283))

To date, only a single brief mention has been made about the results of a CIMT program with an individual who had little to no movement. Taub et al described the client as having "almost no ability to move his fingers." (4(p245)) This person did not meet the minimum motor criteria of being able to grasp, lift, and release a washcloth. The authors described the outcome of a CIMT treatment program with this individual as a "treatment failure." (4(p245)) They did not provide details of the intervention program.

No studies have been published about application of CIMT with people who do not meet the motor criteria of 10 degrees of active MCP and IP joint extension and 20 degrees of active wrist extension. Our hospital receives numerous inquiries about CIMT from people with severe sensorimotor deficits as result of stroke. We estimate that over a 20-month period (June 2000-February 2002), approximately 42% (147 out of 352) of the potential client inquiries we received for CIMT were from individuals who could not perform the washcloth test. Because a CIMT program with an individual with this much movement impairment has not been described, the purpose of this case report is to describe a CIMT program with a woman who had little movement of her wrist and hand more than 15 years after a stroke.

Case Description

Client

"JM" was a 53-year-old right-handed woman. About 15 years before, at the age of 37 years, she had weakness and confusion while approaching the finish line of a 10-km running race. She was immediately transported to the local emergency department and was noted to have aphasia and right hemiparesis of her upper and lower extremities. Prior to the stroke, she had no history of cardiac or hypertensive disease. Two computed tomography scans were essentially normal but suggested mild edema in the left hemisphere, with no evidence of intracranial hemorrhage or ischemic infarct. Angiography revealed slow blood flow and some narrowing of the left internal carotid artery as compared with the right side. According to her medical records, these findings may have been due to arterial spasm or some increased resistance to the left anterior circulation such as a cortical vein thrombosis. Based on her symptoms, she most likely had a left middle cerebral artery occlusion. During her hospitalization, physicians discovered that JM had a mild, intermittent mitral valve prolapse, which may have precipitated her stroke. JM reported participating in inpatient rehabilitation for 10 weeks, followed by home health physical therapy and speech and language therapy intermittently for 2 years.

Screening

The client responded to posted information about a CIMT research project at Spalding Rehabilitation Hospital (Aurora, Colo). An initial telephone interview was completed with her. During this interview, she stated that she was essentially one-handed (using her left arm alone) for ADL and functional activities. She drove and maintained a busy family and social life. She no longer worked, but prior to her stroke she was a social worker. Her goals for the program were to regain some use of her right upper extremity and to hold her grandchild. JM met the initial screening criteria of:

* stroke more than 1 year ago

* no serious uncontrolled medical complications

* ability to follow directions (written, verbal, or demonstration)

* availability of a caregiver 24 hours a day for 3 weeks

* not currently receiving skilled therapy or treatment for the involved upper extremity

During her initial screening visit and her pretreatment testing, she appeared to meet the minimum motor criterion for a study being conducted at our facility: the ability to grasp a washcloth from a tabletop, lift it, and release it. During her treatment program and at posttreatment testing and follow-up testing, however, she could not grasp, lift, and release the washcloth. Upon further review of her pretreatment videotape, it is not clear whether she was doing 100% of the task with her right arm; it appeared that she may have assisted her right hand with her left hand. For this reason, we thought that she did not meet the minimum motor criterion and that her motor ability was lower than initially determined. We decided that it was more appropriate to describe her case separately.

During her initial visit to the clinic, JM was pleasant and apparently in good health. She walked without an assistive device or orthosis and stated she was able to walk at least 3.2 km (2 miles). She had expressive aphasia but managed to accurately make her needs known. She initiated her involvement in the program by telephoning our clinic to inquire about it. After completion of the screening process, she gave written consent to participate in a 3-week CIMT program.

Prior to her stroke, JM was right-hand dominant. Since the stroke, she had been functioning almost completely with her left upper extremity. She did not use her right upper extremity for writing, eating, household activities such as turning on lights, answering the telephone, or opening doors or drawers. She also did not use it for grooming, hygiene, or dressing, and she could not unlock a door with a key or open the refrigerator with her right hand. She did use her right upper extremity as a gross assist to remove clothes from the dryer. She could not move the fingers of her right hand.

JM had full passive range of motion (PROM) in her right upper extremity, with the exception of shoulder flexion and wrist extension. We did not record PROM measurements because range of motion was measured grossly as a part of the Fugl-Meyer Evaluation of Physical Performance. On the Fugl-Meyer test, passive shoulder abduction is tested only to 90 degrees. She reported pain in the shoulder with passive lateral (external) rotation but with no other passive movements of the extremity. When she attempted to perform activities that required right shoulder movement, JM used trunk substitution or scapular elevation to "move" her right arm. When asked to bring food to her mouth, for example, instead of shoulder and elbow flexion, JM used scapular retraction, shoulder elevation, trunk lateral flexion to the left, thoracic flexion, cervical lateral flexion to the left and cervical protraction. When palpating her right shoulder, the client appeared to have an approximately 1.27-cm (0.5-in) glenohumeral subluxation, but she reported no pain. Later in the day, JM clarified her response to state that she had intermittent right shoulder pain when she attempted to move her right upper extremity. She described a "popping or clicking" sensation in her shoulder when she attempted to elevate her arm.

During one of the initial administrations of the Motor Activity Log (MAL), while trying to use the rating scales, JM reported that she had a little trouble with reading. She appeared to read and write well, however, when she completed all of the prestudy paperwork herself and corresponded with one of the authors on a number of occasions via e-mail. Her husband, LM, filled out her daily diary with her verbal input, but this was necessary because JM could not write with her right hand. The MAL rating scales (which are lengthy) were read to her on 3 occasions to assist with the test.

Preintervention Testing

One day prior to her 3-week CIMT program, JM completed preintervention testing, which consisted of:

* a motor assessment (the washcloth test)

* the Graded Wolf Motor Function Test (GWMFT)

* the MAL

* upper-extremity portions of the Fugl-Meyer Evaluation of Physical Performance

* the Brief Neuropsychological Cognitive Evaluation (BNCE)

* the Mini-Mental State Examination (MMSE)

Her husband served as her assistant for the duration of the program and also completed the MAL with regard to JM's amount of use and quality of movement of her involved (right) upper extremity. All of the tests, with the exception of the BNCE, were selected because they have commonly been used in CIMT research. (3,18-20,25,28)

Test Descriptions

The WMFT was developed for use with people with mild to moderate stroke. (3,29) It is useful for measurement in approximately the upper 50% of people with deficits of the upper extremity resulting from stroke or brain injury. (30) The WMFT is reported to have high interrater and intrarater reliability for both performance time and functional ability scores. (29,31) Morris et al (29) used intraclass correlation coefficients (ICCs) to determine interrater reliability and reported ICCs of [greater than or equal to] .97 for performance time and [greater than or equal to] .88 for functional ability. They used Pearson correlation coefficients (r) to determine intrarater reliability and reported values of r=.90 for performance time and r=.95 for functional ability. Wolf et al (31) also used ICCs to determine interrater reliability of WMFT scores and reported ICCs ranging from .97 to .99. Wolf et al (31) supported construct validity by documenting differences in scores between individuals with upperextremity hemiparesis as a result of stroke and control subjects. They also supported criterion validity by demonstrating a correlation between the WMFT scores and the Fugl-Meyer test scores for the more affected extremity of people poststroke using the Spearman rank correlation coefficient ([r.sub.s]= -.57 and -.54 for session 1 and [r.sub.s] = -.67 and -.68 for session 2). (31)

The original WMFT was modified for use with clients who demonstrate moderate to severe motor impairment of the upper extremity. (30) This modified version, the GWMFT, which we used with JM, incorporates 14 upper-extremity motor tasks to assess movement components required for daily tasks. The time required to perform each task and the quality of movement are measured. The client may take up to 120 seconds to perform each task, but a low time score is optimal. The median time score is reported. Quality of movement is assessed by blind raters using videotapes of the tests in random order. Scoring is completed using an 8-point Functional Ability scale, with scores ranging from 0 ("not attempted") to 7 ("normal movement"). The mean Functional Ability scale score is reported. The GWMFT Functional Ability scale is described in Appendix 1. Examples of tasks performed on this test include placing the hand on a table from a sitting position, lifting a washcloth, and flipping a light switch. A complete list of GWMFT task items is listed in Appendix 2.

The psychometric properties of the GWMFT are not known. Because it is a variation of the original test, we might speculate that it also yields valid and reliable measurements. We did not estimate the reliability of our measurements.

The MAL * was developed to assess a client's report of 30 common daily tasks. (32) It consists of 2 assessment subscales for rating the affected upper extremity: an amount of use scale (the Amount scale) and a quality of movement scale (the How Well scale). The test is administered as a semistructured interview with 6-point rating scales (0="no use of the affected extremity," 5="normal use"). (24) Examples of items on the questionnaire include combing hair, donning and doffing shoes, and opening a refrigerator. The Amount and How Well scales are depicted in Appendix 3.

Currently, there appears to be some disagreement about the psychometric properties of the MAL. (33,34) To date, no study investigating the psychometric properties of this test is available. Uswatte and Taub (24) (p222) reported high interrater reliability (mean "interclass" correlation [type 3,1]=.90) with both scales (Amount and How Well scales) and with reports by subjects and significant others reporting on subjects. Miltner et al (20) suggested that the lack of difference between MAL and WMFT scores taken at baseline and then again at pretreatment (2 weeks later, without any treatment) indicates good intertest reliability for both the MAL and the WMFT. Van der Lee et al (34) disagreed with the reliability claims, stating that Taub and Uswatte presented no data in their chapter to support their claims and that a difference in MAL scores from baseline to pretreatment did exist in the study by Miltner et al. It is evident that scientific investigation into the psychometric properties of the MAL is necessary; however, we decided to use this tool in an effort to model our work on the work completed in Taub's laboratory for general comparison purposes. We did not evaluate the reliability of our measurements.

The Fugl-Meyer test is a cumulative assessment that measures motor recovery, balance, sensation, and some joint function in people with hemiparesis. It also includes components that address coordination and speed. The test is based on the premise that motor recovery occurs in a predictable progression. We used the Upper Extremity Motor Score (0-66 points) and an overall total score consisting of the Upper Extremity, Sensation, Joint Range of Motion, and Pain scale scores (0-126 points). We did not use Lower Extremity and Balance scale scores because they did not pertain to this case. Duncan et al (35) estimated intratester reliability for all subsections and total scores of the Fugl-Meyer test using a repeated-measures analysis of variance and Pearson correlation coefficients. Intertester reliability, estimated using Pearson correlation coefficients, was high for the total scores of upper- and lower-extremity motor performance (at least r =.984 and r =.886, respectively).

The MMSE is a brief screening instrument used to assess cognitive abilities. Thirty is the maximum score. Both reliability and validity have been estimated for the MMSE using Pearson correlation coefficients. (36) Concurrent validity was estimated as r=.776 for the MMSE versus the Verbal IQ portion of the Wechsler Adult Intelligence Scale and as r=.660 for the MMSE versus the Performance IQ of the Wechsler Adult Intelligence Scale. Test-rest reliability was estimated as r =.887. Intertester reliability was estimated as r=.827. We used the MMSE as a quick screening tool to evaluate for any cognitive changes that may have occurred from pretreatment to posttreatment or at follow-up. Consistency of MMSE scores would suggest that a client had no cognitive changes throughout the intervention period that may have affected test performance or carryover of the program.

The BNCE is a rapid cognitive assessment (requiring approximately 30 minutes to administer) that addresses 10 cognitive domains. Questions address narrowly defined functions and are ordered with increasing complexity. Three results are generally reported for the BNCE: a cumulative numeric score, an impairment level based on the cumulative score, and a validity index. The test is divided into 2 sections. Part I tests common knowledge of topics such as orientation, presidential memory, naming, comprehension, and constructive praxis. Part II deals with more novel information and includes an assessment of shifting set, similarities, attention, and working memory. The cumulative score (total of both parts) ranges from 0 to 30. An impairment level is determined based on this cumulative score. A score of 0 to 9 indicates a severe impairment level. A score of 10 to 21 denotes a moderate level of impairment. Clients with a score of 0 to 21 generally are not able to live independently. A score of 22 to 27 suggests a mild level of impairment, and a score of 28 to 30 indicates no impairment. Clients in these last 2 categories may be able to live independently. The validity index is the ratio of part I to part II scores and can indicate whether the test results are not valid. A validity index of <0.8 denotes an invalid representation of the client's abilities, as measured by this test. Aphasia, malingering, and auditory or visual impairment are examples of situations that may produce a validity index of <0.8.

The BNCE is useful for distinguishing between psychological and neurological impairments, and its scores are useful for gauging the severity of neurological impairment. Comparison studies with the MMSE show the BNCE to be more sensitive with regard to identification and measurement of cognitive status. (37)

Two studies discussed in the BNCE manual (37) examined test-retest reliability of BNCE scores. One study with clients with psychiatric problems, aged 65 to 74 years, determined test-retest reliability to be .95, interrater reliability to be .97, and internal consistency (Cronbach alpha) to be .90 for the total score. The statistic used to determine reliability was not identified. In a second study involving 41 clients (background details were not provided), the test-retest correlation for the total score was .88 and reliability of subtest scores ranged from .64 to .97, with a median of .83. Internal consistency was .88. In terms of construct validity, in a comparison with 9 other full-scale cognitive assessments, the BNCE scores were most highly correlated with the full-scale IQ score of the Wechsler Adult Intelligence Scale, (correlation [type not identified] =.71).

For discriminant validity, the BNCE impairment scores were found to be directly related to lesion severity in patients with neurological problems: mild ([bar]X=25.8, SD=5.4), moderate ([bar]X =17.5, SD=5.1), and severe ([bar]X =8.1, SD=5.1). High discriminant validity also was found for the part I and II subsections. When compared with the MMSE, the BNCE was found to be more sensitive for detecting mild cognitive deficits. (37)

We used the MMSE at each test administration to detect any gross cognitive changes over the course of the program. The BNCE was used at pretreatment testing only to gather detailed baseline cognitive data.

The test administrators had 2 to 25 years of experience with evaluating and treating people with neurological deficits. All administrations of the GWMFT, the patient's MAL, the caregiver's MAL, and the motor assessment were done by the same physical therapist (NB). The MAL and GWMFT instructions include detailed descriptions and scripts for test administration. The physical therapist reviewed and consistently followed all testing material guidelines for these tests. The motor assessment was conducted (NB) using the same table and chair and with the same verbal directions at each administration. All administrations of the Fugl-Meyer test and the BNCE were completed by the same occupational therapist (KMA) following testing guidelines given in the BNCE manual (37) and described by Fugl-Meyer et al. (38) A different physical therapist and occupational therapist team (HB and PK) always graded the Functional Ability scale score on the GWMFT. This team reviewed the GWMFT manual prior to training and received two 1-hour training sessions and a 4-hour videotaped scoring competency test. The MMSE was administered by the physical therapist (NB) or a nurse (FH). Both were instructed in the application of the MMSE and followed testing guidelines as given by Folstein et al. (36)

Initial Status on Preintervention Testing

On preintervention testing, JM's results on the BNCE were a score of 24 out of a maximum possible score of 30, an impairment level of mild, and a validity index of 1.18. The total score and impairment level suggest that she could possibly live independently. The validity index indicates that the test results were valid. (37) Subset scores indicated thatJM had more difficulty with novel or incomplete information than with frequently used information.

The client's MMSE score on preintervention testing was 29 out of 30. This score remained relatively constant throughout her intervention and follow-up period (29 on posttest, 29 on 1-month follow-up, 28 on 6-month follow-up), suggesting that she had no change in cognitive abilities throughout the study period. These scores fall within established age-specific norms. (39)

Initially, Fugl-Meyer testing revealed the limited ability in the affected (right) upper extremity. The client's initial Fugl-Meyer Upper Extremity scale score was 17 (out of a maximum possible score of 66). Her cumulative score was 52 (out of a maximum possible score of 126). She had decreased PROM in right shoulder flexion and right wrist extension and passive right shoulder abduction to at least 90 degrees. She had dysesthesia to light touch on the right upper arm and palm of the hand. This finding was corroborated during treatment when JM said that she could not feel a mug placed in her hand. Proprioception was intact at the right shoulder and elbow, decreased at the right wrist, and absent at the right thumb. No isolated movement was noted in the right upper extremity, although she was able to elicit some movement into synergy patterns (flexor and extensor).

The MAL asks the client about attempted use of the affected upper extremity in the recent past. JM reported attempting to use her right arm on 4 out of the 30 items on the MAL for the week prior to testing but that her right arm was not useful for these tasks. Her initial MAL scores were 0.13 (out of a maximum possible score of 5.00) for the Amount scale and 0.13 (out of a maximum possible score of 5.00) on the How Well scale.

The client's performance on the GWMFT was videotaped, and each item was timed according to test guidelines. The videotape was rated using the Functional Ability scale by 2 blinded raters (a physical therapist and an occupational therapist who reached a consensus on the score). The GWMFT was administered by the same therapist, in the same room, using the same equipment, same testing template, and same chair and table position for each administration. JM's initial median time score was 118.47 seconds, indicating inefficient movement or difficulty performing a majority of the tasks. Her initial mean Functional Ability scale score was 1.43 (on a scale of 0-7), suggesting that her right upper extremity was not participating in most of the tasks or was participating but with assistance, multiple attempts, compensatory movements, or extra time.

Intervention

JM completed a 21-day intervention program in which she restrained her left upper extremity with a soft mitt attached around the wrist with Velcro ([dagger]) and a standard upper extremity sling. The goal was for her to wear the restraint for 90% of her waking hours.

For 15 of the 21 intervention days (the weekdays), JM was in the clinic for 6 hours of training each day. Three weeks of treatment was chosen as the intervention period instead of 2 weeks at the suggestion of David Morris, PT, MS (oral communication, July 2000) based on clinical experience at the University of Alabama CIMT research laboratory.

The training approaches implemented for this program were: (1) massed practice, (2) shaping, (3) a 1:1 physical therapist-to-client ratio, (4) occasional tactile and verbal feedback (mainly given to instruct the client about trunk substitution), (5) home treatment agreement, and (6) daily treatment diary.

Massed practice involves repetitive attempts to use the involved body part for many hours per day and for consecutive days. (4) Repeatedly attempting to move an involved limb for extended periods of time is thought to be the driving force behind the use-dependent cortical reorganization described in neuroimaging studies involving CIMT. (16,17,23) The principle of massed practice was integrated into JM's program by restraining her less affected side for most of her day and repeatedly attempting to move and use her affected arm for a period of 21 days.

Shaping is a behavioral technique. Taub and Uswatte defined shaping as: "1) selecting tasks that were tailored to address the motor deficits of the individual patient, 2) helping the patient to carry out parts of a movement sequence if they were incapable of completing the movement on their own at first, and 3) providing explicit verbal feedback and verbal reward for small improvements in task performance." (4) (p243) Performance on tasks is measured quantitatively, and clients are given positive feedback as they improve on a particular task. The difficulty level of the task should be compatible with the client's current movement ability. The parameters of the task are adjusted so that as the client improves, the therapist progressively increases the challenge of the task. (4) According to Taub and Morris, (27) shaping is particularly important in the management of clients with less movement ability.

An example of shaping in JM's program was a repetitive task to facilitate opening of the affected hand. JM's right hand was generally fisted. Opening her hand to grasp or release objects was difficult and often impossible for her. The smaller the object, the tighter her grasp and the more difficulty she had releasing the object. Initially, she could not release small objects, such as a marble or a 2.54-cm (1-in) cube block. She could inconsistently release larger items, such as a tennis ball or bunched up washcloth, although this frequently took a long time (sometimes 10+ minutes to release). Her task was to see how long it took her to release a tennis ball. Each attempt was timed, and she rested between attempts, generally for a period equivalent to the time it took her to release the ball. On the first day, she did 10 repetitions of grasping and releasing the ball. The time it took her to release the ball ranged from 1/2 minute to 5 minutes (with 8 out of the 10 attempts being 1 minute). On subsequent days, her ability to release the tennis ball improved. She would attempt to release smaller objects (Mancala stones, cotton balls, or marbles) as she improved. As her release times improved, we set a timer for a specified amount of time (eg, 10 minutes) and had her count how many times she could release the object (eg, marble) in 10 minutes. The number was then recorded by the therapist. She would then rest for 10 minutes and repeat the sequence 2 to 3 more times. JM would work on one specific task (eg, releasing marbles) for 30 to 45 minutes, including rests. Positive feedback was supplied verbally at the time of her success as well as by graphing her progress (decreased time to complete the task or number released in a given time period) over several days.

Another example of how we used shaping was drinking liquids with meals. Originally, JM could not drink from a cup or mug using her right upper extremity. We had her attempt to drink by grasping and slightly lifting a covered mug with a straw. Initially, she required the therapist to stabilize the mug, place her fingers and thumb around the handle, and assist with lifting the mug so the straw reached her mouth. As motor skill in this task improved, the amount of assistance was decreased (the client placed her fingers, but the therapist placed her thumb, and the amount of assistance given to lift the mug was decreased and eventually removed). Further progression would include the client grasping the mug handle without assistance. The ultimate goal was for JM to progress to drinking from an uncovered cup or mug without a straw using the involved hand.

The therapist worked with JM 1:1. Taub and Uswatte (33) emphasized the importance of the 1:1 therapist-to-client ratio in the application of CIMT. Because shaping is important when working with individuals with limited movement ability of the upper extremity, (4) we used shaping as well as a 1:1 therapist-to-client ratio.

The home treatment agreement and daily treatment diary are considered to be essential behavioral components for carryover of a structured CIMT program. (40) The home treatment agreement is a contract that is created with input from the client, caregiver, and therapist. It specifically details what activities will be done with the restraint on and with the restraint off, when the restraint will be worn in community or social situations and when the restraint should be removed for potentially unsafe situations. The treatment diary is a detailed daily log of what the client does when not in the clinic and how much the affected extremity is used. The client and caregiver use this form for daily documentation and include as much detail and description as possible. For instance, the client may have reported that for the previous evening's dinner she ate 75% of a steak with a built-up fork and that the steak had to be cut by the caregiver. The client also may report how much time it took to perform an activity, such as 10 minutes to remove both shoes using only the affected upper extremity. Each morning, the therapist reviewed the diary with the client and caregiver. The diary is a means for clients and caregivers to note measurable improvements in performance quality, quantity, or time. The diary assisted with ongoing evaluation of program adherence and problem solving of tasks that are difficult for the client.

In clients with moderate to severe disability, the caregiver plays an essential role in assisting the clients, assuring safety and providing support and encouragement. JM's husband, LM, was active in the program. Initially, he was in the clinic for 2 full days of training. He then assisted JM in carrying out her program at home in the mornings and evenings and on weekends. He was present every morning prior to the start of JM's daily therapy activities to review with her therapist how JM managed at home the previous evening and that morning. This time was for discussion of the client's accomplishments and problem solving of activities that were challenging. He picked her up at the end of each clinic day and was instructed in her home practice tasks for that evening or weekend. A typical treatment day forJM is shown in Table 1.

Instruction

One of the main focuses of intervention was to teach the client and caregiver about learned nonuse, cortical reorganization, and recovery of function. In addition, instruction was provided with emphasis on safety while engaging in a CIMT'program. The expectation was that JM would continue to carry out a modified CIMT program at home over the following 6 to 12 months. She was given specific instruction in repetitive home exercises and functional use of the right upper extremity, including a home exercise videotape with instructions for proper body mechanics and explanations of how to progress tasks as she gained movement. We also recommended that the client continue with some form of restraint at home as much as possible when it did not compromise her safety. In general, we encourage clients to use a restraint for 2 to 3 hours per day if possible.

Postintervention Testing

The day following JM's 3-week CIMT program, she completed postintervention testing consisting of: motor assessment (washcloth test), GWMFT, MAL, portions of the Fugl-Meyer test, and the MMSE. Her husband also completed the MAL. We also elicited feedback from JM and LM about the program via an exit questionnaire form and discussion.

Follow-up Testing

Motor assessment, GWMFT, client and caregiver MALs, Fugl-Meyer test, MMSE, and client comments were evaluated at 1 month and 6 months postintervention.

Outcomes

JM wore the restraint an average of 84% of her waking hours for the 3-week program. Based on her inability to grasp a washcloth, lift it, and release it, her upper-extremity motor ability continued to be considered severely impaired at posttreatment testing and each of her follow-up visits.

JM's function did not appear to have improved at posttreatment testing. Following intervention, she still could not cook or clean, do meal setup, eat, dress, or perform the majority of her ADL with her right hand. She still could not consistently "stab" her food with a fork and continued to require assistance at meals if she ate using her right upper extremity. She was still preparing meals, getting her clothes out, and dressing primarily with her left upper extremity. She used her left upper extremity for drinking, housework, and self-care. Her test score on the Fugl-Meyer test indicated improvement of right upper-extremity active movement within the flexor synergy. At posttreatment testing, she was able to perform scapular retraction and forearm supination within the flexor synergy, neither of which she was able to fully perform at pretreatment testing. She did not progress, however, to the point that she could accomplish multijoint movements necessary to complete functional tasks. She still could not consistently release objects from her grasp in the right hand. Comparisons of pretreatment and posttreatment scores as well as 1-month and 6-month follow-up scores are shown in Tables 2 and 3 and Figures 1 through 3.

JM scored higher on the MAL with the amount of use and the quality of movement of her right upper extremity after the CIMT intervention (Fig. 1). On the Amount scale, she changed from 0.13 at pretreatment testing to 2.00 at 1-month follow-up testing. This finding suggests that she had almost no use of the affected arm before treatment. She progressed to some use of the right upper extremity but was still primarily dependent on the left upper extremity 1 month after her treatment program. The Amount scale is not used for evaluation immediately after treatment because the client has just completed a period of forced use of the affected upper extremity and reports may be considered not truly representative of how much the client uses the arm in the real world. On the How Well scale, JM changed from 0.13 at pretreatment testing (again indicating the right arm was really not helpful) to 1.53 after treatment (the quality of movement of the right arm could be described as between very poor and poor). This score further increased slightly at the 1-month follow-up visit (1.68).

The scores on both subscales of the MAL appear to have decreased to slightly above pretreatment status by the 6-month follow-up point (0.27 for the Amount scale and 0.47 for the How Well scale), which suggests that JM did not maintain any improvements she may have made over the long term. These lower MAL scores at 6-month follow-up were corroborated by the client's and caregiver's reports of her not performing her daily exercises or attempting to use her right upper extremity for functional activities at the time of the 6-month follow-up. JM reported that at some point between the 1-month and 6-month follow-up visits, she stopped integrating the CIMT concepts into her daily life and discontinued her home program. The caregiver's MAL scores (Fig. 2), which LM reported from his observations of JM, demonstrated a similar trend, with JM increasing from pretreatment to posttreatment testing but returning almost to baseline by the time of the 6-month follow-up.

Also of interest is the number of activities on the MAL that the client routinely attempted to perform. On pretreatment testing, JM reported that she had attempted 4 out of the 30 MAL tasks (picking up a telephone, opening a door, using a television remote control, and pulling a chair away from a table) in the week prior to testing. At the 1-month follow-up, she reported attempting 27 out of 30 tasks. By the 6-month follow-up, she reported attempting 6 out of 30 tasks. These findings again suggest that she attempted to use her right upper extremity more for at least the month following her 3-week program but, sometime between 1 and 6 months posttreatment, she returned to relying primarily on her left arm. The 6 tasks she reported attempting at the time of the 6-month follow-up visit were different from the original 4 tasks. They included turning on a light at a light switch, opening a refrigerator door, washing and drying her hands, brushing her teeth, and carrying an object.

Table 2 illustrates JM's results on the Fugl-Meyer Evaluation of Physical Performance. The Upper Extremity scale score is the upper-extremity motor component on the Fugl-Meyer test. The Comprehensive score includes the Upper Extremity, Sensation, Joint Range of Motion, and Pain scale scores.

JM demonstrated higher scores on the Fugl-Meyer test after the intervention period. The greatest gain was noted from pretreatment to immediate posttreatment (17-21 on the Upper Extremity scale and Comprehensive scores of 69-76). Specifically, JM improved in active movement performed within flexor and extensor synergies. However, she did not demonstrate isolated movement out of synergy during testing. Both subscale scores remained the same at the 1-month follow-up. Both Upper Extremity scale score and Comprehensive score decreased somewhat but still remained higher than pretreatment level at the 6-month follow-up (20 and 73 versus 17 and 69, respectively). Despite discontinuing her program, JM still maintained some of the score changes 6 months after her intervention.

JM's Functional Ability scale scores on the GWMFT (Fig. 3) increased slightly from pretreatment testing to posttreatment testing (from 1.43 to 1.57). The highest score was noted at the 6-month follow-up (2.36), which suggests that the quality of her movement may have improved. JM was able to perform 3 out of the 14 tasks on the GWMFT at pretreatment testing. These tended to be gross motor tasks such as lifting her forearm to a table at her side, lifting her forearm from the table onto a box at her side, and lifting her hand onto the table in front of her. She was unable to perform tasks such as flipping a light switch, picking up a pen, or releasing her grasp on a golf ball. She was able to complete 4 of 14 tasks at both posttreatment testing and 1-month follow-up. She had the ability to complete 6 out of the 14 tasks at the 6-month follow-up. The additional tasks she was able to perform involved active elbow extension and active shoulder lateral rotation. Although her ability to complete these 2 tasks at the 6-month follow-up was an improvement from the pretreatment testing, she completed them with some trunk substitution. She was still unable to complete the more functional tasks on the test at this time. The increased score at 6 months was somewhat unexpected, given her self-scored MAL results and her reports.

The client was given up to 120 seconds to perform each of the 14 GWMFT tasks. Performance of each task was timed, and the median performance time was calculated for both the unaffected and affected upper extremities. The difference between unaffected and affected upper-extremity median performance times was then recorded as the time score. The lower the time score (from 0-120 seconds), the more proficient the client is at performing active movement with the affected arm.

JM's GWMFT time scores (Tab. 3) did not differ throughout the period of time, and performance times close to 120 seconds (the maximum allowable time) reflected the difficulty JM had performing the tasks with her affected upper extremity.

Interviews

At the time of posttreatment testing, JM's and her husband's comments included rating the program 7 to 8 in terms of difficulty on a scale of 0 to 10 (0="not difficult," 10="most difficult thing she had ever done"). They also said they had not anticipated "that old grief issues still were active and would rise up again." They said that "considerable emotion was generated," at times she was very fatigued, and other times she was encouraged by what she saw as improvements in her movement. She commented that her shoulder pain had decreased as the program progressed.

At the 1-month follow-up, the client reported that she was doing about 3 of her exercises a day as well as attempting to use her right upper extremity frequently throughout the day. She reported her right upper extremity was a gross assist for activities such as opening drawers, opening the refrigerator door, and doing laundry. At times, she continued to eat with the right upper extremity, although not in public. She reported that she was not using a restraint at home. She reported that her shoulder did not hurt anymore or "pop or click." She stated that she felt she had "gained as a result of this program."

At the 6-month follow-up, JM reported that she had "tried to integrate (the use of her right upper extremity) into everyday tasks when I can" but "finds, however, that some everyday stuff/crisis absorb time and energy to do so." Usually, it was faster and easier to complete functional tasks with her stronger left arm. She reported little to no use of her right upper extremity at home except occasional use to carry bags, pick up clothes, or turn on lights. She stated that she no longer used her right upper extremity to eat and did not use a restraint of any kind. She continued to report less shoulder pain than prior to the program. Because an assessment of shoulder pain was not initially an objective, it was not quantified. The client reported that she decreased her focus on her home CIMT program at the time a family member became ill (sometime between 2 and 6 months post-intervention). She also indicated that she was not seeing the improvements she wanted, and this influenced her program maintenance.

Discussion

Prior to her involvement in CIMT, this client, who had a stroke 15 years previously, had only synergistic movement in her affected upper extremity and typically would not have been expected to make improvements in her physical abilities. However, she did achieve higher scores on the MAL, Fugl-Meyer Evaluation of Physical Performance, and GWMFT Functional Ability scale from pretreatment to posttreatment. These gains were maintained at the 1-month follow-up visit. Motor scores on the Fugl-Meyer test and the GWMFT remained higher at the 6-month follow-up than at pretreatment testing, but her MAL scores decreased to almost baseline level at 6 months. Her performance speed on the GWMFT showed no change from pretreatment testing to posttreatment testing or at either of the follow-ups.

These changes in test scores may be attributed to several factors. They could reflect improvement in physical capabilities of the upper extremity, a change in learned nonuse behaviors, a Hawthorne effect, or use-dependent cortical changes. There is also the possibility that some or all of the changes in values are not at all significant and occurred simply by chance.

Her improvements in test scores after the CIMT program may have been due to an improvement in her strength and coordination in the right upper extremity as a result of training. Use-dependent cortical reorganization, similar to what has been seen in subjects with greater movement ability, may have occurred as a result of JM's CIMT program. Because she attempted to move and use her right upper extremity during this program more than was her habit, the fact that her test scores increased may be similar to the outcomes of published studies with individuals who demonstrated a greater degree of movement ability. (3,4,16-20)

JM may have unknowingly been capable of performing more movement with her right upper extremity than she was doing prior to this intervention. With the restraint on and the intensive training directed toward her right upper extremity, she attempted to move and use her arm much more during the 3-week CIMT program than previously. This was evidenced by her MAL Amount scores as well as observing her. If she did possess a higher level of ability than she demonstrated prior to intervention and her test scores increased as a result of attempting to move her arm, this would suggest a change in learned nonuse behaviors.

We cannot conclude from this case that the CIMT intervention affected the changes in the test scores. The fact that JM received no other therapeutic intervention for the duration of this case lends credence to the notion that the improvements were the result of CIMT. The changes, however, could be due to the attention the client received or the almost constant attention given by almost everyone involved with her on moving her right upper extremity. We do not know whether actual changes occurred at the cortical level for this individual, because neuroimaging techniques were not included in this case. Currently, it is unknown whether use-dependent cortical reorganization can occur in an individual with this level of motor ability. To our knowledge, no one has undertaken such an investigation.

The significance of the changes in her assessment scores also is a question. How do GWMFT, Fugl Meyer test, and MAL scores relate to increased function in the real world with an individual such as this? With what change in scores would we (as well as third-party payers) consider this intervention to be successful and justifiable? In our opinion, the client's and caregiver's MAL scores appeared to represent a change. The GWMFT Functional Ability scale score increase of approximately 0.9 on a 7-point scale, however, is a little more difficult to interpret. Further investigation with more subjects would help to address this question.

JM's MAL questionnaire scores decreased at the 6-month follow-up, as would be expected based on her reports. The fact that her Fugl-Meyer test scores did not also decrease to the baseline level and her GWMFT Functional Ability scale scores further increased at the 6-month follow-up is interesting and difficult to explain.

Perhaps JM was unknowingly moving her affected upper extremity more for household activities and ADL even though she had decreased her amount of formal exercise. Perhaps, if cortical changes occurred, they were maintained over the 6-month time period despite her discontinuance of the program. A neuroimaging investigation along with a program such as this would be beneficial.

Several other limitations exist in the description of this case. First is the lack of psychometric data concerning the GWMFT and MAL. Although some description of the psychometric properties of the MAL exists, more focused investigation and documentation of both of these tests are needed. We chose to use these tools because they have been commonly used for published CIMT studies.

The fact that we did not measure the client's shoulder subluxation also is a shortcoming of this description. Further investigation in this area, including radiographic imaging, could be beneficial.

This client and caregiver were highly motivated and carried out the program for the 3-week intervention period, as evidenced by JM's effort in the clinic and detailed documentation and discussion between client, caregiver, and therapist. Although she demonstrated gains according to the assessment tools, 6 months postintervention she reported she had not achieved a higher functional level using her right upper extremity. In our opinion, this lack of functional change is the most salient conclusion to draw from this case. Although her test scores may have improved, this client's overall functional level did not appear to be changed as a result of this program.

When comparing changes in MAL How Well scores, it appears that JM had a notable increase in scores from pretreatment testing to posttreatment testing. However, she did not maintain the improvement at follow-up testing, as was the case in other studies. (4,18-20)

JM's mean pretest to posttest change in Functional Ability scale scores on the GWMFT was less than the changes seen in other CIMT studies. The improvement in Functional Ability scale scores for JM at the 6-month follow-up appears to be comparable to follow-up results of Kunkel et al (18) and Miltner et al (20) suggesting a possible change. We, however, used the GWMFT, whereas Kunkel et al and Miltner et al used the WMFT. Information concerning the comparability of the 2 tests is not available. The fact that the MAL score returned to near baseline level and the client reported no overall improvement in ADL despite the apparent improvement in GWMFT Functional Ability scale scores brings up several questions. Is the GWMFT Functional Ability scale score pertinent with this patient population? Was this client capable of more movement than she was using functionally? Is a more substantial change in Functional Ability scale score needed in this population to equate to functional improvement?

A CIMT program can be difficult and frustrating. The program is, by nature, intense. If physical progress occurs, it happens slowly and only as the result of a tremendous amount of effort and time by all participants. In this case, although our client put forth a tremendous effort, it was difficult for her to truly carry out the program due to her lack of isolated movement. Because most of her right upper extremity movement was accomplished via synergy or substitution, we spent more time than we wanted to working on isolated, single-joint, straight-plane movements (eg, shoulder lateral rotation, forearm supination, wrist extension). To try to combine movements (eg, shoulder flexion with elbow extension and forearm supination to pick up a plate) was not feasible and was frustrating for the client. Our client also said that her motivation diminished over time because she was not seeing what she considered significant changes fast enough. We concluded, therefore, that this treatment approach was not beneficial for long-term functional carryover for this client.

We theorize that more effective ways to administer CIMT for this type of client may exist. Altering the treatment approaches used within the context of CIMT would be an interesting topic for further investigation in clients with severe impairments. One theory we propose for further investigation is to address the client's lack of core stability at the shoulder girdle. This client lacked shoulder girdle stability, as evidenced by her shoulder subluxation, lack of coordinated scapulothoracic or glenohumeral movement, and reliance on trunk substitution and scapular elevation to accomplish "movement" with the right upper extremity. It would be worthwhile to investigate whether this client (and those with similar movement patterns) would benefit more and demonstrate greater long-term functional carryover from a CIMT approach that emphasizes proximal stability-enhancing, forced-use activities. This could be accomplished with the distal upper extremity placed in a position to promote full or partial weight bearing, rather than with primarily distally oriented arm on body movements. It would be interesting to explore whether this weight-bearing approach, in combination with the techniques of massed practice, and shaping, could result in an improved functional level for this individual.

Conclusion

This case report describes the application of a CIMT intervention program with an individual 15 years post-stroke who had no volitional movement in her affected upper extremity prior to intervention. Her scores on tests improved immediately following a CIMT program. In general, however, these increased scores were not maintained over the long term. Perhaps more importantly, the client did not demonstrate nor did she report improved functional ability as an overall result of the CIMT program.

This case report is meant to serve as a baseline descriptive effort of CIMT intervention for clients with severe impairment. We hope that it will lead to further research to ascertain the optimal interventions for individuals with very limited motor ability due to upper motor neuron lesion.

Appendix 1.

Graded Wolf Motor Function Test Functional Ability (FA) Scale (a)

Level B

0        Does not attempt with involved arm.
1        Involved arm does not participate functionally; however,
         attempt is made to use the arm. In unilateral tasks, the
         uninvolved extremity may be used to move the involved
         extremity.
2        Does, but requires assistance of uninvolved extremity for
         minor readjustments or change of position, or requires more
         than 2 attempts to complete, or accomplishes very slowly.
3        Does, but movement is performed slowly, and/or with effort,
         and/or with excessive compensatory movements.

Level A

4 (b)    Does, but requires assistance of uninvolved extremity for
         minor readjustments or change of position, or requires more
         than 2 attempts to complete, or accomplishes very slowly.
5 (b)    Does, but movement is performed slowly, and/or with effort,
         and/or with excessive compensatory movements.
6        Does; movement is close to normal, (c) but slightly slower;
         may lack precision, fine coordination, or fluidity.
7        Does; movement appears to be normal. (c)

(a) Reprinted with permission from the UAB CI therapy research project:
Constraint-Induced Movement Therapy Research Group. Manual: Graded Wolf
Motor Function Test [test instructions]. Birmingham, Ala: University of
Alabama at Birmingham and Birmingham Veteran's Administration Center;
August 2000.

(b) The language of ratings 4 and 5 on level A is the same as for
ratings 2 and 3 on level B and is applied to rating arm movement in the
same manner. The difference is that the task is more difficult at level
A; clients (subjects), in effect, receive extra points for performing
the task at level A.

(c) For the determination of "normal," the uninvolved limb can be used
for comparison, with premorbid limb dominance taken into consideration.

Appendix 2.

Graded Wolf Motor Function Test Task List (a)

All items, except for items 10 and 14, are from a seated position. Table height = 73.5 cm (29 in)

1. Forearm to table (side)

2. Forearm to box on table (side)

3. Extend elbow on table (to the side)

4. Extend elbow on table with weight (to the side)

5. Hand to table (front)

6. Hand to box on table (front)

7. Reach and retrieve weight using elbow flexion (front)

8. Moving foam stick (supination/pronation) (front)

9. Lift washcloth (front)

10. Flip light switch (standing and to the front)

11. Lift pen (front)

12. Lift foam triangles (front)

13. Drop golf ball or washcloth (forearm supported)

14. Lift basket from table and place on rolling table on affected side (standing)

(a) Adapted from: Constraint-induced Movement Therapy Research Group. Manual: Graded Wolf Motor Function Test [test instructions]. Birmingham, Ala: University of Alabama at Birmingham and Birmingham Veteran's Administration Center; August 2000.

Appendix 3.

Motor Activity Log Rating Scale (a)

Amount Scale                         How Well Scale

0-Did not use my weaker arm (not     0-The weaker arm was not used at
  used).                               all for that activity (never).
1-Occasionally tried to use my       1-The weaker arm was moved during
  weaker arm (very rarely).            that activity but was not help-
                                       ful (very poor).
2-Sometimes used my weaker arm       2-The weaker arm was of some use
  but did most of the activity         during that activity but needed
  with my stronger arm (rarely).       some help from the stronger arm
                                       or moved very slowly or with
                                       difficulty (poor).
3-Used my weaker arm about half      3-The weaker arm was used for the
  as much as before the stroke         purpose indicated, but movements
  (half prestroke).                    were slow or were made with only
                                       some effort (fair).
4-Used my weaker arm almost as       4-The movements made by the weaker
  much as before the stroke (3/4       arm were almost normal but not
  prestroke).                          quite as fast or accurate as
                                       normal (almost normal).
5-Used my weaker arm as normal as    5-The ability to use the weaker
  before the stroke (same as           arm for that activity was as
  prestroke).                          well as before the injury
                                       (normal).

(a) Reprinted with permission from the UAB CI therapy research project:
Constraint-Induced Movement Therapy Research Group. Manual: Upper
Fxtremity Motor Activity Log (UE/MAL). Birmingham, Ala: University of
Alabama at Birmingham and Birmingham Veteran's Administration Center;
August 2000.

Table 1.

A Typical Treatment Day for JM

6:15 AM      JM awoke, performed morning hygiene and dressing tasks,
               and had breakfast with both hands (using the right upper
               extremity as much as possible and documenting how much
               it was used).
7:00 AM      JM applied restraint, performed chores around house as she
               was able, got newspaper, and so forth, and traveled to
               the clinic.
9:00 AM      JM and caregiver (LM) reviewed the treatment diary and
               discussed the previous day's events with the therapist.
9:15 AM      Caregiver leaves. Portion of Motor Activity Log adminis-
               tered to client.
9:30 AM      Right upper-extremity training implementing the concepts
               of massed practice and shaping: activities focusing on
               eliciting and strengthening the client's ability to move
               and functionally use her right upper extremity. On ave-
               rage, each task was performed for 30 to 45 minutes with
               appropriate rest periods interspersed with movement
               attempts. Tasks were modified to increase difficulty
               when appropriate.
11:45 AM     Lunch: included meal preparation and eating with the
               restraint on. Client used built-up utensils and
               assistance from therapist. She also chose finger foods
               or foods that could be cut into chunks and more easily
               speared with a fork (eg, fruit, energy bars).
1:00 PM      Right upper-extremity training implementing the concepts
               of massed practice and shaping.
2:55 PM      Caregiver picked up client, and we reviewed the day's hap-
               penings and assigned home practice tasks for the eve-
               ning. In a separate section of the treatment diary, a
               homework log listed the tasks to be completed that
               evening and included space for the client or caregiver
               to comment on performance. On weeknights, an average of
               1 to 1.5 hours of repetitive exercise tasks were
               assigned, similar to what was done in the clinic. In
               addition, JM performed as much of her activities of
               daily living (ADL) with her right arm as possible. On
               weekends, she generally performed about 3 to 4 hours of
               assigned repetitive exercise tasks as well as ADL.
3:00 PM      Client and caregiver left the clinic for the day and con-
               tinued with home practice activities and functional
               tasks using the involved upper extremity as much as pos-
               sible. The restraint was worn on the left upper extremi-
               ty for most of the evening, including dinner. Events
               were documented in detail until they arrived at the
               clinic the next morning.
Weekends:    Client continued to wear restraint and use the affected
               upper extremity for exercises and functional tasks as
               much as possible with assistance from her caregiver. A
               written home exercise plan was detailed for each
               Saturday and Sunday.

Table 2.

Fugl-Meyer Evaluation of Physical Performance Scores

                  UE Scale        Comprehensive
                  Score (0-66)    Score (a) (0-126)

Pretreatment      17              69
Posttreatment     21              76
1-mo follow-up    21              76
6-mo follow-up    20              73

(a) The Comprehensive score is a combination of the Upper Extremity
(UE), Sensation, Joint Range of Motion, and Pain scale scores.

Table 3.

Graded Wolf Motor Function Test (GWMFT) Time Scores (a)

                  Time (s)

Pretreatment        118.47
Posttreatment       118.63
1-mo follow-up      118.59
6-mo follow-up      118.10

(a) Each of the 14 tasks of the GWMFT is timed (up to 120 seconds), and
the median time is calculated for the unaffected and affected upper
extremities. A low time score is optimal in either upper extremity. The
difference between the unaffected and affected upper-extremity median
time scores is recorded as the time score. The smaller the overall time
score, the more comparable the 2 upper extremities are in motor
ability.

Figure 1.

Client's Motor Activity Log results. Amount scale represents how much
the client attempts to use the involved upper extremity for functional
tasks. Scores range from 0 ("not used") to 5 ("used weaker arm as
normal as before the stroke"). The How Well scale represents the
quality of movement when the involved upper extremity is used for
functional tasks. Scores range from 0 ("not used") to 5 ("ability is
as good as before the injury or normal"). The Amount scale was not
assessed immediately posttreatment because the client's uninvolved
upper extremity was restrained for 3 weeks and she at least attempted
most tasks with the involved upper extremity.

                 Amount Scale    How Well Scale

Pretreatment         0.13             0.13
Posttreatment                         1.53
1-mo                 2                1.68
Follow-up
6-mo                 0.27             0.47
Follow-up

Note: Table made from bar graph.

Figure 2.

Caregiver's Motor Activity Log results. See Figure 1 footnote for
explanation of scales and scoring.

                 Amount Scale    How Well Scale

Pretreatment         0.21             0.22
Posttreatment        1.75             1.8
1-mo                 1.66             1.29
Follow-up
6-mo                 0.44             0.67
Follow-up

Note: Table made from bar graph.

Figure 3.

Graded Wolf Motor Function Test Functional Ability Scores. A mean
rating of the quality of movement for 14 upper-extremity motor tasks
completed with the involved upper extremity. Scoring is completed on
an 8-point Functional Ability scale, from 0="movement is not
attempted" to 7="normal movement."

Pretreatment          1.43
Posttreatment         1.57
1-mo Follow-up        1.5
6-mo Follow-up        2.36

Note: Table made from line graph.

* The test and instructions were e-mailed to us from David Morris, PT, MS, at the University of Alabama at Birmingham.

([dagger]) Velcro USA Inc, 406 Brown Ave, Manchester, NH 03103.

References

(1) Williams GR. Incidence and characteristics of total stroke in the United States. BMC Neurol. 2001;1:2.

(2) National Stroke Association. The cost of stroke to all Americans. Available at: http://www.stroke.org/brain-stat.cfm. Accessed December 14, 2001.

(3) Wolf SL, LeCraw D, Barton L, Jann B. Forced use of hemiplegic upper extremities to reverse the effect of learned nonuse among chronic stroke and head-injured patients. Exp Neurol. 1989;104: 125-132.

(4) Taub E, Uswatte G, Pidikiti R. Constraint-induced movement therapy: a new family of techniques with broad application to physical rehabilitation--a clinical review. J Rehabil Res Dev. 1999;36:237-251.

(5) Blanton S, Wolf SL. An application of upper-extremity constraint-induced movement therapy in a patient with subacute stroke. Phys Ther. 1999; 79:847-853.

(6) Taub E. Somatosensory deafferentation research with monkeys: implications for rehab medicine. In: Ince LP, ed. Behavioral Psychology in Rehabilitation Medicine: Clinical Applications. New York, NY: Williams & Wilkins; 1980:371-401.

(7) Taub E. Movement in nonhuman primates deprived of somatosensory feedback. Exerc Sport Sci Rev. 1977;4:335-374.

(8) Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. Compensation in recovery of upper extremity function after stroke: the Copenhagen stroke study. Arch Phys Med Rehabil. 1994;75:852-857.

(9) Newborn B. Quality of life for long term recovery in stroke. Topics in Stroke Rehabilitation. 1998;5 (2):61-63.

(10) Bach-y-Rita P. Theoretical and practical considerations in the restoration of function after stroke. Topics in Stroke Rehabilitation. 2001;8(3):1-15.

(11) DiMauro S. A survivor's perspective. In: Gillen G, Burkhardt A, eds. Stroke Rehabilitation: A Function Based Approach. St Louis, Mo: Mosby-Year Book; 1998:508-511.

(12) Broeks JG, Lankhorst GJ, Rumping K, Prevo AJH. The long-term outcome of arm function after stroke: results of a follow-up study. Disability and Rehabilitation. 1999;21:357-364.

(13) Sato Y, Kaji M, Tsuru T, Oizumi K. Carpal tunnel syndrome involving unaffected limbs of stroke patients. Stroke. 1999;30:414-418.

(14) Mayer TK. One-handed in a two-handed world. Topics in Stroke Rehabilitation. 2000;7(2):50-56.

(15) Morris D, Crago JE, DeLuca SC, et al. Constraint-induced movement therapy for motor recovery after stroke. Neurorehabilitation. 1997;9:29-43.

(16) Liepert J, Miltner WHR, Bauder H, et al. Motor cortex plasticity during constraint-induced movement therapy in stroke patients. Neurosci Lett. 1998;250:5-8.

(17) Liepert J, Bauder H, Wolfgang HR, et al. Treatment-induced cortical reorganization for stroke in humans. Stroke. 2000;31:1210-1216.

(18) Kunkel A, Kopp B, Muller G, et al. Constraint-induced movement therapy for motor recovery in chronic stroke patients. Arch Phys Med Rehabil. 1999;80:624-628.

(19) Taub E, Miller NE, Novack TA, et al. Technique to improve chronic motor deficit after stroke. Arch Phys Med Rehabil. 1993:74:347-354.

(20) Miltner WHR, Bauder H, Sommer M, et al. Effects of constraint-induced movement therapy on patients with chronic motor deficits after stroke: a replication. Stroke. 1999;30:586-592.

(21) Taub E. New discovery equals change in clinical practice [guest editorial]. J Rehabil Res Dev. 1999;36(3):vii-viii.

(22) Bauder H, Sommer M, Taub E, Miltner WHR. Effects of CI therapy on movement-related brain potentials [abstract]. Psychophysiol. 1999: 36(suppl 1):S31.

(23) Kopp B, Kunkel A, Muhlnickel W, et al. Plasticity in the motor system related to therapy-induced improvement of movement after stroke. Neuroreport. 1999;10:807-810.

(24) Uswatte G, Taub E. Constraint-induced movement therapy: new approaches to outcome measurement in rehabilitation. In: Struss DT, Winocur G, Robertson IH, eds. Cognitive Neurorehabilitation: A Comprehensive Approach. Cambridge, Mass: Cambridge University Press; 1999: 215-222.

(25) Ostendorf CG, Wolf SL. Effect of forced use of the upper extremity of a hemiplegic patient on changes in function. Phys Ther. 1981;61: 1022-1028.

(26) Morris DM, Taub E. Constraint-induced therapy approach to restoring function after neurological injury. Topics in Stroke Rehabilitation. 2001;8 (3) :16-30.

(27) Taub E, Morris DM. Constraint-induced movement therapy to enhance recovery after stroke. Curr Atheroscler Rep. 2001;3:279-286.

(28) van der Lee JH, Wagenaar RC, Lankhorst GJ, et al. Forced use of the upper extremity in chronic stroke patients: results from a single-blind randomized clinical trial. Sovke. 1999;30:2369-2375.

(29) Morris DM, Uswatte G, Crago JE, et al. The reliability of the wolf motor function test for assessing upper extremity function after stroke. Arch Phys Med Rehabil. 2001;82:750-755.

(30) Constraint-Induced Movement Therapy Research Group. Manual: Graded Wolf Motor Function Test [test instructions]. Birmingham, Ala: University of Alabama at Birmingham and Birmingham Veteran's Administration Center; August 2000.

(31) Wolf SL, Catlin PA, Ellis M, et al. Assessing Wolf motor function test as outcome measure for research in patients after stroke. Stroke. 2001;32:1635-1639.

(32) Constraint-Induced Movement Therapy Research Group. Manual: Upper Extremity Motor Activity Log (UE/MAL). Birmingham, Ala: University of Alabama at Birmingham and Birmingham Veteran's Administration Center; August 2000.

(33) Taub E, Uswatte G. Constraint-induced movement therapy and massed practice [letter to the editor]. Stroke. 2000;31:986-988.

(34) van der Lee JH, Lankhorst GJ, Bouter LM. Constraint-induced movement therapy and massed practice [letter to the editor; response]. Stroke. 2000;31 (4):988.

(35) Duncan PW, Propst M, Nelson SG. Reliability of the Fugl-Meyer assessment of sensorimotor recovery following cerebrovascular accident. Phys Ther. 1983;63:1606-1610.

(36) Folstein MF, Folstein SE, McHugh PR. "Mini-Mental State": a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.

(37) Tonkonogy J. The Brief Neuropsychological Cognitive Examination (BNCE) Manual. Los Angeles, Calif: Western Psychological Services; 1997:43-51.

(38) Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S Steglind S. The post-stroke hemiplegic patient. Scand J Rehabil Med. 1975;7:13-31.

(39) Bleecker ML, Bolla-Wilson K, Kawas C, Agnew J. Age-specific norms for the Mini-Mental State Exam. Neurology. 1988;38:1565-1568.

(40) Morris DM, Crago JE, Uswatte G, Taub E. Behavioral components of constraint-induced movement therapy [poster presentation]. Neurology Report. 2000;24:201.

N Bonifer, PT, MS, NCS, is Physical Therapy Lead, Spalding Rehabilitation Hospital, 900 Potomac St, Aurora, CO 80011 (USA) (Nancy. [e-mail]). Address all correspondence to Ms Bonifer.

KM Anderson, OTR, is Staff Occupational Therapist, Spalding Rehabilitation Hospital.

Both authors provided concept/idea/project design, writing, data collection, and subjects. Ms Bonifer provided data analysis, project management, fund procurement, and clerical support. Ms Anderson provided consultation (including review of manuscript before submission). David Reinhard, MD, Kim Gorgens, PhD, and Elena Draznin, MD, provided consultation. The authors greatly appreciate the efforts of Holle Balzer, PT, MPT, Peggy Kelly, OTR, and Fred Hudson, RN, for assistance with data collection. The authors thank David Morris, PT, MS, and Edward Taub, PhD, for providing expert clinical information via consultation and providing us with copies of the Motor Activity Log, Wolf Motor Function Test, and Graded Wolf Motor Function Test. The authors also thank Steven L Wolf, PT, PhD, FAPTA, for providing expert clinical information. A special thanks is extended to the client and caregiver, who made this case report possible.

The program protocol was approved by the HealthONE Alliance Institutional Review Board.

This project was supported by a grant from the Spalding Community Foundation and HealthONE Spalding Rehabilitation Hospital, Aurora, Colo.

This article was submitted February 22, 2002, and was accepted November 3, 2002.