| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Reports |
JM Wagner, PT, PhD, ATC, is a doctoral candidate, Program in Physical Therapy, Washington University School of Medicine, St Louis, Mo
CE Lang, PT, PhD, is Assistant Professor, Program in Physical Therapy, Program in Occupational Therapy, and Department of Neurology, Washington University School of Medicine, Campus Box 8502, 4444 Forest Park Pkwy, St Louis, MO 63108 (USA)
SA Sahrmann, PT, PhD, FAPTA, is Professor of Physical Therapy/Neurology/Cell Biology & Physiology, Program in Physical Therapy, Washington University School of Medicine
DF Edwards, PhD, is Associate Professor, Program in Occupational Therapy and Department of Neurology, Washington University School of Medicine
AW Dromerick, MD, is Associate Professor, Department of Neurology, Program in Occupational Therapy, and Program in Physical Therapy, Washington University School of Medicine
Address all correspondence to Dr Lang at: langc{at}wustl.edu
Submitted May 5, 2006;
Accepted January 30, 2007
 | Abstract |
|---|
Subjects and Methods: Upper-extremity sensorimotor impairments and reaching performance were evaluated in 39 subjects with hemiparesis at 2 time points: during the acute phase (8.7±3.6 [
±SD] days) and the subacute phase (108.7±16.5 days) after stroke. Ten subjects who were healthy (control subjects) were evaluated once. Regression analyses were used to determine which impairments were the best predictors of variance in reaching performance in the subacute phase after stroke.
Results: Only a small amount of variance (<30%) in reaching performance was explained at the subacute time point, using either acute or subacute impairments as predictor variables. Of the impairments measured, UE strength deficits were the strongest, most consistent predictors of the variance in reaching performance during the first 3 months after stroke.
Discussion and Conclusion: Surprisingly, the detailed clinical assessment of UE sensorimotor impairment, measured at the acute or subacute phase after stroke, did not explain much of the variance in reaching performance during the subacute phase after stroke. The findings that UE strength deficits (ie, decreased active range of motion and isometric force production) were the most common predictors of the variance in reaching performance during the first 3 months after stroke are consistent with the current viewpoint that impaired volitional muscle activation, clinically apparent as UE weakness, is a prominent contributing factor to UE dysfunction after stroke.
| Introduction |
|---|
 TopAbstract Introduction MethodResultsDiscussionConclusionsReferences |
|---|
Over the past decade, kinematic studies in people with chronic hemiparesis have yielded significant information about how movement control is altered after stroke and provided insight into compensatory movement control strategies (eg, see Cirstea and Levin14). Compared with clinical rating scales, kinematic studies offer a sensitive, quantitative assessment of the components of abnormal motor performance (ie, decreased efficiency and speed, poor accuracy, impaired interjoint coordination). Information from this type of assessment could be beneficial in investigating relationships between sensorimotor impairments and motor performance, as well as the predictive value of sensorimotor impairments in determining later motor performance. A better understanding of these issues in the acute and subacute phases after stroke would help clinicians in designing and implementing UE rehabilitation protocols, especially because this is the time when the majority of recovery occurs15 and the time when individuals with stroke receive rehabilitation.16,17
We recently examined the relationships between sensorimotor impairments and the kinematics of reaching performance in a group of subjects with hemiparesis during the acute phase after stroke.18 We found that sensorimotor impairments explained a moderate amount of variance in reaching performance and that measurements of UE strength (force-generating capacity) predicted the largest proportion of variance in reaching performance at this early time point after stroke (average of 9 days poststroke). In the current study, we extended our investigation to look at how sensorimotor impairments relate to reaching performance in the subacute phase after stroke and to look at how sensorimotor impairments measured in the acute phase after stroke relate to reaching performance measured several months later.
We chose to study forward reaching as a representative movement task because: (1) reaching is a fundamental component of many activities of daily living, (2) reaching requires the coordinated movement of multiple UE segments, and (3) reaching has been extensively studied in adults who are healthy and in people with chronic hemiparesis to better understand UE motor control.14,19–28 Of the many variables that can be used to quantify reaching performance using kinematic techniques (eg, see Tab. 2 in Cirstea and Levin14), we chose to quantify the speed, accuracy, and efficiency of the reach. We considered these 3 movement characteristics to be important because, presumably, once in the community setting, patients will not use their arm if movements are not timely or accurate, or if it takes too much effort or too many attempts to perform the movement. We hypothesized that reaching performance and sensorimotor deficits would recover from the acute phase to subacute phase and that strength deficits would be the largest and most consistent predictor of variance in reaching performance during the first few months after stroke.
| Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
Subjects were included if they had: (1) an ischemic or hemorrhagic stroke within 28 days of admission for inpatient rehabilitation; (2) persistent hemiparesis, as indicated by a score of 1 to 3 on the motor arm item of the National Institutes of Health Stroke Scale (NIHSS); (3) the presence of some UE voluntary activity, as indicated by the ability to move proximal or distal joints against gravity; (4) evidence of preserved cognitive function, as indicated by a score of 0 or 1 on the consciousness and communication items of the NIHSS, the ability to follow 2-step commands, as determined by clinical staff, and a score of 8 or lower on the Short Blessed Memory Orientation and Concentration Scale29; and (5) no injury or condition that limited use of the UEs prior to the stroke.
Subjects were excluded if they: (1) could not give informed consent, (2) had clinically significant fluctuations in mental status in the 72 hours prior to enrollment, (3) had hemispatial neglect, or (4) were not expected to survive 1 year due to other illnesses (eg, malignancy). Characteristics and descriptive lesion information of the subjects with hemiparesis are provided in Table 1. Descriptive statistics on performance on the Action Research Arm Test (ARAT) (Tab. 1) indicated that, on average, the affected UE was moderately affected at the acute time point and mildly to moderately affected by the subacute time point. The ARAT assesses UE activity limitation in people with hemiparesis after stroke,30 and ARAT scores are strongly correlated with scores on the UE scale of the Fugl-Meyer test (r=.91–.94).31,32
|
Subjects were randomly assigned to 1 of 3 treatment groups: (1) subjects who received 2 hours daily of occupational therapy, including compensatory techniques for activities of daily living, UE strength, and range of motion and traditional positioning (control group); (2) subjects who received dose-matched constraint-induced movement therapy (CIMT) (2 hours per day of CIMT-based occupational therapy, with 6 hours per day of constraint); or (3) subjects who received high-intensity CIMT (3 hours per day of CIMT-based occupational therapy, with 90% waking hours constraint). Subjects in the control group also participated in a circuit-training program allowing them to perform bilateral self-range-of-motion and functional activities in a supervised setting. Subjects in the dose-matched and high-intensity CIMT groups received a CIMT-based occupational therapy intervention that directed their attention and effort toward the hemiparetic UE and minimized the use of the uninvolved UE during functional activities. Subjects in those 2 groups also participated in a circuit-training program encouraging the use of the hemiparetic arm with a variety of UE and functional tasks. To discourage use of the unaffected hand outside of therapy sessions, subjects in the dose-matched and high-intensity CIMT groups wore a padded mitten (eg, constraint) on their uninvolved UE during the 2-week treatment period. The duration of occupational therapy and mitten use varied depending on treatment group assignment.
The overall emphasis of all treatments was task completion, and there was no special emphasis to practice tasks focusing solely on improving speed, accuracy, or efficiency of movement. Further physical therapy and occupational therapy after this 14-day period were provided outside the clinical trial as considered necessary by each subject's physician. Data in this report are from the same cohort of subjects with hemiparesis at 2 time points: a prerandomization baseline visit (acute time point: 8.7±3.6 days after stroke) and a subacute follow-up visit (subacute time point: 108.7±16.5 days after stroke). All data presented here were obtained from examiners who were masked to treatment group.
The age- and sex-matched control subjects (mean age=59.1 years, SD=12.5, range=70–78; 5 women, 5 men) were free of neurologic or orthopedic conditions that might affect their UEs. Subjects with hemiparesis were tested twice, at an acute time point and at a subacute time point. The control subjects were tested once. Subjects with hemiparesis were tested on their affected side (contralateral to the lesion), and control subjects were tested on their dominant side. The decision to test the dominant UE of control subjects was made a priori as part of the VECTORS clinical trial. The motor task was a forward reaching movement and not a task that involved hand dexterity, thus minimizing the possible influence of hand dominance in our data. Informed consent was obtained from all subjects prior to participation.
Measurement of reaching performance.
Subjects were studied performing a forward reaching task while seated in a straight-back chair. The trunk was stabilized to the back of the chair using a strap placed at chest height to minimize compensatory trunk movements.14 The start position (Fig. 1) was: UE resting on a pillow on the ipsilateral thigh, with the shoulder in approximately 0 degrees of flexion and extension and 0 degrees of internal rotation and with the elbow in 75 to 90 degrees of flexion. The wrist rested palm down, with the finger joints in slight flexion on the pillow. Minor modifications (eg, increased shoulder internal rotation) to the start position were allowed for some subjects to minimize any positional discomfort.
|
From the start position, subjects were instructed to reach forward as fast as possible and touch a 40-mm spherical target positioned 90% of arm's length directly in front of the affected (dominant) shoulder at shoulder height. Subjects were given 1 or 2 practice trials prior to recording to familiarize themselves with the task and the instructions. Three trials of reaching movement were recorded. Data collection was limited to 3 trials of reaching because the subjects with hemiparesis fatigued quickly, particularly at the acute time point and because the subjects were undergoing additional clinical assessments in conjunction with their participation in the VECTORS study.
Measurement of UE impairments.
We measured sensorimotor impairments that are commonly assessed by physical therapists and occupational therapists in stroke rehabilitation settings. Light touch sensation was measured at 4 locations on the arm using Semmes-Weinstein monofilaments.
The smallest monofilament sensed at each location was recorded and given an ordinal score using a previously described scale.33 Joint position sense was measured at the first metacarpophalengeal joint following standard clinical techniques.34 Joint position sense was scored as "intact" (correct response in 
3 of 5 trials) or "absent" (correct response in <3 of 5 trials). Elbow joint spasticity was measured using the Modified Ashworth Scale (MAS).35 Shoulder pain, as perceived at the beginning of the testing session, was measured using a visual analog scale.36
Maximal grip strength was measured using a Jamar handheld dynamometer.
Strength of the shoulder, elbow, and wrist flexor and extensor muscle groups was measured bilaterally using a handheld dynamometer (MICROFET2)
following a previously described protocol37 except that subjects were seated during testing. Maximal voluntary isometric strength values were recorded (in pounds) for each muscle group tested. Subjects who were unable to produce force against the dynamometer were given a score of 0 lb for that particular muscle group. The relative strength of each muscle group was calculated by taking the ratio of the maximal isometric force of the affected (nondominant) limb to the maximal isometric force of the less-affected (dominant) limb.
To assess the ability make isolated (fractionated) UE movements, subjects performed isolated flexion and extension movements of the shoulder, elbow, and wrist joints while 3-dimensional movements were recorded in the same manner as during the reaching task. The starting position for each isolated movement was with the shoulder in 0 degrees of flexion and extension, with the elbow extended and the wrist in neutral (ie, arm hanging by side). For isolated shoulder movement, subjects were instructed to flex the shoulder to 90 degrees while keeping the elbow and wrist still. For isolated elbow movement, subjects were instructed to maximally flex the elbow while keeping the shoulder and wrist still. For isolated wrist movement, subjects were instructed to extend the wrist while keeping the shoulder and elbow still.
Data Analysis
Offline, EvaRT, and Kintrak software* were used to extract position, velocity, and angular data during the reaching task. Because of the redundant camera system, no markers were lost. Data were low-pass filtered at 6 Hz. For the reach, start of movement was defined as the time at which the tangential wrist velocity exceeded 5% of maximum velocity. End of the first phase of reach was defined as the time at which the wrist velocity dropped to a minimum prior to subsequent corrective movements. For each trial, we quantified the speed, accuracy, and efficiency of reaching—3 characteristics of performance that are important for normal function. We considered an efficient reach to be one in which the hand moves directly to the target without extraneous or abnormally circuitous movements.
Peak wrist velocity, endpoint error, and reach path ratio were used to quantify the speed, accuracy, and efficiency of reach, respectively. Peak wrist velocity was the maximum tangential linear velocity of the wrist attained between the start of movement and the end of the first phase of reach. Endpoint error was the 3-dimensional distance from the index finger to the center of the target at the end of the first phase of reach. Reach path ratio was calculated as the ratio of the length of the actual wrist path traveled to an ideal straight line between the start position and target touch. For subjects who were unable to touch the target, the reach path ratio was calculated from start of movement to the time and position where the index finger was closest to the target. A reach path ratio of 1 represents a straight path (normal), whereas a reach path ratio >1 represents either an abnormally curved path or multiple attempts to touch the target. In this study, the reach path ratio was chosen as a measure of efficiency because, in our subjects with hemiparesis, higher reach path ratios were due to multiple attempts to touch the target and were not due to decoupling in the shoulder and elbow joint, as they are in subjects with cerebellar damage.38 In addition, the time from start of movement to target touch was calculated and reported as movement time.
To quantify a subject's ability to perform isolated (fractionated) joint motion, individuation indexes were calculated for the shoulder, elbow, and wrist using the angular excursions measured during the isolated movement tasks.33,39–41 The individuation index is a measure of how well the instructed joint is able to move by itself, without other joints moving. The individuation index will be close to 1 for an ideally isolated movement in which the instructed joint moved with no movement at noninstructed joints, and it will be closer to 0 the more the noninstructed joints moved with the instructed one. Subjects who had little to no volitional movement, defined as less than 10% of the control group's average angular excursion, were given a score of 0 for the individuation index for that segment. A composite individuation score33 was calculated as the average of the individuation indexes for the shoulder, elbow, and wrist, where each joint was equally weighted. The composite individuation score reflects the ability to make fractionated UE movements (ie, move out of stereotypical synergistic patterns).33
Upper-extremity strength deficits were quantified 3 ways. First, maximal grip strength was used to quantify distal strength. Second, a composite UE strength score18 was calculated because the strength scores for the various muscle groups were correlated with each other. Relative strength (the ratio of the strength of the affected [nondominant] limb to strength of the less-affected [dominant limb]) scores at the shoulder, elbow, and wrist were averaged (equally weighted) to obtain a composite UE strength score for each subject. Thus, the composite strength score reflects the distributed isometric strength of the UE relative to the nonparetic (dominant) side. This method was useful for subjects who could produce force against gravity and manual resistance but not for those subjects who could move against gravity but were unable to produce force against manual resistance (eg, a floor effect).
We therefore used measurements of active range of motion (AROM) as our third way to quantify strength.42 The AROM at each joint was calculated as the total average excursion of the joint against gravity during the isolated movement tasks. Because subjects were instructed to flex their shoulder to 90 degrees during the isolated movement task, maximum shoulder flexion values were limited to 90 degrees. A composite AROM score, reflecting the sum of the measurements of AROM at the shoulder, elbow, and wrist, was calculated for subsequent statistical analyses. The AROM measurements of the shoulder, elbow, and wrist were equally weighted in the calculation of the composite AROM score because the AROM measurements of the 3 joints were significantly correlated with each other. Composite AROM was related to composite strength (r=.45, P=.05), supporting the use of AROM as a lower-end measure of strength.
A composite UE light touch sensation score was calculated by averaging (equally weighted) the ordinal scores from the 4 test sites because strong correlations existed among the sites. Joint position sense was coded for statistical analysis (present=0 and absent=1).
Statistica software|| was used for all statistical analysis and the criterion for significance as set at P<.05. Distributions of variables were tested for normality using the Shapiro-Wilk W test. Some of the sensorimotor impairment and kinematic variables were not normally distributed and needed to be transformed for further statistical analyses. The type of transformation done on a variable was chosen by examining the raw distribution and then selecting the transformation that would best minimize skewness. Four sensorimotor impairment variables were transformed as follows: composite strength using the natural log function, composite AROM using the natural log function, composite individuation score using the natural log function, and the MAS scores using percentile ranks. Two kinematic variables were transformed as follows: reach path ratio and end point error using the natural log function. All statistical analyses on these 6 variables were done using the transformed data.
We used t tests to look for differences in reaching performance and composite measures of impairment (eg, composite UE strength) between: (1) subjects with hemiparesis and control subjects and (2) the subacute and acute time points for the subjects with hemiparesis. A series of 2 x 3 (time x joint) repeated-measures analyses of variance (ANOVAs) were used to test for changes over time in individuation, AROM, and UE strength in the subjects with hemiparesis. A series of 2 x 3 (group x joint) repeated-measures ANOVAs were used to test for differences in individuation, AROM, and UE strength between the subjects with hemiparesis and the control subjects at the 2 time points. Tukey Honestly Significant Difference tests were used for post hoc comparisons when significant main and interaction effects were present. Spearman rank order and Pearson product moment correlations were used to test for relationships between the various sensorimotor impairments as well as the relationships between reaching performance and measures of UE impairment. Bonferroni corrections were applied to adjust for multiple comparisons.
Stepwise linear multiple regression, which predicts the maximum amount of variance (R2) with a minimum number of independent variables,43 was used as an exploratory tool to identify which sensorimotor impairments were the strongest predictors of reaching performance in the early months after stroke. Specifically, a series of forward stepwise linear multiple regression analyses were used to determine whether UE impairments measured at the subacute time point could predict the variance in the speed, accuracy, and efficiency of reaching performance at the same time point and whether UE impairments measured at the acute time point could predict the variance in the speed, accuracy, and efficiency of reaching performance at the subacute time point.
| Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
Reaching Performance and Sensorimotor Impairments at the Acute Time Point
During the acute phase after stroke, reaching performance was generally poor (black bars, Fig. 2), such that the subjects with hemiparesis had longer movement times (P=.002), lower peak wrist velocities (P<.001), larger endpoint errors (P=.001), and higher reach path ratios (P=.02) compared with the control subjects. The hemiparetic subjects had deficits in all sensorimotor measures at the acute time point (black bars, Fig. 3). Composite light touch sensation and joint position sense were impaired in 59% and 33% of the subjects with hemiparesis, respectively. Forty-nine percent of the subjects with hemiparesis had some degree of elbow spasticity, but it was typically mild (MAS score=1). Twelve (31%) of the subjects with hemiparesis reported shoulder pain, with 11 of those subjects reporting mild to moderate pain on the visual analog scale (VAS score=2 or 4). The subjects with hemiparesis had deficits in AROM (main effect for group: F=11.47; df=1,47; P=.001), strength (main effect for group: F=45.90; df=1,47; P<.001), and the ability to isolate (fractionate) movements (main effect for group: F=13.14; df=1,47; P<.001) compared with the control subjects.
|
|
Figure 3 shows the sensorimotor impairments in the subjects with hemiparesis at the acute (black bars) and subacute (blue bars) time points. Where appropriate, measurement of sensorimotor impairments for control subjects also are shown (white bars). The presence and severity of sensorimotor impairment were generally reduced at the subacute time point compared with the acute time point, except for spasticity and shoulder pain. Composite light touch sensation (Fig. 3A) and joint position sense (Fig. 3B) recovered such that the subjects with hemiparesis did not significantly differ from the control subjects at the subacute time point (P=.12 and P=.36, respectively). The severity of elbow spasticity as represented by MAS scores (Fig. 3C) and shoulder pain (Fig. 3D) in the subjects with hemiparesis increased (worsened) from the acute time point to subacute time point (P=.01 and P=.05, respectively). Improvements were noted in AROM (main effect for time: F=32.80; df=1,38; P<.001), strength (main effect for time: F=30.58; df=1,38; P<.001), and isolated movement control (main effect for time: F=11.96; df=1,38; P=.001) (Fig. 3E–G). Recovery was incomplete because deficits remained in AROM (main effect for group: F=4.99; df=1,47; P=.03), strength (main effect for group: F=5.50; df=1,47; P=.02), and isolated movement control (main effect for group: F=5.94; df=1,47; P=.02) in the subjects with hemiparesis compared with the control subjects.
Relationships Between Impairments and Reaching Deficits at the Subacute Time Point
Relationships between sensorimotor impairments and reaching performance at the subacute time point were examined using Spearman correlation coefficients (Tab. 2, top panel). Based on our sample size, correlation coefficients greater than .31 were statistically significant at the P<.05 level. Composite UE strength and composite AROM were related to speed, accuracy, and efficiency of reaching such that the greater the UE composite strength and AROM, the faster, more accurate, and more efficient the reach. All other sensorimotor impairments were not significantly related to reaching performance.
|
.25 with reaching performance. When available, composite impairment variables were entered into the regression to optimize the statistical analysis. Ultimately, 3 independent variables (composite strength, composite AROM, and the MAS scores) were selected for inclusion in the model. Multicolinearity in each regression equation was evaluated by calculating the tolerance of each independent variable,44 with the results indicating acceptable levels of redundancy among the 3 independent variables. At the subacute time point, 27% of the variance in reaching speed was predicted by composite strength (Tab. 3, middle panel), with no other sensorimotor impairments entered into the model. For reaching accuracy, 15% of the variance was predicted by composite strength (11%) and MAS scores (4%). For reaching efficiency, 26% of the variance was predicted by composite strength (24%) and MAS scores (2%). Thus, at the subacute time point, sensorimotor impairments predicted only a small amount of the variance in reaching performance, regardless of which measure of performance was used, and composite strength was the common contributing factor.
|
Regression analyses were performed to determine which impairment variables measured at the acute time point best predicted variance in reaching performance at the subacute time point. As above, composite strength, composite AROM, and MAS scores were entered into the model. Seventeen percent of the variance in reaching speed at the subacute time point was predicted by composite AROM values at the acute time point (Tab. 3, bottom panel), with no other sensorimotor impairment entered into the model. Sixteen percent of the variance in reaching accuracy at the subacute time point was predicted by composite AROM (12%) and MAS scores (4%) at the acute time point. Twenty-five percent of the variance in reaching efficiency at the subacute time point was predicted by composite AROM (19%), composite strength (3%), and MAS scores (3%) at the acute time point. Thus, sensorimotor impairments measured at the acute time point predicted only a small portion of the variance in reaching performance at the subacute time point, regardless of which characteristic of movement performance was used as the dependent variable.
Do Sensorimotor Impairments Predict Greater Proportions of Variance for UE Function Than for Specific Characteristics of Motor Performance?
We chose reaching as a representative movement task, yet reaching is only one movement in an enormous repertoire of possible UE movements. To determine whether the relatively limited predictive value of sensorimotor impairments was due to selecting a single movement task, we performed additional regression analyses in which the ARAT scores were entered into the regression as the dependent variable in lieu of peak wrist velocity, end point error, and reach path ratio (Tab. 4). The ARAT is a functional assessment of UE movement across a range of movement tasks (ie, gross movement, grasping, gripping, and pinching).
|
Contribution of Grip Strength Impairment to the Variance in Reaching Performance
Grip strength has been proposed as a surrogate marker for recovery of UE function after stroke.4,7,10 In our sample, grip strength measurements were correlated to composite strength measurements at the acute time point (r=.69, P<.05) and at the subacute time point (r=.76, P<.05). At both acute and subacute time points, grip strength was related to speed (r=.47 and r=.58, respectively) and accuracy (r=–.44 and r=–.44, respectively) of reaching such that a greater grip strength was associated with a faster and more accurate reach. At the subacute time point, grip strength also as related to efficiency of reaching (r=–.37) such that greater grip strength was associated with a more efficient reach.
To determine whether grip strength was a better predictor of the variance in reaching performance than our composite UE strength measure, we performed additional regression analyses in which grip strength scores were entered as an independent variable in lieu of UE composite strength. Entering grip strength measured at the subacute time point as an independent variable to predict the variance in reaching performance at the subacute time point, the amounts of explained variance in peak wrist velocity, endpoint error, and reach path ratio were 33%, 9%, and 22%, respectively. Entering grip strength measured at the acute time point to predict the variance in reaching performance at the subacute time point, the amounts of explained variance in peak wrist velocity, endpoint error, and reach path ratio were 17%, 16%, and 25%, respectively. Thus, the same amount of variance in reaching performance was explained whether grip strength or composite UE composite strength was entered into the regression model.
| Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
To our knowledge, this is the first study to examine specific characteristics of movement performance during a functional UE task to assess motor recovery during the acute and subacute phases after stroke. Studies that have predicted motor recovery following stroke have primarily used clinical measures of UE motor performance,2,6–9,13 UE functional recovery,10,11,13,45 or global disability1 as the measure of motor recovery. We chose to study specific characteristics of reaching instead of clinical measures of motor and functional performance as markers of motor recovery because we believed that a better understanding of the characteristics of functional movement (timeliness, accuracy, efficiency) will provide additional insight about central nervous system control of movement after stroke, while providing clinicians with information that may help guide the selection of rehabilitation techniques and exercise parameters (eg, focusing on strength to improve speed of movement).
Reaching performance of the subjects with hemiparesis recovered such that only peak wrist velocity significantly differed between the subjects with hemiparesis and the control subjects. The improvement of reaching accuracy and efficiency was accompanied by a decline in the presence and severity of the majority of sensorimotor impairments; however, sensorimotor impairment recovery was incomplete because there were persistent deficits in strength, AROM, and isolated movement control, as well as increased elbow spasticity and shoulder pain at the subacute time point (Fig. 3).
Two interpretations of these findings are relevant to clinicians and scientists interested in UE motor recovery following stroke. First, our results imply that the performance of a functional movement can be normal or near-normal despite the presence of underlying sensorimotor impairments. This may reflect the idea that not all functional movements require full sensorimotor capacity. For example, our subjects with hemiparesis were able to meet the sensorimotor requirements of the reaching task despite the presence of measurable sensorimotor impairments. Our control subjects used approximately 35 degrees of shoulder flexion, 15 degrees of elbow flexion, and 10 degrees of wrist extension to reach the target. These values are within the range of available AROM for the subjects with hemiparesis (Fig. 3), suggesting that, despite limitations in maximal AROM, subjects with hemiparesis were able to meet the AROM demands of the reaching task. Second, our results imply that the normalization of specific characteristics of reaching (ie, accuracy, efficiency) does not indicate full UE sensorimotor recovery, suggesting that kinematic analyses of UE movement may not adequately describe motor recovery following stroke. Additional research is needed to determine which outcome measure (ie, kinematic variable versus standardized clinical examinations), or combination of measures, best reflects UE motor recovery following stroke.
Sensorimotor impairments predicted similar amounts of variance (R2
43%) for ARAT scores and 2 of the 3 characteristics of reaching (speed and accuracy) at the acute time point; however, by the subacute time point, sensorimotor impairments predicted approximately twice the amount of variance in ARAT scores (R2=.57) than for speed and accuracy of reaching (R2=.27 and R2=.15, respectively). These findings indicate that clinical measures of sensorimotor impairments are more predictive of the variance in UE function (eg, ARAT scores) than for a single functional motor task at the subacute phase. The subacute time point is the time when our subjects were living in their homes and interacting with their communities. One explanation for this outcome may be that, during the early days after stroke, when UE use is limited,46 a few movement tasks (eg, reaching) are sufficient to assess motor control of the UE, but later in the course of recovery, as movement capacity and complexity increase, more motor tasks are needed to capture the motor performance of the UE. Future studies are needed to investigate how sensorimotor impairments may or may not contribute to UE activity and participation outside the hospital, clinic, or laboratory environment.
We believe it is noteworthy that sensorimotor impairments, whether measured at the acute time point or the subacute time point, were not found to be strong predictors of the variance in speed, accuracy, or efficiency of reaching at the subacute phase after stroke. Thus, our detailed clinical examination of sensorimotor status (an examination that is more detailed than done typically at bedside) did not adequately capture the variance of a common functional motor task.
One explanation for the low amount of explained variance is that the sensorimotor impairments evaluated during a typical examination are lacking (ie, some potentially important impairments that may predict how people perform functional movements are missing). For example, as is typically done bedside, strength was evaluated isometrically with one contraction of each muscle group, but it may be that more dynamic assessments of concentric and eccentric muscle strength and muscle power (eg, via isokinetic testing) or assessments of muscle endurance are more informative. Other potentially important impairments that were not assessed include deficits in motor planning and non-motor factors (ie, attention deficits, depression). Further research is needed to evaluate the importance of these other factors for predicting the variance in UE motor performance in people with poststroke hemiparesis.
It is possible that the limited predictive ability of our model (R2
.57) was due to using multiple linear regression analyses (MRA) to predict the variance of complex motor tasks (ie, reaching performance and motor function as assessed by the ARAT). In MRAs, the proportion of total variance explained in the dependent variable is contingent upon the particular set of independent variables used in the analysis, and the number of independent variables is constrained as a function of sample size, where it is generally recommended that there should be 10 to 20 subjects per predictor variable. Due to our sample size (N=39) we were restricted to about 4 predictor variables. We considered both sample size and zero-order correlations when selecting the 3 independent variables (composite strength or grip strength, composite AROM, MAS scores) for our MRAs. It is possible that a larger proportion of variance could have been explained for these complex motor tasks if different or additional independent variables had been entered into the model. Further research, with larger sample sizes, is needed to determine whether other combinations of independent variables yield larger proportions of explained variance for these complex motor tasks.
It is possible that the low amount of variance explained by our model was due to using linear multiple regression to predict the variance in motor performance, when the relationships between sensorimotor impairments and reaching performance may have been better captured by a nonlinear regression model. Although we cannot rule out this possibility, it is unlikely that a nonlinear model would have been a better fit because visual inspection of scatter plots did not reveal curvilinear relationships between individual sensorimotor impairments and reaching performance or ARAT scores.
Regardless of whether movement performance was quantified as speed, accuracy, or efficiency, UE strength deficits, as measured by composite AROM or composite strength, were the most common predictors of the variance in reaching performance during the first 3 months after stroke. These results are consistent with previous reports linking the severity of UE weakness with the outcome of UE movement and function.6,7,9-11,45,47,48
Maximal isometric grip strength was found to be as strong as a predictor as a composite measure of UE isometric strength for predicting reaching performance at the acute and subacute time points. These results suggest that future studies of reaching performance could be more efficiently conducted by obtaining grip strength measures in lieu of maximal isometric strength testing. Additional research is needed to determine whether maximal grip strength is a good proxy for other strength metrics (ie, maximal dynamic peak torque) or whether equivalent results would be found when studying other UE tasks.
The ability to isolate (fractionate) movement, as measured by individuation indexes, was not a significant predictor of the variance in peak wrist velocity, endpoint error, or reach path ratio in our cohort of subjects at both acute and subacute time points. These results differ from those of a report of patients with chronic hemiparesis,33 where individuation indexes, not strength, were the best predictors of reach path and endpoint error. We postulate that the different relationships observed between reaching performance and sensorimotor impairments in our subjects with acute or subacute hemiparesis versus the patients with chronic hemiparesis may be linked to the disparity in the chronicity of hemiparesis between the groups (subacute hemiparesis=109±17 days poststroke, chronic hemiparesis=32±62 months poststroke). One possible explanation is that, in the chronic phase after stroke, spared components of the descending motor system may have been able to activate motor units to produce force, but may not have been able to activate them selectively. Thus, poor isolated movement control (ie, the inability to move out of stereotypic synergies) may limit motor performance more than strength deficits in people with chronic hemiparesis.
Our subjects represent a subset of patients with stroke and were selected based on the presence of hemiparesis. Although our subjects are reasonably representative of subjects with poststroke hemiparesis seen in inpatient rehabilitation facilities in the United States, they appear to be less severely affected overall than the subjects often used to examine movement control in people with chronic hemiparesis.14,22,24,49,50 For example, 33% (13/39) of our subjects at the subacute time point had a maximum score of 57/57 on the ARAT. Caution should be taken when interpreting the present results with respect to previous work because the same movement control problems may not be present in subjects with mild-to-moderate hemiparesis compared with subjects with more severe hemiparesis.24
| Conclusions |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
| Footnotes |
|---|
The study was approved by the Institutional Review Board at Washington University School of Medicine.
This work, in part, was presented at the Combined Sections Meeting of the American Physical Therapy Association; February 1–5, 2006; San Diego, Calif.
This work was supported by National Institutes of Health grants NS41261 and HD047669, James S. McDonnell Foundation grant 21002032, and the Foundation for Physical Therapy Promotion of Doctoral Studies Scholarship.
* Motion Analysis Corp, 3617 Westwind Blvd, Santa Rosa, CA 95403. 
North Coast Medical, 18305 Sutter Blvd, Morgan Hill, CA 95037.
Pro-Med Products Inc, 6445 Powers Ferry Rd, #199, Atlanta, GA 30339.
Hoggan Health Industries, 8020 S 1300 West, West Jordan, UT 84088.
|| StatSoft Inc, 2300 E 14th St, Tulsa, OK 74104.
| References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. F. Levin, J. A. Kleim, and S. L. Wolf What Do Motor "Recovery" and "Compensation" Mean in Patients Following Stroke? Neurorehabil Neural Repair, May 1, 2009; 23(4): 313 - 319. [Abstract] [PDF]
|
|
|
|
 |
|
 J. A. Beebe and C. E. Lang Active Range of Motion Predicts Upper Extremity Function 3 Months After Stroke Stroke, May 1, 2009; 40(5): 1772 - 1779. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M Wagner, J. A Rhodes, and C. Patten Reproducibility and Minimal Detectable Change of Three-Dimensional Kinematic Analysis of Reaching Tasks in People With Hemiparesis After Stroke Physical Therapy, May 1, 2008; 88(5): 652 - 663. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
