Arm and Trunk Movement Kinematics During Seated Reaching Within and Beyond Arm's Length in People With Stroke: A Validity Study Ching-yi Wu, Rong-jiuan Liing, Hsieh-ching Chen, Chia-ling Chen and Keh-chung Lin PHYS THER. 2014; 94:845-856. Originally published online January 30, 2014 doi: 10.2522/ptj.20130101
The online version of this article, along with updated information and services, can be found online at: http://ptjournal.apta.org/content/94/6/845 Collections
This article, along with others on similar topics, appears in the following collection(s): Kinesiology/Biomechanics Stroke (Geriatrics) Stroke (Neurology) Tests and Measurements
e-Letters
To submit an e-Letter on this article, click here or click on "Submit a response" in the right-hand menu under "Responses" in the online version of this article.
E-mail alerts
Sign up here to receive free e-mail alerts
Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Research Report Arm and Trunk Movement Kinematics During Seated Reaching Within and Beyond Arm’s Length in People With Stroke: A Validity Study Ching-yi Wu, Rong-jiuan Liing, Hsieh-ching Chen, Chia-ling Chen, Keh-chung Lin
Background. Kinematic analysis is commonly used to objectively measure upper extremity movement performance after stroke. However, the concurrent validity and predictive validity of arm-trunk kinematics during reaching within and beyond arm’s length have not been studied. Objective. The aim of this study was to estimate the concurrent validity of kinematic measures before and after treatment and the predictive validity for reaching within and beyond arm’s length after stroke.
Design. This was a secondary analysis study. Methods. Ninety-seven participants with stroke (mean age⫽55.9 years [SD⫽10.9]) received intensive treatment every weekday for 3 to 4 weeks. Kinematic reaching tasks and the Wolf Motor Function Test (WMFT) were used before and after treatment. The validity of the kinematic measures was estimated in relation to WMFT scores.
Results. Of the 8 kinematic variables that were measured, index movement time
before treatment (R2⫽.227–.362) and trunk movement time and trunk displacement after treatment (R2⫽.095–.346) had the strongest association with the WMFT at both reaching distances. Trunk movement time and trunk displacement before treatment explained 6.9% to 14.9% of the variance in the WMFT after treatment. Kinematic variables explained 6.9% to 49.3% and 9.4% to 38.7% of the variance in the WMFT during a task within arm’s length and beyond arm’s length, respectively.
Limitations. The study has limited generalizability. Conclusions. Different kinematic variables may partially reflect motor function before and after treatment to a limited degree. Although the predictive validity was modest, trunk movement may be considered a prognostic determinant of motor function after treatment. A reaching task within arm’s length may be a more suitable measure of kinematic performance for describing motor function than a reaching task beyond arm’s length.
C. Wu, ScD, OTR/L, Department of Occupational Therapy and Graduate Institute of Behavioral Sciences, College of Medicine, Chang Gung University, Taoyuan, Taiwan, and Healthy Ageing Research Center, Chang Gung University, Taoyuan, Taiwan. R. Liing, PT, PhD, School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan. H. Chen, PhD, Department and Graduate Institute of Industrial Engineering and Management, National Taipei University of Technology, Taipei, Taiwan. C. Chen, PhD, MD, Department of Industrial Engineering, and Rehabilitation, LinKou Chang Gung Memorial Hospital, Taoyuan, Taiwan. K. Lin, ScD, OTR/L, School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan, and Division of Occupational Therapy, Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, 17, F4, Xuzhou Road, Taipei 100, Taiwan. Address all correspondence to Dr Lin at: [email protected]. [Wu C, Liing R, Chen H, et al. Arm and trunk movement kinematics during seated reaching within and beyond arm’s length in people with stroke: a validity study. Phys Ther. 2014;94:845– 856.] © 2014 American Physical Therapy Association Published Ahead of Print: January 30, 2014 Accepted: January 27, 2014 Submitted: March 10, 2013
Post a Rapid Response to this article at: ptjournal.apta.org June 2014
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
845
Arm and Trunk Movement Kinematics in People With Stroke
I
mpaired upper extremity (UE) function is a common deficit after stroke. About 70% of patients with stroke experience hemiparesis with an impaired UE,1 and 55% to 75% will still have limitations in UE function 3 to 6 months after stroke.2,3 Upper extremity motor impairment often leads to poor movement control and long-term disabilities.4 Because of persistent UE dysfunction after stroke, an accurate and precise measurement of UE performance is crucial for rehabilitation. Kinematic analysis has been commonly used in objective and quantitative measures of UE movement performance after stroke. Kinematic performance may reveal aspects of movement planning in terms of the joint angle coordinate strategy and endpoint coordinate strategy.5 In the joint angle coordinate strategy, the central nervous system plans the movement of the UE around a set of intrinsic coordinates of the body (eg, shoulder, elbow, and wrist joints) to arrive at the target, whereas in the endpoint coordinate strategy, extrinsic coordinates in the space are used to direct the hand to the final (endpoint) location (ie, the target). In addition, kinematic variables are often used to detect trunk compensatory movement (trunk displacement and trunk movement time).5,6 Kinematic assessment has been widely applied for discriminating differences in movement strategies between people with stroke and adults who are healthy7,8 and for evaluating the effects of various therapeutic interventions on the UE9 –11 or UE recovery after stroke.12,13 Incorporating kinematic assessment into clinical evaluation is necessary to understand the impact of a disability or to capture or predict important effects of treatments on various health-related conditions, ranging from motor control strategies to activity participation.14
846
f
Physical Therapy
The ability of kinematic data to function as a discriminative, monitoring, or evaluative measurement tool depends on the presence of sound psychometric properties.15 Some studies have established good relative15,16 and absolute15 (eg, minimal detectable change value) reliability of kinematic measures. Although several studies using kinematic variables have shown a relationship between sensorimotor impairment and motor dysfunction,17–21 research investigating the concurrent validity of kinematics has been limited,14,22,23 and no research using comprehensive kinematic variables to correlate motor impairment and motor function in people with stroke has been done. Information about how different types of kinematic variables at various stroke phases (eg, before and after intensive intervention) relate to motor function is of great value for better selection and interpretation of kinematic variables reflecting motor function in rehabilitation research. Furthermore, Massie et al22 pointed out the need for studying how kinematic variables predict movement performance over time (ie, the predictive validity of kinematics). Adding plausible kinematic predictors to established predictive models19,24 may improve the determination of prognosis.25 In 2 previous studies of kinematic validity, variables relevant to joint angle coordinate strategy (eg, recruitment of shoulder flexion, shoulder abduction, and elbow extension) were used to investigate the concurrent validity of kinematics reflecting motor impairment23 and motor function.22 In another kinematic validity study, selected endpoint coordinate strategy variables (eg, endpoint movement trajectory, movement time, and time to peak velocity) were used.14 Given that the central nervous system may not program movements exclusively through one strategy or the other26
and that task demands (eg, target distance within or beyond arm’s length) and health conditions (eg, adults who are healthy or people with stroke) may be factors,26 both types of strategy variables must be included in studies of kinematic validity. Trunk compensatory movement during reaching has been considered a critical factor in studies of the UE performance of people with stroke27–30 because they often recruit additional degrees of freedom (trunk movement) to compensate for UE motor deficits, thereby hindering the potential for motor recovery after stroke.27 People with stroke may use excessive trunk movement to compensate for a limited range of motion in the UE29; specific rehabilitation training may improve voluntary joint range and coordination and decrease trunk movement.10,31 Previous studies of kinematic validity or the relationships between motor impairment or function and kinematic measures also suggested that trunk displacement is an important variable14,23 for distinguishing the severity of motor impairment.23 It also may be important to incorporate trunk compensatory movement variables in further studies of the concurrent validity of kinematic measures before and after intensive training. In all validity studies, UE tasks with a target distance within approximately 80% to 100% of arm’s length have been used.14,22,23 Different reaching distances induce people with stroke to use different coordinate strategies, possibly together with trunk compensatory strategies, to accomplish reaching tasks.29,32 Reaching to a target within arm’s length involves the endpoint control and interjoint coordination of the UE; that is, central commands preplan a straight, efficient, and smooth endpoint trajectory and then coordinate
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke the motions of the shoulder and elbow to reach the target.33,34 Reaching beyond arm’s length (see “Method” section) requires trunk participation, and the amount of trunk displacement or recruitment is inversely associated with joint recruitment of the shoulder and elbow.28,31,35 Whether the kinematic performance of tasks within arm’s length can fully represent a patient’s coordinate strategies and trunk recruitment and which kinematic task (within or beyond arm’s length) appropriately reflects a patient’s motor performance have yet to be determined. The purpose of this study was to investigate the validity of arm and trunk movement kinematics during seated forward reaching. We attempted to estimate the concurrent validity of using arm-trunk kinematics against Wolf Motor Function Test (WMFT) scores before and after treatment and the validity of these kinematic measures for predicting WMFT scores for reaching tasks within and beyond arm’s length after stroke. The WMFT, with wellknown psychometric properties,36,37 was used as the gold standard for UE motor function measurement.
Method Participants In the present study, we used pooled data for a secondary analysis. The data were drawn from 4 previous38 – 41 and ongoing randomized controlled trials of constraint-induced therapy, bilateral arm training, robot-assisted arm training, and conventional training. Ninety-seven participants at 4 hospitals in Taiwan were recruited. The inclusion criteria used in the randomized controlled trials were as follows: first stroke, Brunnstro ¨ m stage 3 or above for the proximal and distal UE, and ability to follow instructions for the evaluation of arm-trunk kinematics. The exclusion criteria were as follows: cognitive impairment June 2014
(Mini-Mental State Examination score of ⬍21), excessive spasticity at any joint of the arm (modified Ashworth scale score of ⬎2), and participation in experimental rehabilitation or drug studies within the preceding 6 months. All participants signed a consent form. Only data from participants who were evaluated with the WMFT and a kinematic assessment before and after intervention with tasks both within arm’s length and beyond arm’s length were included in the present study. Interventions and Procedure The randomization procedures used in the trials were similar. Participants were randomly assigned to the constraint-induced therapy, bilateral arm training, robot-assisted arm training, or conventional training treatment group for 90 to 120 minutes of intervention every weekday for 3 to 4 weeks. Six evaluators with training in administering clinical assessments and unaware of the group assignments performed the clinical evaluations. Clinical Assessments The WMFT, which includes 15 function-based tasks, was used to evaluate UE function before and after 3 to 4 weeks of treatment. Tasks 1 to 6 of the WMFT are timed joint segment movements, and tasks 7 to 15 are timed integrative functional movements. The WMFT has 2 main scores: the mean performance time (WMFT-TIME) and the mean functional ability score (WMFT-FAS). The reliability and validity of the WMFTTIME and the WMFT-FAS have been well established.36,37 In the WMFTTIME, the instructions for each task focus on the speed of completion. In the WMFT-FAS, task completion and movement strategies (eg, movements made in synergy) are rated on a 6-point ordinal scale (from 0 to 5 points).
Measurement of Reaching Performance A reaching task is a fundamental component of many daily activities, requires the coordinated movement of multiple UE joints, and is thus appropriate for use as an experimental task.19 Reaching to press a desk bell was used as the kinematic task. The bell was placed along the participant’s midsagittal plane at a distance corresponding to the distance from the medial border of the axilla to the distal wrist crease (a reaching distance within arm’s length) and at 125% of arm’s length (a reaching distance beyond arm’s length). Arm’s length was defined as the distance from the medial border of the axilla to the tip of the third finger, as measured with the arm hanging down alongside the trunk. The instruction was to use the index finger of the affected arm to press the desk bell as quickly as possible.42 Each participant sat on an adjustable chair with the seat height set to 100% of the lower leg’s length, as measured from the lateral knee joint to the foot. An adjustable table was adjacent to the chair. The table height was adjusted to 5 cm below the elbow while the participant sat on the chair. The initial position of the hand was on the edge of the table with the elbow flexed at 90 degrees. After one practice trial, the participant was instructed to reach to press the desk bell as quickly as possible in 3 trials. Kinematic Analysis of Reaching Performance Arm and trunk movements were modeled with 13 markers, which were placed on the spinal processes of the seventh cervical vertebra (C7) and the fourth thoracic vertebra (T4), midsternum, bilateral clavicular heads and acromions, the anterior aspect of the upper arm midway between the acromion and the lateral epicondyle, lateral epicondyle, styloid processes of the ulna and the
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
847
Arm and Trunk Movement Kinematics in People With Stroke radius, thumb nail, and index nail on the affected side. The 3-dimensional marker positions were measured at 120 Hz with a 7-camera motion capture system (VICON MX; Oxford Metrics Inc, Oxford, United Kingdom), and the data were subjected to low-pass filtering at 5 Hz with a second-order Butterworth filter. LabVIEW software (National Instruments Inc, Austin, Texas) was used to process the kinematic data. Data Analysis Eight kinematic variables were selected to indicate the endpoint coordinate and joint angle coordinate strategies and the trunk compensatory movement for the reaching task.23,43– 45 The endpoint coordinate strategy was described by index movement time, index displacement, and the percentage of index movement time at which index peak velocity (Index PPV) occurs. These variables were computed from the marker placed on the index nail. Index movement time was defined as the time between the onset and the offset of the index movement. The index movement onset was defined for each trial as the time at which the tangential velocity rose above baseline by 5% of the peak tangential velocity of the index marker, and the index movement offset was defined as the time at which the tangential velocity fell and remained below 5% of the peak tangential velocity.27,43 The index displacement was computed to represent the length of the trajectory of the marker placed on the index nail in the 3-dimensional space from index movement onset to index movement offset during the reaching task. The Index PPV is used to characterize movement strategies.46 The shoulder joint angle coordinate strategy was represented by the maximal angle of shoulder flexion in the sagittal plane and the maximal shoulder abduction in the frontal plane. 848
f
Physical Therapy
The elbow joint angle coordinate strategy was represented by the minimal angle of elbow flexion. The shoulder angle was determined as the angle between the vector joining the ipsilateral acromion-lateral epicondyle markers and the vector joining the C7–T4 markers. The elbow angle was determined as 180 degrees minus the angle between the vector formed by the ipsilateral acromionlateral epicondyle markers and a vector defined by the lateral epicondyle and the styloid process of the ulna. Trunk compensatory movement was described by trunk movement time, and trunk displacement was defined by the marker placed on the sternum. The trunk movement onset was defined for each trial as the time at which the tangential velocity rose above baseline by 5% of the peak tangential velocity of the sternum marker, and the trunk movement offset was defined as the time at which the tangential velocity fell and remained below 5% of the peak tangential velocity.27,43 Trunk movement time was defined as the time between the onset and the offset of the trunk movement. Trunk displacement was defined as the length of the trajectory of the marker placed on the sternum in the 3-dimensional space from trunk movement onset to trunk movement offset during the reaching task. The normality of the mean WMFT scores and the 8 kinematic predictors for the 2 distance tasks before and after treatment was statistically verified by examining the skewness (⫾1) and visually verified with histograms. All data sets met the criteria of normality. The collinearity between predictors was examined by analyzing variance inflation factors (ⱕ10), tolerance (ⱖ0.1), and condition index (⬎20). The datasets showed no collinearity problems.
Two steps were used to identify kinematic variables. First, the Pearson correlation coefficient (r) was used to examine the associations between kinematic variables and the mean WMFT-TIME and WMFT-FAS. The selection criterion for the entry of kinematic variables into the model was set to a P value of .25 to avoid the exclusion of important factors in model development.47 Although a typical threshold is a P value of .15, a P value of .25 is acceptable when no serious multicollinearity problem is found. Next, the remaining variables were used in a stepwise procedure to develop linear regression models for establishing concurrent and predictive validity. The concurrent validity of kinematic variables before and after treatment was estimated against the WMFT-TIME and WMFT-FAS (dependent variables) before and after treatment, respectively. The predictive validity of kinematic variables (predictors) before treatment was estimated against WMFT scores after treatment. The WMFT-TIME and WMFT-FAS before treatment were used as covariates to statistically control for individual differences that existed before treatment. Stepwise linear regressions were used to assess how much variance in the WMFT could be explained by kinematic variables and to determine which kinematic variables explained the largest amount of variance. The probability for entry into stepwise regression was set at .05. The adjusted R2 value, standardized coefficient (), and incremental R2 provided better estimations of the contribution of each predictor to the model. Role of the Funding Source This work was partly supported by the National Health Research Institutes (NHRI-EX-102-9920PI and NHRI-EX-102-10010PI), the National
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke Science Council (NSC-100-2314-B002-008-MY3 and NSC-99-2314-B182-014-MY3), and Chang Gung Memorial Hospital (CMRPD1C0401) in Taiwan. The funding sources supported the conduct of the research team, including study participant transportation expense, intervention expense, examiners’ salary, equipment maintenance and update, and data reduction. The funding sources had no influence on analysis, interpretation, or manuscript writing.
Table 1. Participant Characteristics and Clinical Measures of Motor Function (N⫽97) Variablea
Valueb
Age (y)
55.9⫾10.9
Mo after stroke onset
18.5⫾14.3
Sex Female
30 (31)
Male
67 (69)
Side of hemiparesis Right
43 (44)
Left
54 (56)
Lesion location
Results Participant characteristics and the WMFT-TIME and WMFT-FAS results before and after treatment are shown in Table 1. Kinematic data for reaching distance within and beyond arm’s length before and after treatment are shown in Table 2. Correlation Analysis Correlation analysis showed that there were 6 correlation coefficients with P values of greater than .25 and that 5 of the 6 were for the variable minimal angle of elbow flexion (Tab. 3). The kinematic variables with P values of less than .25 in the Pearson correlation coefficient analysis were entered into the stepwise multiple regression models. Multiple Regression Modeling Table 4 shows the results of stepwise multiple regression analysis for concurrent validity before treatment for the 2 reaching distances. For reaching distance within arm’s length, index movement time before treatment was a significant predictor for the WMFT-FAS model, accounting for 35.5% of the variance. Index movement time, index displacement, and maximal shoulder abduction before treatment were significant predictors for the WMFT-TIME model, accounting for 30.0% of the variance. For reaching distance beyond arm’s length, index movement time and minimal angle of elbow flexion before treatment were June 2014
Cortical
35 (36.1)
Subcortical
62 (63.9)
Type of stroke Ischemic
38 (39.2)
Hemorrhagic
59 (60.8)
Brunnstro¨m stage of upper extremity Proximal part
4.5 (3–6)
Distal part
4.5 (3–6)
WMFT-FAS before treatment
3.8⫾1.4
WMFT-TIME before treatment
4.8⫾1.8
WMFT-FAS after treatment
3.8⫾0.6
WMFT-TIME after treatment
3.6⫾0.7
a
WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time. b Continuous variables are shown as mean⫾standard deviation or median (range), and categoric variables are shown as number (percentage).
significant predictors for the WMFTFAS model, accounting for 33.7% of the variance. Index movement time and maximal angle of shoulder flexion before treatment were significant predictors for the WMFT-TIME model, accounting for 24.8% of the variance. Table 5 shows the results of stepwise multiple regression analysis for concurrent validity after treatment for the 2 reaching distances. For reaching distance within arm’s length, trunk displacement and Index PPV after treatment were significant predictors for the WMFT-FAS model, accounting for 40.5% of the variance. Trunk displacement, Index PPV, and trunk movement time after treatment were significant predictors for
the WMFT-TIME model, accounting for 47.7% of the variance. For reaching distance beyond arm’s length, index movement time and trunk displacement were significant predictors for the WMFT-FAS model, accounting for 37.4% of the variance. Trunk movement time, trunk displacement, and Index PPV were significant predictors for the WMFTTIME model, accounting for 36.4% of the variance. Table 6 summarizes the results of stepwise multiple regression analysis for predictive validity for the 2 reaching distances. After the WMFT-FAS before treatment was entered as the covariate, trunk movement time before treatment was found to be a significant predictor, explaining
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
849
Arm and Trunk Movement Kinematics in People With Stroke Table 2. Mean Values for Kinematic Variables XⴞSD Variable
a
Before Treatment
After Treatment
Within arm’s length reaching distance Index movement time (s)
1.4⫾0.7
1.1⫾0.5
Index displacement (mm)
374.7⫾131.2
353.8⫾101.3
Index PPV (%)
28.2⫾13
Maximal angle of shoulder flexion (°)
28.4⫾18.9
28.2⫾12.3 29⫾17.6
Maximal angle of shoulder abduction (°)
41.7⫾15.7
39.5⫾14.7
Minimal angle of elbow flexion (°)
93.2⫾15.4
94.5⫾14.1
Trunk movement time (s)
1.5⫾0.7
1.2⫾0.4
Trunk displacement (mm)
78.5⫾37.7
80⫾39.3
Beyond arm’s length reaching distance Index movement time (s)
1.7⫾0.7
1.5⫾0.6
Index displacement (mm)
701.6⫾129.8
685.4⫾132.8
Index PPV (%)
a
24⫾10.2
25.1⫾9
Maximal angle of shoulder flexion (°)
69.1⫾21.1
68.4⫾19.3
Maximal angle of shoulder abduction (°)
73.6⫾17.2
71.7⫾17.5
Minimal angle of elbow flexion (°)
71.8⫾17.5
72.2⫾15.3
Trunk movement time (s)
1.7⫾0.6
1.6⫾0.6
Trunk displacement (mm)
279.9⫾64.7
266.2⫾56.9
Index PPV⫽percentage of index movement time at which index peak velocity occurs.
6.9% of the variance (P⫽.001) in the WMFT-FAS for reaching distance within arm’s length; trunk displacement and trunk movement time before treatment were found to be significant predictors, explaining 9.4% of the variance (P⫽.013) for reaching distance beyond arm’s length. After the WMFT-TIME before treatment was entered as the covariate, trunk movement time before treatment was found to be a significant predictor, explaining 14.4% of the variance (P⬍.001) in the WMFTTIME for reaching distance within arm’s length; trunk displacement and trunk movement time before treatment were found to be significant predictors, explaining 14.9% of the variance (P⫽.004) for reaching distance beyond arm’s length.
Discussion To our knowledge, the present study is the first to investigate the concur850
f
Physical Therapy
rent validity of arm-trunk kinematics against motor function (WMFT) before and after 3 to 4 weeks of intensive treatment and the validity of kinematic variables for predicting motor function for reaching within and beyond arm’s length. Our results showed that the kinematic variables that can explain a proportion of the variance in the WMFT-FAS and the WMFT-TIME depend on the treatment phase (before or after treatment) and the reaching distance. For concurrent validity of the 8 kinematic variables that were examined, index movement time was most closely associated with the WMFTFAS and WMFT-TIME before treatment, whereas combinations of trunk compensatory movement variables (trunk displacement, trunk movement time, or both) and endpoint coordinate strategy variables (Index PPV and index movement time) were the significant contribu-
tors after treatment. For predictive validity, less than 15% of the variance in the WMFT was explained by trunk movement time combined with trunk displacement. In general, motor function could be better explained by the kinematic variables when the participants performed the reaching task within arm’s length rather than beyond arm’s length. These findings corroborated the results of previous studies showing that kinematic measures can reflect relatively closely a patient’s actual movement performance during functional activities.14,22,23 The findings of the present study also extend the results of previous studies14,22,23 by concurrently using endpoint coordinate strategies, joint angle coordinate strategies, and trunk compensatory movement variables to establish concurrent validity before and after treatment and to establish predictive validity. Concurrent Validity Before Treatment Index movement time was the primary contributor to the largest amount of variance (23%–36%) in the WMFT for both reaching distances. The positive relationship between index movement time and the WMFT-FAS indicated that a longer time to reach for pressing a desk bell reflected a better movement quality in performing a movement or functional task (eg, extending elbow movement, stacking checkers). One possible account for this relationship is the speed–accuracy trade-off.48 Participants spent more time to reach and press the desk bell to more accurately and successfully achieve the task goal. The accuracy and success of accomplishing tasks are associated with the movement quality measured with the WMFT. Another possible explanation is that participants spent more time to adjust or reduce pathologic synergies for performing a reaching task. A reduction in pathologic synergies
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke Table 3. Correlations (P Values) Between Kinematic Variables and Mean Wolf Motor Function Test Scores Before and After Treatmenta Correlation (P) Reaching Distance Within arm’s length
Kinematics Before Treatment
Before Treatment WMFT-FAS
WMFT-TIME b
Kinematics After Treatment
After Treatment WMFT-FAS b
⫺.542 (⬍.001)
WMFT-TIME b
⫺.525 (⬍.001)
b
Index movement time
.602 (⬍.001)
.508 (⬍.001)
Index displacement
.415 (⬍.001)b
.478 (⬍.001)b ⫺.468 (⬍.001)b ⫺.454 (⬍.001)b
Index PPV
⫺.334 (⬍.001)b ⫺.273 (.007)b b
.276 (.006)
b
Maximal angle of shoulder flexion
.297 (.03)
Maximal angle of shoulder abduction
.219 (.032)b
.208 (.041)b
Minimal angle of elbow flexion
⫺.273 (.007)b
⫺.228 (.025)b
.35 (⬍.001)b ⫺.193 (.058)
b
⫺.179 (.079)b
.095 (.353)
.292 (.004)b ⫺.192 (.06)
.133 (.195)b
MinElbFlex
.107 (.297)
.054 (.601)
.337 (.001)b
.261 (.01)b
Maximal angle of shoulder flexion
.181 (.076)
Maximal angle of shoulder abduction
.117 (.255)
Minimal angle of elbow flexion
⫺.233 (.022)b
.227 (.025)
⫺.31 (.002)b
⫺.372 (⬍.001)b ⫺.323 (.001)b
.188 (.065)b
⫺.221 (.03)b
⫺.161 (.116)
.47 (⬍.001)b
⫺.301 (.003)b
.477 (⬍.001)b ⫺.501 (⬍.001)b ⫺.456 (⬍.001)b
.372 (⬍.001)b
.422 (⬍.001)b
⫺.339 (.001)b
.564 (⬍.001)b
b
⫺.454 (⬍.001)b ⫺.406 (⬍.001)b
MaxShAbd
.146 (.155)b
b
Index Displ Index PPV
⫺.454 (⬍.001)b
⫺.161 (.114)b
.255 (.012)b
⫺.381 (⬍.001)b ⫺.29 (.004)b
⫺.474 (⬍.001)
⫺.212 (.037)b
Trunk displacement
Index PPV
Index MT
⫺.223 (.028)
.409 (⬍.001)b ⫺.525 (⬍.001)b ⫺.569 (⬍.001)b
Index displacement
WMFT-TIME b
MaxShFlex
.526 (⬍.001)b
Beyond arm’s Index length movement time
WMFT-FAS
b
b
Trunk movement time
⫺.343 (.001)b
Correlation (P) After Treatment
.321 (.001)b
⫺.435 (⬍.001)b ⫺.446 (⬍.001)b
Trunk Displ
⫺.548 (⬍.001)b ⫺.558 (⬍.001)b
Index MT
⫺.541 (⬍.001)b ⫺.505 (⬍.001)b
Index Displ
⫺.402 (⬍.001)b ⫺.316 (.002)b
Index PPV
.32 (.001)b
.359 (⬍.001)b
b
MaxShFlex
⫺.245 (.016)
⫺.188 (.065)b
⫺.183 (.073)b
MaxShAbd
⫺.242 (.017)b
⫺.237 (.019)b
.121 (.237)b
.119 (.246)b
MinElbFlex
.07 (.498)
.08 (.437)
b
⫺.159 (.119)
Trunk MT
Trunk movement time
.56 (⬍.001)b
.388 (⬍.001)b ⫺.505 (⬍.001)b ⫺.464 (⬍.001)b
Trunk displacement
.26 (.01)b
.229 (.024)b
⫺.374 (⬍.001)b ⫺.427 (⬍.001)b
b
⫺.21 (.039)b
Trunk MT
⫺.538 (⬍.001)b ⫺.521 (⬍.001)b
Trunk Displ
⫺.459 (⬍.001)b ⫺.436 (⬍.001)b
a WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time, Index PPV⫽percentage of index movement time at which index peak velocity occurs. b Kinematic variables entered into multiple regression models for further analysis.
may be reflected by better movement quality. Concurrent Validity After Treatment Of the 8 kinematic variables that were measured, trunk compensatory June 2014
movement variables were most closely associated with the WMFTFAS and WMFT-TIME (partial R2⫽ .30 –.35). The Index PPV (partial R2⫽.04 –.15) was the second most important contributor to the variance in the WMFT-TIME and WMFT-
FAS for both reaching distances after treatment, except for the regression model of WMFT-FAS for reaching distance beyond arm’s length. Previous studies found similar results; they showed that trunk displacement during reaching, reach-to-grasp,23
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
851
Arm and Trunk Movement Kinematics in People With Stroke Table 4. Results of Stepwise Multiple Regression Analyses and Collinearity Test for Concurrent Validity Before Treatment
Regression Modela
 Coefficient (Standardized)
Incremental R2
Adjusted R2 (Model P)
Tolerance
Variance Inflation Factor
Condition Index
Within arm’s length reaching distance Model 1 dependent variable: WMFT-FAS before treatment Index movement time
.355 (⬍.001) .602
.362
Model 2 dependent variable: WMFT-TIME before treatment
4.282 1
1
.3 (.048)
11.121
Index movement time
.285
.259
.521
1.92
Index displacement
.294
.293
.523
1.88
Maximal angle of shoulder abduction
.176
.322
.943
1.06
Beyond arm’s length reaching distance Model 3 dependent variable: WMFT-FAS before treatment Index movement time Minimal angle of elbow flexion
.337 (.032) .547
.318
.991
1.009
⫺.182
.351
.991
1.009
Model 4 dependent variable: WMFT-TIME before treatment
a
10.998
.248 (.034)
8.621
Index movement time
.462
.227
.994
1.006
Maximal angle of shoulder flexion
.191
.264
.994
1.006
WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time.
and drinking14 tasks could reflect not only motor impairment but also motor function in people with stroke. Although trunk movement time has not been used often to represent trunk compensation, the results of the present study showed that trunk movement time was a significant factor and that up to 31% of the variance in the WMFT was attributable to trunk movement time. A previous study49 also suggested that a shorter trunk movement time may reflect less trunk recruitment, indicating that trunk movement time could be considered an important indicator of trunk compensation. In addition, the combination of trunk displacement and trunk movement time seems to relate to motor function better than either one alone. The kinematic variables reflecting motor function differed before and 852
f
Physical Therapy
after intensive treatments. Possible reasons for trunk compensatory movement variables being important indicators only after treatment are that people with stroke may use trunk compensatory movement as a habitual response learned after stroke31 and that reaching movements are primarily controlled by endpoint coordinate strategy variables. After intensive treatments, people may have a greater capacity to voluntarily, separately control trunk and arm participation in UE movements.27 Because the recruitment of the trunk became an important contributor for performing the reaching task after rehabilitation, it may serve as a variable reflecting motor function. These data suggest the need for careful selection of kinematic variables to describe the motor function of people with stroke at various phases, such as before treatment and after treatment. In addition, these previously unreported
data indicate the trunk movements that may partially reflect motor function after treatment. The joint angle coordinate strategy variables for the UE did not play important roles in explaining the variance in motor function. A possible reason is that the central nervous system preprograms UE movement through endpoint trajectory, together with arm and trunk coordination, rather than joint angles.34 The findings of the present study were similar to those of Subramanian et al23 but in contrast to those of Massie et al,22 who found that elbow extension during reaching was the only predictor to reflect UE motor function. The discrepancy between the findings of the present study and those of the study of Massie et al22 may be attributable to the nature of the kinematic task. In the study of Massie et al,22 participants were asked to reach between 2 targets
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke Table 5. Results of Stepwise Multiple Regression Analyses and Collinearity Test for Concurrent Validity After Treatment  Coefficient (Standardized)
Regression Modela
Incremental R2
Adjusted R2 (Model P)
Tolerance
Variance Inflation Factor
Condition Index
Within arm’s length reaching distance Model 1 dependent variable: WMFT-FAS after treatment
.405 (⬍.001) ⫺.495
Trunk displacement Index PPV
.347
7.167
.3
.997
1.024
.417
.997
1.024
Model 2 dependent variable: WMFT-TIME after treatment
.477 (.02)
Trunk displacement Index PPV Trunk movement time
11.986
⫺.452
.311
.917
1.09
.328
.462
.858
1.166
⫺.195
.493
.808
1.238
Beyond arm’s length reaching distance Model 3 dependent variable: WMFT-FAS after treatment
.374 (⬍.001)
Index movement time
⫺.442
.292
Trunk displacement
⫺.324
.387
Model 4 dependent variable: WMFT-TIME after treatment
11.511 .907
1.103
.907
1.103
.364 (.019)
14.918
Trunk movement time
⫺.358
.271
.807
1.239
Trunk displacement
⫺.284
.346
.88
1.136
.205
.384
.897
1.114
Index PPV a
WMFT-FAS⫽Wolf Motor Function Test functional ability score, Index PPV⫽percentage of index movement time at which index peak velocity occurs, WMFT-TIME⫽Wolf Motor Function Test performance time.
for 4 continuous cycles, whereas in the present study, the discrete functional task of pressing a desk bell on a table was used. Predictive Validity of Kinematic Measures The trunk compensatory movement variables before treatment (trunk movement time and trunk displacement) were significant predictors, explaining 6.9% to 9.4% of the variance in the WMFT-FAS and 14.4% to 14.9% of the variance in the WMFTTIME after treatment. A shorter trunk movement time and less trunk displacement before intensive treatment are related to better temporal efficiency for performing functional tasks (WMFT-TIME) after intensive treatment in people with stroke and mild to moderate levels of UE motor function. Although only limited variJune 2014
ance in the WMFT-TIME after treatment was explained by the trunk variables before treatment, Kaplan and Saccuzzo50 suggested that R2 values of greater than .1 with statistical significance could be considered the minimal threshold for a predictor in a predictive model. Therefore, trunk variables may be predictors and may have some prognostic value. Our results confirmed the finding of Cirstea and Levin27 that better trunk control performance before treatment in people with stroke may indicate better execution capacity and increased opportunities to obtain greater functional gains from intensive treatment. Fritz et al24 found the motor ability of finger extension to be a plausible predictor of motor function after treatment. Adding trunk variables to the previously suggested predictive models may
enhance the accuracy of prediction of motor function after treatment. That trunk variables explained less than 10% of the variance in the WMFT-FAS may have been attributable to the WMFT-FAS being less objective, possibly resulting in the high variability of the scores. Kinematic Validity for Reaching to a Target Within and Beyond Arm’s Length The kinematic variables selected in the regression models explained and predicted larger amounts of variance in the WMFT scores for the task within arm’s length (R2⫽.069 –.493) than for the task beyond arm’s length (R2⫽.094 –.387). The results indicated that a reaching task within arm’s length may be more suitable for evaluating kinematic performance to describe motor function in people
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
853
Arm and Trunk Movement Kinematics in People With Stroke Table 6. Results of Stepwise Multiple Regression Analyses and Collinearity Test for Predictive Validity
Regression Modela
 Coefficient (Standardized)
Incremental R2
Adjusted R2 (Model P)
Tolerance
Variance Inflation Factor
Condition Index
Within arm’s length reaching distance Model 1 dependent variable: WMFT-FAS after treatment
.384 (.001)
6.719
WMFT-FAS before treatment
⫺.410
.328
.723
1.383
Trunk movement time before treatment
⫺.309
.397
.723
1.383
Model 2 dependent variable: WMFT-TIME after treatment
.429 (⬍.001)
6.2
WMFT-TIME before treatment
⫺.375
.297
.833
1.2
Trunk movement time before treatment
⫺.416
.441
.833
1.2
Beyond arm’s length reaching distance Model 3 dependent variable: WMFT-FAS after treatment
.403 (.013)
WMFT-FAS before treatment
⫺.379
.328
.667
1.499
Trunk displacement before treatment
⫺.220
.382
.923
1.084
Trunk movement time before treatment
⫺.242
.422
.679
1.473
Model 4 dependent variable: WMFT-TIME after treatment
a
12.384
.428 (.004)
12.338
WMFT-TIME before treatment
⫺.383
.297
.829
1.207
Trunk displacement before treatment
⫺.283
.394
.925
1.081
Trunk movement time before treatment
⫺.251
.446
.830
1.205
WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time.
with stroke than a reaching task beyond arm’s length. Unlike people who are healthy, who use only interjoint recruitment and coordination of the UE to perform reaching tasks within arm’s length, people with stroke may use trunk compensatory movement to assist the UE to perform reaching tasks within arm’s length to a different degree.27,29,30 Limitations The present study has a few limitations. One is the generalizability of the study results. The data used for the present study were pooled from several clinical trials investigating different treatment interventions. These clinical trials recruited people with stroke and moderate to high levels of motor ability. Therefore, our results cannot be attributed to the effect of a specific treatment and can be generalized only to people 854
f
Physical Therapy
with chronic stroke and similar levels of motor impairment. Second, the Brunnstro ¨ m stage, an inclusion criterion for this secondary analysis study, may not be a desirable measurement for clinical settings. Other suitable clinical measurements, such as the Fugl-Meyer Assessment for the UE, could be used as inclusion criteria in further studies. Third, variability among the testers in the different trials may have affected the results; however, all of the testers across the clinical trials received training before they evaluated participants to ensure that the same evaluation standard was used.
Conclusion The uniqueness of the present study is that it separately examined the current validity of kinematic variables for different reaching distances before and after intensive UE train-
ing and examined the predictive validity for both distances. The results provide a better understanding of the meaning and nature of different types of kinematic variables applied in rehabilitation research. The selection of different types of kinematic variables to describe UE motor function at various phases, such as before and after treatment, in patients with stroke is necessary. Before treatment, index movement time seemed to be a kinematic variable with a relatively better association with motor function. After treatment, trunk compensatory movement variables (trunk movement time and trunk displacement) became salient contributors, possibly because participants then had a greater capacity to separately control trunk participation and arm participation in UE movements. For predictive validity, trunk variables may be considered
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke possible predictors of UE motor function after treatment. However, it would be better to use these trunk variables as adjuncts to enhance the accuracy of prediction because the amount of variance in motor function that kinematic variables could explain is small. A reaching task within arm’s length may be better than a reaching task beyond arm’s length for evaluating kinematic performance to partially describe motor function in people with stroke. Dr Wu and Dr Lin provided concept/idea/ research design, project management, and fund procurement. Dr Wu and Dr Liing provided writing. Dr Wu provided data collection and institutional liaisons. Dr Wu, Dr Liing, and Professor H. Chen provided data analysis. Professor C. Chen provided study participants. Dr Wu, Professor C. Chen, and Dr Lin provided facilities/equipment. Dr Lin provided consultation (including review of manuscript before submission). This study was approved by the institutional review board at each participating site. This work was partly supported by the National Health Research Institutes (NHRIEX-102-9920PI and NHRI-EX-102-10010PI), the National Science Council (NSC-1002314-B-002-008-MY3 and NSC-99-2314-B182-014-MY3), and Chang Gung Memorial Hospital (CMRPD1C0401) in Taiwan. DOI: 10.2522/ptj.20130101
References 1 Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study. Arch Phys Med Rehabil. 1994;75:394 –398. 2 Mayo NE, Wood-Dauphine´e S, Ahmed S, et al. Disablement following stroke. Disabil Rehabil. 1999;21:258 –268. 3 Olsen TS. Arm and leg paresis as outcome predictors in stroke rehabilitation. Stroke. 1990;21:247–251. 4 Feys HM, De Weerdt WJ, Selz BE, et al. Effect of a therapeutic intervention for the hemiplegic upper limb in the acute phase after stroke: a single-blind, randomized, controlled multicenter trial. Stroke. 1998; 29:785–792. 5 Shumway-Cook A, Woollacott MH. Normal reach, grasp, and manipulation. In: Lupash E, ed. Motor Control: Translating Research Into Clinical Practice. 4th ed. Philadelphia, PA: Wolters Kluwer, Lippincott Williams & Wilkins; 2012:477–501.
June 2014
6 Shumway-Cook A, Woollacott MH. Clinical management of the patient with reach, grasp, and manipulation disorders. In: Lupash E, ed. Motor Control: Translating Research Into Clinical Practice. 4th ed. Philadelphia, PA: Wolters Kluwer, Lippincott Williams & Wilkins; 2012:552–594. 7 Roby-Brami A, Feydy A, Combeaud M, et al. Motor compensation and recovery for reaching in stroke patients. Acta Neurol Scand. 2003;107:369 –381. 8 Reisman DS, Scholz JP. Aspects of joint coordination are preserved during pointing in persons with post-stroke hemiparesis. Brain. 2003;126:2510 –2527. 9 Caimmi M, Carda S, Giovanzana C, et al. Using kinematic analysis to evaluate constraint-induced movement therapy in chronic stroke patients. Neurorehabil Neural Repair. 2008;22:31–39. 10 Thielman GT, Dean CM, Gentile AM. Rehabilitation of reaching after stroke: taskrelated training versus progressive resistive exercise. Arch Phys Med Rehabil. 2004;85:1613–1618. 11 Massie CL, Malcolm MP, Greene D, Thaut M. The effects of constraint-induced therapy on kinematic outcomes and compensatory movement patterns: an exploratory study. Arch Phys Med Rehabil. 2009;90: 571–579. 12 Wagner JM, Dromerick AW, Sahrmann SA, Lang CE. Upper extremity muscle activation during recovery of reaching in subjects with post-stroke hemiparesis. Clin Neurophysiol. 2007;118:164 –176. 13 Kitago T, Liang J, Huang VS, et al. Improvement after constraint-induced movement therapy: recovery of normal motor control or task-specific compensation? Neurorehabil Neural Repair. 2013; 27:99 –109. 14 Alt Murphy M, Willen C, Sunnerhagen KS. Movement kinematics during a drinking task are associated with the activity capacity level after stroke. Neurorehabil Neural Repair. 2012;26:1106 –1115. 15 Wagner JM, Rhodes JA, Patten C. Reproducibility and minimal detectable change of three-dimensional kinematic analysis of reaching tasks in people with hemiparesis after stroke. Phys Ther. 2008;88:652– 663. 16 Schneiberg S, McKinley P, Gisel E, et al. Reliability of kinematic measures of functional reaching in children with cerebral palsy. Dev Med Child Neurol. 2010;52: e167– e173. 17 Wagner JM, Lang CE, Sahrmann SA, et al. Relationships between sensorimotor impairments and reaching deficits in acute hemiparesis. Neurorehabil Neural Repair. 2006;20:406 – 416. 18 Lang CE, Wagner JM, Dromerick AW, Edwards DF. Measurement of upperextremity function early after stroke: properties of the Action Research Arm Test. Arch Phys Med Rehabil. 2006;87:1605– 1610. 19 Wagner JM, Lang CE, Sahrmann SA, et al. Sensorimotor impairments and reaching performance in subjects with poststroke hemiparesis during the first few months of recovery. Phys Ther. 2007;87:751–765.
20 Zackowski KM, Dromerick AW, Sahrmann SA, et al. How do strength, sensation, spasticity and joint individuation relate to the reaching deficits of people with chronic hemiparesis? Brain. 2004;127:1035–1046. 21 Edwards DF, Lang CE, Wagner JM, et al. An evaluation of the Wolf Motor Function Test in motor trials early after stroke. Arch Phys Med Rehabil. 2012;93:660 – 668. 22 Massie CL, Fritz S, Malcolm MP. Elbow extension predicts motor impairment and performance after stroke. Rehabil Res Pract. 2011;2011:381978. 23 Subramanian SK, Yamanaka J, Chilingaryan G, Levin MF. Validity of movement pattern kinematics as measures of arm motor impairment poststroke. Stroke. 2010;41:2303–2308. 24 Fritz SL, Light KE, Patterson TS, et al. Active finger extension predicts outcomes after constraint-induced movement therapy for individuals with hemiparesis after stroke. Stroke. 2005;36:1172–1177. 25 Loewen SC, Anderson BA. Predictors of stroke outcome using objective measurement scales. Stroke. 1990;21:78 – 81. 26 Hollerbach JH. Planning arm movements. In: Osherson DN, Kosslyn SM, Hollerbach JM, eds. Visual Cognition and Action. Vol 2. Cambridge, MA: MIT; 1990:183–211. 27 Cirstea MC, Levin MF. Compensatory strategies for reaching in stroke. Brain. 2000; 123:940 –953. 28 Robertson JV, Roby-Brami A. The trunk as a part of the kinematic chain for reaching movements in healthy subjects and hemiparetic patients. Brain Res. 2011;1382: 137–146. 29 Levin MF, Michaelsen SM, Cirstea CM, Roby-Brami A. Use of the trunk for reaching targets placed within and beyond the reach in adult hemiparesis. Exp Brain Res. 2002;143:171–180. 30 Archambault P, Pigeon P, Feldman AG, Levin MF. Recruitment and sequencing of different degrees of freedom during pointing movements involving the trunk in healthy and hemiparetic subjects. Exp Brain Res. 1999;126:55– 67. 31 Woodbury ML, Howland DR, McGuirk TE, et al. Effects of trunk restraint combined with intensive task practice on poststroke upper extremity reach and function: a pilot study. Neurorehabil Neural Repair. 2009;23:78 –91. 32 van Vliet PM, Sheridan MR. Ability to adjust reach extent in the hemiplegic arm. Physiotherapy. 2009;95:176 –184. 33 Kaminski TR, Bock C, Gentile AM. The coordination between trunk and arm motion during pointing movements. Exp Brain Res. 1995;106:457– 466. 34 Ma S, Feldman AG. Two functionally different synergies during arm reaching movements involving the trunk. J Neurophysiol. 1995;73:2120 –2122. 35 McCrea PH, Eng JJ, Hodgson AJ. Biomechanics of reaching: clinical implications for individuals with acquired brain injury. Disabil Rehabil. 2002;24:534 –541.
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
855
Arm and Trunk Movement Kinematics in People With Stroke 36 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. 37 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. 38 Wu CY, Yang CL, Chuang LL, et al. Effect of therapist-based versus robot-assisted bilateral arm training on motor control, functional performance, and quality of life after chronic stroke: a clinical trial. Phys Ther. 2012;92:1006 –1016. 39 Wu CY, Chuang LL, Lin KC, et al. Randomized trial of distributed constraint-induced therapy versus bilateral arm training for the rehabilitation of upper-limb motor control and function after stroke. Neurorehabil Neural Repair. 2011;25:130 –139. 40 Lin KC, Chen YA, Chen CL, et al. The effects of bilateral arm training on motor control and functional performance in chronic stroke: a randomized controlled study. Neurorehabil Neural Repair. 2010; 24:42–51.
856
f
Physical Therapy
41 Wu CY, Chen CL, Tang SF, et al. Kinematic and clinical analyses of upper-extremity movements after constraint-induced movement therapy in patients with stroke: a randomized controlled trial. Arch Phys Med Rehabil. 2007;88:964 –970. 42 DeJong SL, Schaefer SY, Lang CE. Need for speed: better movement quality during faster task performance after stroke. Neurorehabil Neural Repair. 2012;26:362–373. 43 Levin MF, Kleim JA, Wolf SL. What do motor “recovery” and “compensation” mean in patients following stroke? Neurorehabil Neural Repair. 2009;23:313–319. 44 Alt Murphy M, Willen C, Sunnerhagen KS. Kinematic variables quantifying upperextremity performance after stroke during reaching and drinking from a glass. Neurorehabil Neural Repair. 2011;25:71– 80. 45 Subramanian SK, Lourenco CB, Chilingaryan G, et al. Arm motor recovery using a virtual reality intervention in chronic stroke: randomized control trial. Neurorehabil Neural Repair. 2013;27:13–23.
46 Skidmore ER, Becker JT, Whyte EM, et al. Cognitive impairments and depressive symptoms did not impede upper limb recovery in a clinical repetitive task practice program after stroke: a pilot study. Am J Phys Med Rehabil. 2012;91:327–331. 47 Hosmer DW, Lemeshow S. Applied Logistic Regression. New York, NY: John Wiley & Sons Inc; 1989. 48 Fitts PM. The information capacity of the human motor system in controlling the amplitude of movement. J Exp Psychol. 1954;47:381–391. 49 Ma HI, Hwang WJ, Wang CY, et al. Trunkarm coordination in reaching for moving targets in people with Parkinson’s disease: comparison between virtual and physical reality. Hum Mov Sci. 2012;31:1340 –1352. 50 Kaplan KM, Saccuzzo DP. Psychological Testing: Principles, Application, and Issues. 5th ed. Belmont, CA: Wadsworth/ Thomson Learning; 2001.
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics During Seated Reaching Within and Beyond Arm's Length in People With Stroke: A Validity Study Ching-yi Wu, Rong-jiuan Liing, Hsieh-ching Chen, Chia-ling Chen and Keh-chung Lin PHYS THER. 2014; 94:845-856. Originally published online January 30, 2014 doi: 10.2522/ptj.20130101
References
This article cites 46 articles, 24 of which you can access for free at: http://ptjournal.apta.org/content/94/6/845#BIBL
Subscription Information
http://ptjournal.apta.org/subscriptions/
Permissions and Reprints http://ptjournal.apta.org/site/misc/terms.xhtml Information for Authors
http://ptjournal.apta.org/site/misc/ifora.xhtml
Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
The online version of this article, along with updated information and services, can be found online at: http://ptjournal.apta.org/content/94/6/845 Collections
This article, along with others on similar topics, appears in the following collection(s): Kinesiology/Biomechanics Stroke (Geriatrics) Stroke (Neurology) Tests and Measurements
e-Letters
To submit an e-Letter on this article, click here or click on "Submit a response" in the right-hand menu under "Responses" in the online version of this article.
E-mail alerts
Sign up here to receive free e-mail alerts
Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Research Report Arm and Trunk Movement Kinematics During Seated Reaching Within and Beyond Arm’s Length in People With Stroke: A Validity Study Ching-yi Wu, Rong-jiuan Liing, Hsieh-ching Chen, Chia-ling Chen, Keh-chung Lin
Background. Kinematic analysis is commonly used to objectively measure upper extremity movement performance after stroke. However, the concurrent validity and predictive validity of arm-trunk kinematics during reaching within and beyond arm’s length have not been studied. Objective. The aim of this study was to estimate the concurrent validity of kinematic measures before and after treatment and the predictive validity for reaching within and beyond arm’s length after stroke.
Design. This was a secondary analysis study. Methods. Ninety-seven participants with stroke (mean age⫽55.9 years [SD⫽10.9]) received intensive treatment every weekday for 3 to 4 weeks. Kinematic reaching tasks and the Wolf Motor Function Test (WMFT) were used before and after treatment. The validity of the kinematic measures was estimated in relation to WMFT scores.
Results. Of the 8 kinematic variables that were measured, index movement time
before treatment (R2⫽.227–.362) and trunk movement time and trunk displacement after treatment (R2⫽.095–.346) had the strongest association with the WMFT at both reaching distances. Trunk movement time and trunk displacement before treatment explained 6.9% to 14.9% of the variance in the WMFT after treatment. Kinematic variables explained 6.9% to 49.3% and 9.4% to 38.7% of the variance in the WMFT during a task within arm’s length and beyond arm’s length, respectively.
Limitations. The study has limited generalizability. Conclusions. Different kinematic variables may partially reflect motor function before and after treatment to a limited degree. Although the predictive validity was modest, trunk movement may be considered a prognostic determinant of motor function after treatment. A reaching task within arm’s length may be a more suitable measure of kinematic performance for describing motor function than a reaching task beyond arm’s length.
C. Wu, ScD, OTR/L, Department of Occupational Therapy and Graduate Institute of Behavioral Sciences, College of Medicine, Chang Gung University, Taoyuan, Taiwan, and Healthy Ageing Research Center, Chang Gung University, Taoyuan, Taiwan. R. Liing, PT, PhD, School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan. H. Chen, PhD, Department and Graduate Institute of Industrial Engineering and Management, National Taipei University of Technology, Taipei, Taiwan. C. Chen, PhD, MD, Department of Industrial Engineering, and Rehabilitation, LinKou Chang Gung Memorial Hospital, Taoyuan, Taiwan. K. Lin, ScD, OTR/L, School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan, and Division of Occupational Therapy, Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, 17, F4, Xuzhou Road, Taipei 100, Taiwan. Address all correspondence to Dr Lin at: [email protected]. [Wu C, Liing R, Chen H, et al. Arm and trunk movement kinematics during seated reaching within and beyond arm’s length in people with stroke: a validity study. Phys Ther. 2014;94:845– 856.] © 2014 American Physical Therapy Association Published Ahead of Print: January 30, 2014 Accepted: January 27, 2014 Submitted: March 10, 2013
Post a Rapid Response to this article at: ptjournal.apta.org June 2014
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
845
Arm and Trunk Movement Kinematics in People With Stroke
I
mpaired upper extremity (UE) function is a common deficit after stroke. About 70% of patients with stroke experience hemiparesis with an impaired UE,1 and 55% to 75% will still have limitations in UE function 3 to 6 months after stroke.2,3 Upper extremity motor impairment often leads to poor movement control and long-term disabilities.4 Because of persistent UE dysfunction after stroke, an accurate and precise measurement of UE performance is crucial for rehabilitation. Kinematic analysis has been commonly used in objective and quantitative measures of UE movement performance after stroke. Kinematic performance may reveal aspects of movement planning in terms of the joint angle coordinate strategy and endpoint coordinate strategy.5 In the joint angle coordinate strategy, the central nervous system plans the movement of the UE around a set of intrinsic coordinates of the body (eg, shoulder, elbow, and wrist joints) to arrive at the target, whereas in the endpoint coordinate strategy, extrinsic coordinates in the space are used to direct the hand to the final (endpoint) location (ie, the target). In addition, kinematic variables are often used to detect trunk compensatory movement (trunk displacement and trunk movement time).5,6 Kinematic assessment has been widely applied for discriminating differences in movement strategies between people with stroke and adults who are healthy7,8 and for evaluating the effects of various therapeutic interventions on the UE9 –11 or UE recovery after stroke.12,13 Incorporating kinematic assessment into clinical evaluation is necessary to understand the impact of a disability or to capture or predict important effects of treatments on various health-related conditions, ranging from motor control strategies to activity participation.14
846
f
Physical Therapy
The ability of kinematic data to function as a discriminative, monitoring, or evaluative measurement tool depends on the presence of sound psychometric properties.15 Some studies have established good relative15,16 and absolute15 (eg, minimal detectable change value) reliability of kinematic measures. Although several studies using kinematic variables have shown a relationship between sensorimotor impairment and motor dysfunction,17–21 research investigating the concurrent validity of kinematics has been limited,14,22,23 and no research using comprehensive kinematic variables to correlate motor impairment and motor function in people with stroke has been done. Information about how different types of kinematic variables at various stroke phases (eg, before and after intensive intervention) relate to motor function is of great value for better selection and interpretation of kinematic variables reflecting motor function in rehabilitation research. Furthermore, Massie et al22 pointed out the need for studying how kinematic variables predict movement performance over time (ie, the predictive validity of kinematics). Adding plausible kinematic predictors to established predictive models19,24 may improve the determination of prognosis.25 In 2 previous studies of kinematic validity, variables relevant to joint angle coordinate strategy (eg, recruitment of shoulder flexion, shoulder abduction, and elbow extension) were used to investigate the concurrent validity of kinematics reflecting motor impairment23 and motor function.22 In another kinematic validity study, selected endpoint coordinate strategy variables (eg, endpoint movement trajectory, movement time, and time to peak velocity) were used.14 Given that the central nervous system may not program movements exclusively through one strategy or the other26
and that task demands (eg, target distance within or beyond arm’s length) and health conditions (eg, adults who are healthy or people with stroke) may be factors,26 both types of strategy variables must be included in studies of kinematic validity. Trunk compensatory movement during reaching has been considered a critical factor in studies of the UE performance of people with stroke27–30 because they often recruit additional degrees of freedom (trunk movement) to compensate for UE motor deficits, thereby hindering the potential for motor recovery after stroke.27 People with stroke may use excessive trunk movement to compensate for a limited range of motion in the UE29; specific rehabilitation training may improve voluntary joint range and coordination and decrease trunk movement.10,31 Previous studies of kinematic validity or the relationships between motor impairment or function and kinematic measures also suggested that trunk displacement is an important variable14,23 for distinguishing the severity of motor impairment.23 It also may be important to incorporate trunk compensatory movement variables in further studies of the concurrent validity of kinematic measures before and after intensive training. In all validity studies, UE tasks with a target distance within approximately 80% to 100% of arm’s length have been used.14,22,23 Different reaching distances induce people with stroke to use different coordinate strategies, possibly together with trunk compensatory strategies, to accomplish reaching tasks.29,32 Reaching to a target within arm’s length involves the endpoint control and interjoint coordination of the UE; that is, central commands preplan a straight, efficient, and smooth endpoint trajectory and then coordinate
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke the motions of the shoulder and elbow to reach the target.33,34 Reaching beyond arm’s length (see “Method” section) requires trunk participation, and the amount of trunk displacement or recruitment is inversely associated with joint recruitment of the shoulder and elbow.28,31,35 Whether the kinematic performance of tasks within arm’s length can fully represent a patient’s coordinate strategies and trunk recruitment and which kinematic task (within or beyond arm’s length) appropriately reflects a patient’s motor performance have yet to be determined. The purpose of this study was to investigate the validity of arm and trunk movement kinematics during seated forward reaching. We attempted to estimate the concurrent validity of using arm-trunk kinematics against Wolf Motor Function Test (WMFT) scores before and after treatment and the validity of these kinematic measures for predicting WMFT scores for reaching tasks within and beyond arm’s length after stroke. The WMFT, with wellknown psychometric properties,36,37 was used as the gold standard for UE motor function measurement.
Method Participants In the present study, we used pooled data for a secondary analysis. The data were drawn from 4 previous38 – 41 and ongoing randomized controlled trials of constraint-induced therapy, bilateral arm training, robot-assisted arm training, and conventional training. Ninety-seven participants at 4 hospitals in Taiwan were recruited. The inclusion criteria used in the randomized controlled trials were as follows: first stroke, Brunnstro ¨ m stage 3 or above for the proximal and distal UE, and ability to follow instructions for the evaluation of arm-trunk kinematics. The exclusion criteria were as follows: cognitive impairment June 2014
(Mini-Mental State Examination score of ⬍21), excessive spasticity at any joint of the arm (modified Ashworth scale score of ⬎2), and participation in experimental rehabilitation or drug studies within the preceding 6 months. All participants signed a consent form. Only data from participants who were evaluated with the WMFT and a kinematic assessment before and after intervention with tasks both within arm’s length and beyond arm’s length were included in the present study. Interventions and Procedure The randomization procedures used in the trials were similar. Participants were randomly assigned to the constraint-induced therapy, bilateral arm training, robot-assisted arm training, or conventional training treatment group for 90 to 120 minutes of intervention every weekday for 3 to 4 weeks. Six evaluators with training in administering clinical assessments and unaware of the group assignments performed the clinical evaluations. Clinical Assessments The WMFT, which includes 15 function-based tasks, was used to evaluate UE function before and after 3 to 4 weeks of treatment. Tasks 1 to 6 of the WMFT are timed joint segment movements, and tasks 7 to 15 are timed integrative functional movements. The WMFT has 2 main scores: the mean performance time (WMFT-TIME) and the mean functional ability score (WMFT-FAS). The reliability and validity of the WMFTTIME and the WMFT-FAS have been well established.36,37 In the WMFTTIME, the instructions for each task focus on the speed of completion. In the WMFT-FAS, task completion and movement strategies (eg, movements made in synergy) are rated on a 6-point ordinal scale (from 0 to 5 points).
Measurement of Reaching Performance A reaching task is a fundamental component of many daily activities, requires the coordinated movement of multiple UE joints, and is thus appropriate for use as an experimental task.19 Reaching to press a desk bell was used as the kinematic task. The bell was placed along the participant’s midsagittal plane at a distance corresponding to the distance from the medial border of the axilla to the distal wrist crease (a reaching distance within arm’s length) and at 125% of arm’s length (a reaching distance beyond arm’s length). Arm’s length was defined as the distance from the medial border of the axilla to the tip of the third finger, as measured with the arm hanging down alongside the trunk. The instruction was to use the index finger of the affected arm to press the desk bell as quickly as possible.42 Each participant sat on an adjustable chair with the seat height set to 100% of the lower leg’s length, as measured from the lateral knee joint to the foot. An adjustable table was adjacent to the chair. The table height was adjusted to 5 cm below the elbow while the participant sat on the chair. The initial position of the hand was on the edge of the table with the elbow flexed at 90 degrees. After one practice trial, the participant was instructed to reach to press the desk bell as quickly as possible in 3 trials. Kinematic Analysis of Reaching Performance Arm and trunk movements were modeled with 13 markers, which were placed on the spinal processes of the seventh cervical vertebra (C7) and the fourth thoracic vertebra (T4), midsternum, bilateral clavicular heads and acromions, the anterior aspect of the upper arm midway between the acromion and the lateral epicondyle, lateral epicondyle, styloid processes of the ulna and the
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
847
Arm and Trunk Movement Kinematics in People With Stroke radius, thumb nail, and index nail on the affected side. The 3-dimensional marker positions were measured at 120 Hz with a 7-camera motion capture system (VICON MX; Oxford Metrics Inc, Oxford, United Kingdom), and the data were subjected to low-pass filtering at 5 Hz with a second-order Butterworth filter. LabVIEW software (National Instruments Inc, Austin, Texas) was used to process the kinematic data. Data Analysis Eight kinematic variables were selected to indicate the endpoint coordinate and joint angle coordinate strategies and the trunk compensatory movement for the reaching task.23,43– 45 The endpoint coordinate strategy was described by index movement time, index displacement, and the percentage of index movement time at which index peak velocity (Index PPV) occurs. These variables were computed from the marker placed on the index nail. Index movement time was defined as the time between the onset and the offset of the index movement. The index movement onset was defined for each trial as the time at which the tangential velocity rose above baseline by 5% of the peak tangential velocity of the index marker, and the index movement offset was defined as the time at which the tangential velocity fell and remained below 5% of the peak tangential velocity.27,43 The index displacement was computed to represent the length of the trajectory of the marker placed on the index nail in the 3-dimensional space from index movement onset to index movement offset during the reaching task. The Index PPV is used to characterize movement strategies.46 The shoulder joint angle coordinate strategy was represented by the maximal angle of shoulder flexion in the sagittal plane and the maximal shoulder abduction in the frontal plane. 848
f
Physical Therapy
The elbow joint angle coordinate strategy was represented by the minimal angle of elbow flexion. The shoulder angle was determined as the angle between the vector joining the ipsilateral acromion-lateral epicondyle markers and the vector joining the C7–T4 markers. The elbow angle was determined as 180 degrees minus the angle between the vector formed by the ipsilateral acromionlateral epicondyle markers and a vector defined by the lateral epicondyle and the styloid process of the ulna. Trunk compensatory movement was described by trunk movement time, and trunk displacement was defined by the marker placed on the sternum. The trunk movement onset was defined for each trial as the time at which the tangential velocity rose above baseline by 5% of the peak tangential velocity of the sternum marker, and the trunk movement offset was defined as the time at which the tangential velocity fell and remained below 5% of the peak tangential velocity.27,43 Trunk movement time was defined as the time between the onset and the offset of the trunk movement. Trunk displacement was defined as the length of the trajectory of the marker placed on the sternum in the 3-dimensional space from trunk movement onset to trunk movement offset during the reaching task. The normality of the mean WMFT scores and the 8 kinematic predictors for the 2 distance tasks before and after treatment was statistically verified by examining the skewness (⫾1) and visually verified with histograms. All data sets met the criteria of normality. The collinearity between predictors was examined by analyzing variance inflation factors (ⱕ10), tolerance (ⱖ0.1), and condition index (⬎20). The datasets showed no collinearity problems.
Two steps were used to identify kinematic variables. First, the Pearson correlation coefficient (r) was used to examine the associations between kinematic variables and the mean WMFT-TIME and WMFT-FAS. The selection criterion for the entry of kinematic variables into the model was set to a P value of .25 to avoid the exclusion of important factors in model development.47 Although a typical threshold is a P value of .15, a P value of .25 is acceptable when no serious multicollinearity problem is found. Next, the remaining variables were used in a stepwise procedure to develop linear regression models for establishing concurrent and predictive validity. The concurrent validity of kinematic variables before and after treatment was estimated against the WMFT-TIME and WMFT-FAS (dependent variables) before and after treatment, respectively. The predictive validity of kinematic variables (predictors) before treatment was estimated against WMFT scores after treatment. The WMFT-TIME and WMFT-FAS before treatment were used as covariates to statistically control for individual differences that existed before treatment. Stepwise linear regressions were used to assess how much variance in the WMFT could be explained by kinematic variables and to determine which kinematic variables explained the largest amount of variance. The probability for entry into stepwise regression was set at .05. The adjusted R2 value, standardized coefficient (), and incremental R2 provided better estimations of the contribution of each predictor to the model. Role of the Funding Source This work was partly supported by the National Health Research Institutes (NHRI-EX-102-9920PI and NHRI-EX-102-10010PI), the National
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke Science Council (NSC-100-2314-B002-008-MY3 and NSC-99-2314-B182-014-MY3), and Chang Gung Memorial Hospital (CMRPD1C0401) in Taiwan. The funding sources supported the conduct of the research team, including study participant transportation expense, intervention expense, examiners’ salary, equipment maintenance and update, and data reduction. The funding sources had no influence on analysis, interpretation, or manuscript writing.
Table 1. Participant Characteristics and Clinical Measures of Motor Function (N⫽97) Variablea
Valueb
Age (y)
55.9⫾10.9
Mo after stroke onset
18.5⫾14.3
Sex Female
30 (31)
Male
67 (69)
Side of hemiparesis Right
43 (44)
Left
54 (56)
Lesion location
Results Participant characteristics and the WMFT-TIME and WMFT-FAS results before and after treatment are shown in Table 1. Kinematic data for reaching distance within and beyond arm’s length before and after treatment are shown in Table 2. Correlation Analysis Correlation analysis showed that there were 6 correlation coefficients with P values of greater than .25 and that 5 of the 6 were for the variable minimal angle of elbow flexion (Tab. 3). The kinematic variables with P values of less than .25 in the Pearson correlation coefficient analysis were entered into the stepwise multiple regression models. Multiple Regression Modeling Table 4 shows the results of stepwise multiple regression analysis for concurrent validity before treatment for the 2 reaching distances. For reaching distance within arm’s length, index movement time before treatment was a significant predictor for the WMFT-FAS model, accounting for 35.5% of the variance. Index movement time, index displacement, and maximal shoulder abduction before treatment were significant predictors for the WMFT-TIME model, accounting for 30.0% of the variance. For reaching distance beyond arm’s length, index movement time and minimal angle of elbow flexion before treatment were June 2014
Cortical
35 (36.1)
Subcortical
62 (63.9)
Type of stroke Ischemic
38 (39.2)
Hemorrhagic
59 (60.8)
Brunnstro¨m stage of upper extremity Proximal part
4.5 (3–6)
Distal part
4.5 (3–6)
WMFT-FAS before treatment
3.8⫾1.4
WMFT-TIME before treatment
4.8⫾1.8
WMFT-FAS after treatment
3.8⫾0.6
WMFT-TIME after treatment
3.6⫾0.7
a
WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time. b Continuous variables are shown as mean⫾standard deviation or median (range), and categoric variables are shown as number (percentage).
significant predictors for the WMFTFAS model, accounting for 33.7% of the variance. Index movement time and maximal angle of shoulder flexion before treatment were significant predictors for the WMFT-TIME model, accounting for 24.8% of the variance. Table 5 shows the results of stepwise multiple regression analysis for concurrent validity after treatment for the 2 reaching distances. For reaching distance within arm’s length, trunk displacement and Index PPV after treatment were significant predictors for the WMFT-FAS model, accounting for 40.5% of the variance. Trunk displacement, Index PPV, and trunk movement time after treatment were significant predictors for
the WMFT-TIME model, accounting for 47.7% of the variance. For reaching distance beyond arm’s length, index movement time and trunk displacement were significant predictors for the WMFT-FAS model, accounting for 37.4% of the variance. Trunk movement time, trunk displacement, and Index PPV were significant predictors for the WMFTTIME model, accounting for 36.4% of the variance. Table 6 summarizes the results of stepwise multiple regression analysis for predictive validity for the 2 reaching distances. After the WMFT-FAS before treatment was entered as the covariate, trunk movement time before treatment was found to be a significant predictor, explaining
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
849
Arm and Trunk Movement Kinematics in People With Stroke Table 2. Mean Values for Kinematic Variables XⴞSD Variable
a
Before Treatment
After Treatment
Within arm’s length reaching distance Index movement time (s)
1.4⫾0.7
1.1⫾0.5
Index displacement (mm)
374.7⫾131.2
353.8⫾101.3
Index PPV (%)
28.2⫾13
Maximal angle of shoulder flexion (°)
28.4⫾18.9
28.2⫾12.3 29⫾17.6
Maximal angle of shoulder abduction (°)
41.7⫾15.7
39.5⫾14.7
Minimal angle of elbow flexion (°)
93.2⫾15.4
94.5⫾14.1
Trunk movement time (s)
1.5⫾0.7
1.2⫾0.4
Trunk displacement (mm)
78.5⫾37.7
80⫾39.3
Beyond arm’s length reaching distance Index movement time (s)
1.7⫾0.7
1.5⫾0.6
Index displacement (mm)
701.6⫾129.8
685.4⫾132.8
Index PPV (%)
a
24⫾10.2
25.1⫾9
Maximal angle of shoulder flexion (°)
69.1⫾21.1
68.4⫾19.3
Maximal angle of shoulder abduction (°)
73.6⫾17.2
71.7⫾17.5
Minimal angle of elbow flexion (°)
71.8⫾17.5
72.2⫾15.3
Trunk movement time (s)
1.7⫾0.6
1.6⫾0.6
Trunk displacement (mm)
279.9⫾64.7
266.2⫾56.9
Index PPV⫽percentage of index movement time at which index peak velocity occurs.
6.9% of the variance (P⫽.001) in the WMFT-FAS for reaching distance within arm’s length; trunk displacement and trunk movement time before treatment were found to be significant predictors, explaining 9.4% of the variance (P⫽.013) for reaching distance beyond arm’s length. After the WMFT-TIME before treatment was entered as the covariate, trunk movement time before treatment was found to be a significant predictor, explaining 14.4% of the variance (P⬍.001) in the WMFTTIME for reaching distance within arm’s length; trunk displacement and trunk movement time before treatment were found to be significant predictors, explaining 14.9% of the variance (P⫽.004) for reaching distance beyond arm’s length.
Discussion To our knowledge, the present study is the first to investigate the concur850
f
Physical Therapy
rent validity of arm-trunk kinematics against motor function (WMFT) before and after 3 to 4 weeks of intensive treatment and the validity of kinematic variables for predicting motor function for reaching within and beyond arm’s length. Our results showed that the kinematic variables that can explain a proportion of the variance in the WMFT-FAS and the WMFT-TIME depend on the treatment phase (before or after treatment) and the reaching distance. For concurrent validity of the 8 kinematic variables that were examined, index movement time was most closely associated with the WMFTFAS and WMFT-TIME before treatment, whereas combinations of trunk compensatory movement variables (trunk displacement, trunk movement time, or both) and endpoint coordinate strategy variables (Index PPV and index movement time) were the significant contribu-
tors after treatment. For predictive validity, less than 15% of the variance in the WMFT was explained by trunk movement time combined with trunk displacement. In general, motor function could be better explained by the kinematic variables when the participants performed the reaching task within arm’s length rather than beyond arm’s length. These findings corroborated the results of previous studies showing that kinematic measures can reflect relatively closely a patient’s actual movement performance during functional activities.14,22,23 The findings of the present study also extend the results of previous studies14,22,23 by concurrently using endpoint coordinate strategies, joint angle coordinate strategies, and trunk compensatory movement variables to establish concurrent validity before and after treatment and to establish predictive validity. Concurrent Validity Before Treatment Index movement time was the primary contributor to the largest amount of variance (23%–36%) in the WMFT for both reaching distances. The positive relationship between index movement time and the WMFT-FAS indicated that a longer time to reach for pressing a desk bell reflected a better movement quality in performing a movement or functional task (eg, extending elbow movement, stacking checkers). One possible account for this relationship is the speed–accuracy trade-off.48 Participants spent more time to reach and press the desk bell to more accurately and successfully achieve the task goal. The accuracy and success of accomplishing tasks are associated with the movement quality measured with the WMFT. Another possible explanation is that participants spent more time to adjust or reduce pathologic synergies for performing a reaching task. A reduction in pathologic synergies
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke Table 3. Correlations (P Values) Between Kinematic Variables and Mean Wolf Motor Function Test Scores Before and After Treatmenta Correlation (P) Reaching Distance Within arm’s length
Kinematics Before Treatment
Before Treatment WMFT-FAS
WMFT-TIME b
Kinematics After Treatment
After Treatment WMFT-FAS b
⫺.542 (⬍.001)
WMFT-TIME b
⫺.525 (⬍.001)
b
Index movement time
.602 (⬍.001)
.508 (⬍.001)
Index displacement
.415 (⬍.001)b
.478 (⬍.001)b ⫺.468 (⬍.001)b ⫺.454 (⬍.001)b
Index PPV
⫺.334 (⬍.001)b ⫺.273 (.007)b b
.276 (.006)
b
Maximal angle of shoulder flexion
.297 (.03)
Maximal angle of shoulder abduction
.219 (.032)b
.208 (.041)b
Minimal angle of elbow flexion
⫺.273 (.007)b
⫺.228 (.025)b
.35 (⬍.001)b ⫺.193 (.058)
b
⫺.179 (.079)b
.095 (.353)
.292 (.004)b ⫺.192 (.06)
.133 (.195)b
MinElbFlex
.107 (.297)
.054 (.601)
.337 (.001)b
.261 (.01)b
Maximal angle of shoulder flexion
.181 (.076)
Maximal angle of shoulder abduction
.117 (.255)
Minimal angle of elbow flexion
⫺.233 (.022)b
.227 (.025)
⫺.31 (.002)b
⫺.372 (⬍.001)b ⫺.323 (.001)b
.188 (.065)b
⫺.221 (.03)b
⫺.161 (.116)
.47 (⬍.001)b
⫺.301 (.003)b
.477 (⬍.001)b ⫺.501 (⬍.001)b ⫺.456 (⬍.001)b
.372 (⬍.001)b
.422 (⬍.001)b
⫺.339 (.001)b
.564 (⬍.001)b
b
⫺.454 (⬍.001)b ⫺.406 (⬍.001)b
MaxShAbd
.146 (.155)b
b
Index Displ Index PPV
⫺.454 (⬍.001)b
⫺.161 (.114)b
.255 (.012)b
⫺.381 (⬍.001)b ⫺.29 (.004)b
⫺.474 (⬍.001)
⫺.212 (.037)b
Trunk displacement
Index PPV
Index MT
⫺.223 (.028)
.409 (⬍.001)b ⫺.525 (⬍.001)b ⫺.569 (⬍.001)b
Index displacement
WMFT-TIME b
MaxShFlex
.526 (⬍.001)b
Beyond arm’s Index length movement time
WMFT-FAS
b
b
Trunk movement time
⫺.343 (.001)b
Correlation (P) After Treatment
.321 (.001)b
⫺.435 (⬍.001)b ⫺.446 (⬍.001)b
Trunk Displ
⫺.548 (⬍.001)b ⫺.558 (⬍.001)b
Index MT
⫺.541 (⬍.001)b ⫺.505 (⬍.001)b
Index Displ
⫺.402 (⬍.001)b ⫺.316 (.002)b
Index PPV
.32 (.001)b
.359 (⬍.001)b
b
MaxShFlex
⫺.245 (.016)
⫺.188 (.065)b
⫺.183 (.073)b
MaxShAbd
⫺.242 (.017)b
⫺.237 (.019)b
.121 (.237)b
.119 (.246)b
MinElbFlex
.07 (.498)
.08 (.437)
b
⫺.159 (.119)
Trunk MT
Trunk movement time
.56 (⬍.001)b
.388 (⬍.001)b ⫺.505 (⬍.001)b ⫺.464 (⬍.001)b
Trunk displacement
.26 (.01)b
.229 (.024)b
⫺.374 (⬍.001)b ⫺.427 (⬍.001)b
b
⫺.21 (.039)b
Trunk MT
⫺.538 (⬍.001)b ⫺.521 (⬍.001)b
Trunk Displ
⫺.459 (⬍.001)b ⫺.436 (⬍.001)b
a WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time, Index PPV⫽percentage of index movement time at which index peak velocity occurs. b Kinematic variables entered into multiple regression models for further analysis.
may be reflected by better movement quality. Concurrent Validity After Treatment Of the 8 kinematic variables that were measured, trunk compensatory June 2014
movement variables were most closely associated with the WMFTFAS and WMFT-TIME (partial R2⫽ .30 –.35). The Index PPV (partial R2⫽.04 –.15) was the second most important contributor to the variance in the WMFT-TIME and WMFT-
FAS for both reaching distances after treatment, except for the regression model of WMFT-FAS for reaching distance beyond arm’s length. Previous studies found similar results; they showed that trunk displacement during reaching, reach-to-grasp,23
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
851
Arm and Trunk Movement Kinematics in People With Stroke Table 4. Results of Stepwise Multiple Regression Analyses and Collinearity Test for Concurrent Validity Before Treatment
Regression Modela
 Coefficient (Standardized)
Incremental R2
Adjusted R2 (Model P)
Tolerance
Variance Inflation Factor
Condition Index
Within arm’s length reaching distance Model 1 dependent variable: WMFT-FAS before treatment Index movement time
.355 (⬍.001) .602
.362
Model 2 dependent variable: WMFT-TIME before treatment
4.282 1
1
.3 (.048)
11.121
Index movement time
.285
.259
.521
1.92
Index displacement
.294
.293
.523
1.88
Maximal angle of shoulder abduction
.176
.322
.943
1.06
Beyond arm’s length reaching distance Model 3 dependent variable: WMFT-FAS before treatment Index movement time Minimal angle of elbow flexion
.337 (.032) .547
.318
.991
1.009
⫺.182
.351
.991
1.009
Model 4 dependent variable: WMFT-TIME before treatment
a
10.998
.248 (.034)
8.621
Index movement time
.462
.227
.994
1.006
Maximal angle of shoulder flexion
.191
.264
.994
1.006
WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time.
and drinking14 tasks could reflect not only motor impairment but also motor function in people with stroke. Although trunk movement time has not been used often to represent trunk compensation, the results of the present study showed that trunk movement time was a significant factor and that up to 31% of the variance in the WMFT was attributable to trunk movement time. A previous study49 also suggested that a shorter trunk movement time may reflect less trunk recruitment, indicating that trunk movement time could be considered an important indicator of trunk compensation. In addition, the combination of trunk displacement and trunk movement time seems to relate to motor function better than either one alone. The kinematic variables reflecting motor function differed before and 852
f
Physical Therapy
after intensive treatments. Possible reasons for trunk compensatory movement variables being important indicators only after treatment are that people with stroke may use trunk compensatory movement as a habitual response learned after stroke31 and that reaching movements are primarily controlled by endpoint coordinate strategy variables. After intensive treatments, people may have a greater capacity to voluntarily, separately control trunk and arm participation in UE movements.27 Because the recruitment of the trunk became an important contributor for performing the reaching task after rehabilitation, it may serve as a variable reflecting motor function. These data suggest the need for careful selection of kinematic variables to describe the motor function of people with stroke at various phases, such as before treatment and after treatment. In addition, these previously unreported
data indicate the trunk movements that may partially reflect motor function after treatment. The joint angle coordinate strategy variables for the UE did not play important roles in explaining the variance in motor function. A possible reason is that the central nervous system preprograms UE movement through endpoint trajectory, together with arm and trunk coordination, rather than joint angles.34 The findings of the present study were similar to those of Subramanian et al23 but in contrast to those of Massie et al,22 who found that elbow extension during reaching was the only predictor to reflect UE motor function. The discrepancy between the findings of the present study and those of the study of Massie et al22 may be attributable to the nature of the kinematic task. In the study of Massie et al,22 participants were asked to reach between 2 targets
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke Table 5. Results of Stepwise Multiple Regression Analyses and Collinearity Test for Concurrent Validity After Treatment  Coefficient (Standardized)
Regression Modela
Incremental R2
Adjusted R2 (Model P)
Tolerance
Variance Inflation Factor
Condition Index
Within arm’s length reaching distance Model 1 dependent variable: WMFT-FAS after treatment
.405 (⬍.001) ⫺.495
Trunk displacement Index PPV
.347
7.167
.3
.997
1.024
.417
.997
1.024
Model 2 dependent variable: WMFT-TIME after treatment
.477 (.02)
Trunk displacement Index PPV Trunk movement time
11.986
⫺.452
.311
.917
1.09
.328
.462
.858
1.166
⫺.195
.493
.808
1.238
Beyond arm’s length reaching distance Model 3 dependent variable: WMFT-FAS after treatment
.374 (⬍.001)
Index movement time
⫺.442
.292
Trunk displacement
⫺.324
.387
Model 4 dependent variable: WMFT-TIME after treatment
11.511 .907
1.103
.907
1.103
.364 (.019)
14.918
Trunk movement time
⫺.358
.271
.807
1.239
Trunk displacement
⫺.284
.346
.88
1.136
.205
.384
.897
1.114
Index PPV a
WMFT-FAS⫽Wolf Motor Function Test functional ability score, Index PPV⫽percentage of index movement time at which index peak velocity occurs, WMFT-TIME⫽Wolf Motor Function Test performance time.
for 4 continuous cycles, whereas in the present study, the discrete functional task of pressing a desk bell on a table was used. Predictive Validity of Kinematic Measures The trunk compensatory movement variables before treatment (trunk movement time and trunk displacement) were significant predictors, explaining 6.9% to 9.4% of the variance in the WMFT-FAS and 14.4% to 14.9% of the variance in the WMFTTIME after treatment. A shorter trunk movement time and less trunk displacement before intensive treatment are related to better temporal efficiency for performing functional tasks (WMFT-TIME) after intensive treatment in people with stroke and mild to moderate levels of UE motor function. Although only limited variJune 2014
ance in the WMFT-TIME after treatment was explained by the trunk variables before treatment, Kaplan and Saccuzzo50 suggested that R2 values of greater than .1 with statistical significance could be considered the minimal threshold for a predictor in a predictive model. Therefore, trunk variables may be predictors and may have some prognostic value. Our results confirmed the finding of Cirstea and Levin27 that better trunk control performance before treatment in people with stroke may indicate better execution capacity and increased opportunities to obtain greater functional gains from intensive treatment. Fritz et al24 found the motor ability of finger extension to be a plausible predictor of motor function after treatment. Adding trunk variables to the previously suggested predictive models may
enhance the accuracy of prediction of motor function after treatment. That trunk variables explained less than 10% of the variance in the WMFT-FAS may have been attributable to the WMFT-FAS being less objective, possibly resulting in the high variability of the scores. Kinematic Validity for Reaching to a Target Within and Beyond Arm’s Length The kinematic variables selected in the regression models explained and predicted larger amounts of variance in the WMFT scores for the task within arm’s length (R2⫽.069 –.493) than for the task beyond arm’s length (R2⫽.094 –.387). The results indicated that a reaching task within arm’s length may be more suitable for evaluating kinematic performance to describe motor function in people
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
853
Arm and Trunk Movement Kinematics in People With Stroke Table 6. Results of Stepwise Multiple Regression Analyses and Collinearity Test for Predictive Validity
Regression Modela
 Coefficient (Standardized)
Incremental R2
Adjusted R2 (Model P)
Tolerance
Variance Inflation Factor
Condition Index
Within arm’s length reaching distance Model 1 dependent variable: WMFT-FAS after treatment
.384 (.001)
6.719
WMFT-FAS before treatment
⫺.410
.328
.723
1.383
Trunk movement time before treatment
⫺.309
.397
.723
1.383
Model 2 dependent variable: WMFT-TIME after treatment
.429 (⬍.001)
6.2
WMFT-TIME before treatment
⫺.375
.297
.833
1.2
Trunk movement time before treatment
⫺.416
.441
.833
1.2
Beyond arm’s length reaching distance Model 3 dependent variable: WMFT-FAS after treatment
.403 (.013)
WMFT-FAS before treatment
⫺.379
.328
.667
1.499
Trunk displacement before treatment
⫺.220
.382
.923
1.084
Trunk movement time before treatment
⫺.242
.422
.679
1.473
Model 4 dependent variable: WMFT-TIME after treatment
a
12.384
.428 (.004)
12.338
WMFT-TIME before treatment
⫺.383
.297
.829
1.207
Trunk displacement before treatment
⫺.283
.394
.925
1.081
Trunk movement time before treatment
⫺.251
.446
.830
1.205
WMFT-FAS⫽Wolf Motor Function Test functional ability score, WMFT-TIME⫽Wolf Motor Function Test performance time.
with stroke than a reaching task beyond arm’s length. Unlike people who are healthy, who use only interjoint recruitment and coordination of the UE to perform reaching tasks within arm’s length, people with stroke may use trunk compensatory movement to assist the UE to perform reaching tasks within arm’s length to a different degree.27,29,30 Limitations The present study has a few limitations. One is the generalizability of the study results. The data used for the present study were pooled from several clinical trials investigating different treatment interventions. These clinical trials recruited people with stroke and moderate to high levels of motor ability. Therefore, our results cannot be attributed to the effect of a specific treatment and can be generalized only to people 854
f
Physical Therapy
with chronic stroke and similar levels of motor impairment. Second, the Brunnstro ¨ m stage, an inclusion criterion for this secondary analysis study, may not be a desirable measurement for clinical settings. Other suitable clinical measurements, such as the Fugl-Meyer Assessment for the UE, could be used as inclusion criteria in further studies. Third, variability among the testers in the different trials may have affected the results; however, all of the testers across the clinical trials received training before they evaluated participants to ensure that the same evaluation standard was used.
Conclusion The uniqueness of the present study is that it separately examined the current validity of kinematic variables for different reaching distances before and after intensive UE train-
ing and examined the predictive validity for both distances. The results provide a better understanding of the meaning and nature of different types of kinematic variables applied in rehabilitation research. The selection of different types of kinematic variables to describe UE motor function at various phases, such as before and after treatment, in patients with stroke is necessary. Before treatment, index movement time seemed to be a kinematic variable with a relatively better association with motor function. After treatment, trunk compensatory movement variables (trunk movement time and trunk displacement) became salient contributors, possibly because participants then had a greater capacity to separately control trunk participation and arm participation in UE movements. For predictive validity, trunk variables may be considered
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics in People With Stroke possible predictors of UE motor function after treatment. However, it would be better to use these trunk variables as adjuncts to enhance the accuracy of prediction because the amount of variance in motor function that kinematic variables could explain is small. A reaching task within arm’s length may be better than a reaching task beyond arm’s length for evaluating kinematic performance to partially describe motor function in people with stroke. Dr Wu and Dr Lin provided concept/idea/ research design, project management, and fund procurement. Dr Wu and Dr Liing provided writing. Dr Wu provided data collection and institutional liaisons. Dr Wu, Dr Liing, and Professor H. Chen provided data analysis. Professor C. Chen provided study participants. Dr Wu, Professor C. Chen, and Dr Lin provided facilities/equipment. Dr Lin provided consultation (including review of manuscript before submission). This study was approved by the institutional review board at each participating site. This work was partly supported by the National Health Research Institutes (NHRIEX-102-9920PI and NHRI-EX-102-10010PI), the National Science Council (NSC-1002314-B-002-008-MY3 and NSC-99-2314-B182-014-MY3), and Chang Gung Memorial Hospital (CMRPD1C0401) in Taiwan. DOI: 10.2522/ptj.20130101
References 1 Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study. Arch Phys Med Rehabil. 1994;75:394 –398. 2 Mayo NE, Wood-Dauphine´e S, Ahmed S, et al. Disablement following stroke. Disabil Rehabil. 1999;21:258 –268. 3 Olsen TS. Arm and leg paresis as outcome predictors in stroke rehabilitation. Stroke. 1990;21:247–251. 4 Feys HM, De Weerdt WJ, Selz BE, et al. Effect of a therapeutic intervention for the hemiplegic upper limb in the acute phase after stroke: a single-blind, randomized, controlled multicenter trial. Stroke. 1998; 29:785–792. 5 Shumway-Cook A, Woollacott MH. Normal reach, grasp, and manipulation. In: Lupash E, ed. Motor Control: Translating Research Into Clinical Practice. 4th ed. Philadelphia, PA: Wolters Kluwer, Lippincott Williams & Wilkins; 2012:477–501.
June 2014
6 Shumway-Cook A, Woollacott MH. Clinical management of the patient with reach, grasp, and manipulation disorders. In: Lupash E, ed. Motor Control: Translating Research Into Clinical Practice. 4th ed. Philadelphia, PA: Wolters Kluwer, Lippincott Williams & Wilkins; 2012:552–594. 7 Roby-Brami A, Feydy A, Combeaud M, et al. Motor compensation and recovery for reaching in stroke patients. Acta Neurol Scand. 2003;107:369 –381. 8 Reisman DS, Scholz JP. Aspects of joint coordination are preserved during pointing in persons with post-stroke hemiparesis. Brain. 2003;126:2510 –2527. 9 Caimmi M, Carda S, Giovanzana C, et al. Using kinematic analysis to evaluate constraint-induced movement therapy in chronic stroke patients. Neurorehabil Neural Repair. 2008;22:31–39. 10 Thielman GT, Dean CM, Gentile AM. Rehabilitation of reaching after stroke: taskrelated training versus progressive resistive exercise. Arch Phys Med Rehabil. 2004;85:1613–1618. 11 Massie CL, Malcolm MP, Greene D, Thaut M. The effects of constraint-induced therapy on kinematic outcomes and compensatory movement patterns: an exploratory study. Arch Phys Med Rehabil. 2009;90: 571–579. 12 Wagner JM, Dromerick AW, Sahrmann SA, Lang CE. Upper extremity muscle activation during recovery of reaching in subjects with post-stroke hemiparesis. Clin Neurophysiol. 2007;118:164 –176. 13 Kitago T, Liang J, Huang VS, et al. Improvement after constraint-induced movement therapy: recovery of normal motor control or task-specific compensation? Neurorehabil Neural Repair. 2013; 27:99 –109. 14 Alt Murphy M, Willen C, Sunnerhagen KS. Movement kinematics during a drinking task are associated with the activity capacity level after stroke. Neurorehabil Neural Repair. 2012;26:1106 –1115. 15 Wagner JM, Rhodes JA, Patten C. Reproducibility and minimal detectable change of three-dimensional kinematic analysis of reaching tasks in people with hemiparesis after stroke. Phys Ther. 2008;88:652– 663. 16 Schneiberg S, McKinley P, Gisel E, et al. Reliability of kinematic measures of functional reaching in children with cerebral palsy. Dev Med Child Neurol. 2010;52: e167– e173. 17 Wagner JM, Lang CE, Sahrmann SA, et al. Relationships between sensorimotor impairments and reaching deficits in acute hemiparesis. Neurorehabil Neural Repair. 2006;20:406 – 416. 18 Lang CE, Wagner JM, Dromerick AW, Edwards DF. Measurement of upperextremity function early after stroke: properties of the Action Research Arm Test. Arch Phys Med Rehabil. 2006;87:1605– 1610. 19 Wagner JM, Lang CE, Sahrmann SA, et al. Sensorimotor impairments and reaching performance in subjects with poststroke hemiparesis during the first few months of recovery. Phys Ther. 2007;87:751–765.
20 Zackowski KM, Dromerick AW, Sahrmann SA, et al. How do strength, sensation, spasticity and joint individuation relate to the reaching deficits of people with chronic hemiparesis? Brain. 2004;127:1035–1046. 21 Edwards DF, Lang CE, Wagner JM, et al. An evaluation of the Wolf Motor Function Test in motor trials early after stroke. Arch Phys Med Rehabil. 2012;93:660 – 668. 22 Massie CL, Fritz S, Malcolm MP. Elbow extension predicts motor impairment and performance after stroke. Rehabil Res Pract. 2011;2011:381978. 23 Subramanian SK, Yamanaka J, Chilingaryan G, Levin MF. Validity of movement pattern kinematics as measures of arm motor impairment poststroke. Stroke. 2010;41:2303–2308. 24 Fritz SL, Light KE, Patterson TS, et al. Active finger extension predicts outcomes after constraint-induced movement therapy for individuals with hemiparesis after stroke. Stroke. 2005;36:1172–1177. 25 Loewen SC, Anderson BA. Predictors of stroke outcome using objective measurement scales. Stroke. 1990;21:78 – 81. 26 Hollerbach JH. Planning arm movements. In: Osherson DN, Kosslyn SM, Hollerbach JM, eds. Visual Cognition and Action. Vol 2. Cambridge, MA: MIT; 1990:183–211. 27 Cirstea MC, Levin MF. Compensatory strategies for reaching in stroke. Brain. 2000; 123:940 –953. 28 Robertson JV, Roby-Brami A. The trunk as a part of the kinematic chain for reaching movements in healthy subjects and hemiparetic patients. Brain Res. 2011;1382: 137–146. 29 Levin MF, Michaelsen SM, Cirstea CM, Roby-Brami A. Use of the trunk for reaching targets placed within and beyond the reach in adult hemiparesis. Exp Brain Res. 2002;143:171–180. 30 Archambault P, Pigeon P, Feldman AG, Levin MF. Recruitment and sequencing of different degrees of freedom during pointing movements involving the trunk in healthy and hemiparetic subjects. Exp Brain Res. 1999;126:55– 67. 31 Woodbury ML, Howland DR, McGuirk TE, et al. Effects of trunk restraint combined with intensive task practice on poststroke upper extremity reach and function: a pilot study. Neurorehabil Neural Repair. 2009;23:78 –91. 32 van Vliet PM, Sheridan MR. Ability to adjust reach extent in the hemiplegic arm. Physiotherapy. 2009;95:176 –184. 33 Kaminski TR, Bock C, Gentile AM. The coordination between trunk and arm motion during pointing movements. Exp Brain Res. 1995;106:457– 466. 34 Ma S, Feldman AG. Two functionally different synergies during arm reaching movements involving the trunk. J Neurophysiol. 1995;73:2120 –2122. 35 McCrea PH, Eng JJ, Hodgson AJ. Biomechanics of reaching: clinical implications for individuals with acquired brain injury. Disabil Rehabil. 2002;24:534 –541.
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
Physical Therapy f
855
Arm and Trunk Movement Kinematics in People With Stroke 36 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. 37 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. 38 Wu CY, Yang CL, Chuang LL, et al. Effect of therapist-based versus robot-assisted bilateral arm training on motor control, functional performance, and quality of life after chronic stroke: a clinical trial. Phys Ther. 2012;92:1006 –1016. 39 Wu CY, Chuang LL, Lin KC, et al. Randomized trial of distributed constraint-induced therapy versus bilateral arm training for the rehabilitation of upper-limb motor control and function after stroke. Neurorehabil Neural Repair. 2011;25:130 –139. 40 Lin KC, Chen YA, Chen CL, et al. The effects of bilateral arm training on motor control and functional performance in chronic stroke: a randomized controlled study. Neurorehabil Neural Repair. 2010; 24:42–51.
856
f
Physical Therapy
41 Wu CY, Chen CL, Tang SF, et al. Kinematic and clinical analyses of upper-extremity movements after constraint-induced movement therapy in patients with stroke: a randomized controlled trial. Arch Phys Med Rehabil. 2007;88:964 –970. 42 DeJong SL, Schaefer SY, Lang CE. Need for speed: better movement quality during faster task performance after stroke. Neurorehabil Neural Repair. 2012;26:362–373. 43 Levin MF, Kleim JA, Wolf SL. What do motor “recovery” and “compensation” mean in patients following stroke? Neurorehabil Neural Repair. 2009;23:313–319. 44 Alt Murphy M, Willen C, Sunnerhagen KS. Kinematic variables quantifying upperextremity performance after stroke during reaching and drinking from a glass. Neurorehabil Neural Repair. 2011;25:71– 80. 45 Subramanian SK, Lourenco CB, Chilingaryan G, et al. Arm motor recovery using a virtual reality intervention in chronic stroke: randomized control trial. Neurorehabil Neural Repair. 2013;27:13–23.
46 Skidmore ER, Becker JT, Whyte EM, et al. Cognitive impairments and depressive symptoms did not impede upper limb recovery in a clinical repetitive task practice program after stroke: a pilot study. Am J Phys Med Rehabil. 2012;91:327–331. 47 Hosmer DW, Lemeshow S. Applied Logistic Regression. New York, NY: John Wiley & Sons Inc; 1989. 48 Fitts PM. The information capacity of the human motor system in controlling the amplitude of movement. J Exp Psychol. 1954;47:381–391. 49 Ma HI, Hwang WJ, Wang CY, et al. Trunkarm coordination in reaching for moving targets in people with Parkinson’s disease: comparison between virtual and physical reality. Hum Mov Sci. 2012;31:1340 –1352. 50 Kaplan KM, Saccuzzo DP. Psychological Testing: Principles, Application, and Issues. 5th ed. Belmont, CA: Wadsworth/ Thomson Learning; 2001.
Volume 94 Number 6 Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
June 2014
Arm and Trunk Movement Kinematics During Seated Reaching Within and Beyond Arm's Length in People With Stroke: A Validity Study Ching-yi Wu, Rong-jiuan Liing, Hsieh-ching Chen, Chia-ling Chen and Keh-chung Lin PHYS THER. 2014; 94:845-856. Originally published online January 30, 2014 doi: 10.2522/ptj.20130101
References
This article cites 46 articles, 24 of which you can access for free at: http://ptjournal.apta.org/content/94/6/845#BIBL
Subscription Information
http://ptjournal.apta.org/subscriptions/
Permissions and Reprints http://ptjournal.apta.org/site/misc/terms.xhtml Information for Authors
http://ptjournal.apta.org/site/misc/ifora.xhtml
Downloaded from http://ptjournal.apta.org/ by guest on June 9, 2015
