Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Stroke. Author manuscript; available in PMC 2010 May 1.
Published in final edited form as:
PMCID: PMC2718540

Active Range of Motion predicts Upper Extremity Function Three months post-stroke

Justin A. Beebe, PT, PhD1 and Catherine E. Lang, PT, PhD1,2,3


Background and Purpose:

After stroke, 80% of patients experience acute paresis of the upper extremity and only approximately 1/3 achieve full functional recovery. Predicting functional recovery for these patients is highly important in order to provide focused, cost effective rehabilitation. Our purpose was to examine if early measures of upper extremity active range of motion (AROM) could predict recovery of upper extremity function, and to describe the trajectory of upper extremity AROM recovery over time.


33 subjects were tested at 1 month and then at 3 months after stroke. Upper extremity function was measured with six standardized clinical tests which were synthesized into a single, sensitive score for upper extremity function using principal component analysis. The ability to move each segment (AROM) was measured using a 3D electromagnetic tracking system.


Stepwise multiple regression revealed that AROM of the shoulder and middle finger segments taken at 1 month, could predict 71% of the variance in upper extremity function at 3 months. All segments of the upper extremity recover similarly and no evidence of a proximal to distal gradient in motor deficits appeared over time.


Simple AROM measurements of the upper extremity taken within 1 month post stroke can be used to predict upper extremity function at 3 months. This information is important for determining the prognosis of upper extremity functional recovery.


Recovery of upper extremity movement and function is a major concern facing individuals who have upper extremity paresis following stroke 1. Of the 80% of patients experiencing acute paresis of the upper extremity after stroke, only approximately 1/3 achieve full functional recovery 2-4. Predicting functional recovery for these patients is highly important in order to provide focused, cost effective rehabilitation. While there are several good models for predicting mortality 5-7, life satisfaction 8, discharge destination 9, or likelihood of independent function 5, 7, 7, 10-16, only a few models have tried to predict recovery of upper extremity function specifically 1, 17-21. Of those, many of the predictors were complicated or time consuming 18-21, or predicted a small portion of the variance 17. A quick and simple method to predict upper extremity functional recovery would be of high utility to patients and clinicians alike.

We have recently found that the simple measure of active range of motion (AROM) can account for 82% of the variance in upper extremity function in people with hemiparesis less than 1 month post stroke 22. Conceptually, AROM against gravity in people post stroke can be considered a quick measure of the capacity of the spared motor system to activate the spinal motoneuron pools that move a given segment. Our result opens up the possibility that AROM values are predictive of later upper extremity function and if indeed their predictive value is strong enough, these values may be used to estimate arm function prognosis in individual patients. Thus, the purpose of the present study was to examine if early measures of upper extremity AROM could predict recovery of upper extremity function.

Here, we extend our earlier work 22 by following patients out to 3 months post stroke. We evaluated AROM and upper extremity function at both time points, and hypothesize that AROM measured within 1 month will be predictive of upper extremity function evaluated at 3 months post stroke. A secondary purpose of this paper was to examine how AROM at various upper extremity segments recovered over this same time period. This was important to evaluate because relatively greater recovery in one segment vs. another could impact the predictive value of the early AROM measurements.

Materials and Methods

We prospectively studied 33 subjects with hemiparesis due to stroke between July 2005 and December 2007. This accounted for approximately 5% of all subjects screened for inclusion into the study. Subjects were recruited from the Cognitive Rehabilitation Research Group Stroke Registry (CRRG) based on the presence of hemiparesis. The CRRG is a tertiary program that captures all patients who are admitted to Barnes-Jewish Hospital and who are seen by the Stroke Management and Rehabilitation Team. Subjects were included if they 1) had a diagnosis of ischemic or hemorrhagic stroke by a stroke neurologist within one month of onset, 2) had CT or MRI imaging data consistent with clinical presentation, 3) had persistent hemiparesis with a score of 1 to 4 on the Motor Arm item of the National Institutes of Health Stroke Scale (NIHSS), 4) had evidence of preserved cognition as indicated by a score of 0 or 1 on the Consciousness and Communication item of the NIHSS, and 5) had the ability to follow 2-step commands. Patients were excluded from the study because of the following reasons 1) had orthopaedic or other medical conditions that limited the affected upper extremity prior to the stroke, 2) had a prior history of hemiparesis or stroke, 3) had hemispatial neglect as evidenced by a score of 2 on the Extinction and Inattention item of the NIHSS, 4) had severe aphasia as evidenced by a score of 2 or 3 on the Language item of the NIHSS, 5) had complete hemianopsia as evidenced by a score of 2 or 3 on the Visual item of the NIHSS, 6) the subject was unable to give informed consent, or 7) the subject chose not to participate. Characteristics of the group are provided in Table 1. This study was approved by the Washington University Human Research Protection Office, and all participants provided informed consent prior to participation.

Table 1
Subject Characteristics

By the 3 month follow up visit, 28 subjects remained. Subject attrition by the 3 month time point was attributed to 1 deceased, 2 with additional medical complications that prohibited them from continuing, and 2 who could not be reached for follow-up. Comparison of baseline values for those subjects lost between time points indicated no significant differences between those who did not finish and those who completed the study.

Testing paradigm

Subjects were tested for their ability to use the affected upper extremity for functional activities and for their ability to move nine segments of that upper extremity against gravity (i.e. AROM) at two time points: within 1 month, and 3 months after stroke. Initial testing was completed, in most cases, during a single testing session lasting approximately two hours. In a few cases, during the 1 month time point only, testing was completed in two separate one hour sessions within a 24-hour period due to rehabilitation schedules or medical testing. All 3 month testing occurred during a single testing session. All kinematic testing occurred in our laboratory.

Measurement of upper extremity function

All subjects underwent a battery of six standardized clinical tests of upper extremity function. A battery was used because a single test may not suitably capture and quantify upper extremity function across this patient population 3, 23. The tests were selected based on published data (at the time of study initiation) regarding reliability, validity, normative values, and appropriateness for use with people with stroke. All clinical tests were performed on both sides such that the unaffected upper extremity served as the matched control for the affected side. The following six tests were used: Grip Strength 24, 25, Pinch Strength 26, 27, Action Research Arm Test (ARAT) 28-33, Jebsen Taylor Test of Hand Function 34, 9-Hole Peg Test 35, and the Stroke Impact Scale-Hand 36-38. The first task of the Jebsen, writing a sentence, was not used 39 because it is dependent on hand dominance and education level.

The results of the test battery were synthesized to yield a single measure of upper extremity function for each subject using principal components analysis 22, 40-42. For the principal component analysis, the 6 Jebsen subtest scores were entered separately, such that a total of 11 test scores were used for each individual42. The first principal component (eigenvalue = 9.37) explained 85.2% of the variance in the test scores, and no other components yielded eigenvalues greater than 0.53. The weighted linear coefficients of the first principal component were then used to generate a single upper extremity function score for each subject.

Measurement of active range of motion

The AROM measurements performed in this study were highly instrumented because our overall testing paradigm at study initiation included the calculation of different variables for separate investigations within the same sample. We note that these same AROM measurements could have been obtained bedside or in the clinic with a standard goniometer. Briefly, kinematic techniques were used to quantify AROM of nine upper extremity segments on the affected side, contralateral to the lesion. Three-dimensional movements of the upper extremity were captured using an electromagnetic tracking system (The Motion Monitor, Innovative Sports Training Inc, Chicago IL). Nine sensors were attached to the trunk (1), upper arm (1), forearm (1), hand (1), and fingers (5). Subjects were seated in a straight-backed chair and were able to maintain a seated position for the duration of testing. While seated with the upper extremity hanging down by the side, subjects were instructed to make movements of one segment while keeping the other upper extremity segments still. Care was taken to ensure that the tested upper extremity did not contact or otherwise be obstructed by the side edge of the chair. The nine instructed movements were: shoulder flexion, elbow flexion, forearm supination/pronation, wrist flexion/extension, thumb flexion, index finger flexion, middle finger flexion, ring finger flexion, and little finger flexion. Subjects were allowed to move at a self-selected pace and instructed to move the segment as far as they could then return to the starting position. Because the start position was with the upper extremity hanging down by the side, two movements, forearm supination/pronation, and thumb flexion were tested in a gravity eliminated position. In these two movements, the moment arms were negligible such that gravity was not likely to have a substantial influence in a subject's ability to perform the movements. Each movement was first demonstrated by the tester, and replicated by the subject prior to each trial. Two trials of each movement were recorded. The instrumentation and software used permitted accurate measurement of AROM irrespective of movement at non-instructed segments, i.e. elbow AROM was the range through which the elbow moved regardless of whether the shoulder and the wrist also moved. Due to upgrades in the data collection system during the study, data was collected at either 60 Hz or 100 Hz and stored offline for subsequent analyses. Kinematic data were low-pass filtered at 6 Hz using a second-order Butterworth filter. Motion Monitor software (Innovative Sports Training Inc, Chicago IL), and custom-written software in MATLAB (The Mathworks, Nadick MA) was used to extract and calculate position and angle data.

For each trial, the main variable extracted was AROM. AROM was a measure of how far a segment can be moved against gravity. For a given segment, AROM was calculated relative to the adjacent, proximal segment, as the angular excursion through which the segment moved when it was the instructed one. For finger AROM, finger angular excursions were calculated from finger angle data 42, 43, a Euler angle calculation that represents the rotation at all three joints of each finger. The nine different segments have different anatomical ranges of motion. In order to examine how one segment may be affected and recover compared to another, raw AROM values were converted from degrees to percent of normal range of the unaffected side of comparable aged individuals with stroke during the same task 22, 42. Normalizing to this group eased the testing burden on the subjects in the present study.

Additional descriptive measures

Additional tests were conducted to provide a more thorough description of the sample (Table 1). To quantify upper extremity strength, maximal voluntary isometric contractions were measured using a hand-held dynamometer. Both flexion and extension of the shoulder, elbow, wrist, and index finger were measured one time in standard manual muscle test positions 44. Both affected and unaffected sides were tested. If the subject was unable to hold the manual muscle testing position, a score of zero was given for that muscle group. If the subject could maintain the test position, they were instructed to “hold the position, and don't let me move you”. All testing for each subject was performed by the same tester. Results were averaged across all joints on each side then expressed as a percentage of unaffected side. (Table 1, Composite Strength) 45. Affected side shoulder pain was assessed using a standard 11 point pain scale, where 0 = no pain (Table 1, Shoulder Pain). Joint position sense was evaluated on both sides at the index finger using standard clinical techniques where normal = correct on ≥ 3/5 trials (Table 1, Index finger joint position sense). Lastly, spasticity was evaluated on the affected side using the Modified Ashworth Scale at four joints: metacarpophalangeal joints of the hand, the wrist, the elbow, and the shoulder (Table 1, Modified Ashworth Scale).

Data analysis

SPSS version 13 was used for all statistical analyses and the criterion for statistical significance was set at p < 0.05. Distributions of variables were examined for normality using the Kolmogorov-Smirnov test. The six subtest scores from the Jebsen test were not normally distributed and were transformed for further statistical analyses. Each subtest score was transformed using the natural log function. All subsequent analyses using these six variables were done with the transformed data.

Paired t-tests were used to evaluate differences between time points for the descriptive measures. A repeated measures within subjects ANOVA with two factors: time and segment, was used to determine if normalized AROM at the nine upper extremity segments were similarly or differentially affected within and between time points. When significant effects were found, Bonferroni corrected post hoc t-tests were used to detect where differences existed.

Pearson product moment correlations were used to evaluate the relationships between AROM at all nine segments at 1 month and upper extremity function (score from principal components analysis) at 3 months. Based on our sample size, correlation coefficients with an absolute value greater than 0.34, were statistically significant. To determine the most parsimonious set of AROM variables that could explain the variance in upper extremity function scores without an a priori theory, we used stepwise multiple regression analysis. The dependent variable was the upper extremity function score obtained from the principal components analysis at 3 months. The independent variables were the 1 month AROM values of the nine upper extremity segments.


Thirty-three subjects (57 ± 10 years, range 31-77) with hemiparesis were tested within 1 month (18.6 ± 5.6 days, range 8-30) and at 3 months (98.3 ± 14.9 days, range 98-138) after stroke. Subject characteristics and lesion information are included in Table 1. Subjects had varying degrees of hemiparesis from nearly complete plegia to just barely detectable paresis, as indicated by the range in composite strength (Table 1). Composite strength improved from 1 to 3 months (p < 0.01). Subjects had minimal shoulder pain initially, and showed a trend toward increasing pain by 3 months (p = 0.067). Only six subjects had impaired joint position sense at the index finger. Spasticity was minimal at 1 month, but did increase in the shoulder (p < 0.05) and elbow (p < 0.01) by 3 months. Interestingly, subjects reported themselves to be an average of 51% and 57% recovered at the 1 and 3 month time points respectively, a difference that was not significant (p = 0.171).

Group AROM raw values, in degrees, are presented for each segment in Figure 1A. Average normalized AROM for each segment of the upper extremity ranged from 44% to 67% at 1 month, and from 72% to 99% at 3 months (Figure 1B). Using a repeated measures ANOVA, we found a significant main effect of time (p = 0.001) indicating that AROM increased between 1 and 3 months. We also found a significant main effect of segment (p = 0.007) indicating that differences existed between segments within each time point. Post hoc comparisons indicated that forearm AROM was greater than the index finger AROM (p = 0.004) at 1 month 22, and forearm AROM was greater than the shoulder (p = 0.006), elbow (p = 0.005), middle (p = 0.006) and index (p < 0.001) fingers, and thumb (p = 0.033) AROM at 3 months. We did not find a time by segment interaction (p = 0.575) indicating that all segments recovered to a similar degree over time.

Figure 1
A: Raw AROM group means ± SDs for affected side for each of the 9 segments at 1 month (■), and at 3 months (□). B: Percent of normal AROM group means ± SEs for each of the 9 segments at 1 month and 3 months. Raw values ...

Table 2 includes group mean scores for each upper extremity function test, and individual subject NIHSS-motor arm scores, upper extremity function scores for each test, and the composite upper extremity function scores, and group mean change scores between 1 and 3 months for each upper extremity function test. Following principal components analysis, upper extremity function scores were normally distributed with a mean of 0 (SD = 1, range = −1.81 – 1.23), where higher functioning was indicated by positive numbers and lower functioning was indicated by negative numbers.

Table 2
Individual subject scores for NIHSS, 6 upper extremity function tests, and composite upper extremity function score.

One month AROM values were strongly correlated with upper extremity function at 3 months (Table 3, Correlations) indicating that greater movement ability at 1 month was related to greater upper extremity function at 3 months. All correlations were significant at the p < 0.01 level. Overall, the correlation between shoulder AROM and upper extremity function was the largest (r = 0.81), and the correlation between forearm AROM and upper extremity function was the smallest (r = 0.67). Stepwise multiple regression analysis was used to determine if AROM taken within 1 month could predict upper extremity function at 3 months (Table 3, Variance Explained). Only the 1 month shoulder (R2 = 0.65) and middle finger (R2 = 0.06) AROM values entered the model. Thus, measurement of AROM of the shoulder and middle finger within 1 month after stroke predicted 71% of the variance in upper extremity function at 3 months (F2, 25 = 30.38, p < 0.001). No other variables we tested entered the prediction model.

Table 3
Correlations and R2 values


In our sample, stroke similarly affected the ability to move in all nine upper extremity segments at one month after injury 22. Here, we found that recovery of volitional movement, as measured by AROM, showed a trend toward being similar across segments. AROM measures at 1 month were strongly correlated with upper extremity function at 3 months. The most parsimonious model for using AROM at 1 month to predict upper extremity function at 3 months included only two segments, the shoulder and the middle finger. These two segments explained the majority (71%) of the variance in upper extremity function.

Our current findings build upon existing literature examining the prediction of upper extremity functional recovery following stroke 1, 17-19, 21, 46 by examining how AROM can predict upper extremity function at 3 months post stroke. Our finding that this simple measure was able to predict 71% of the variance in upper extremity function is important. The portion of variance in upper extremity function predicted by our model is much higher than that predicted by Katrak et al. (1998), where their models predicted less than 25% of the variance on a functional test using different independent variables. Our predicted variance in upper extremity function was more similar to that predicted by the models of Smania et al. (2007), where in three different models (three different functional test scores used as dependent variables), they were able to predict 42%, 75%, and 70% of the variance in upper extremity function at 90 days. A major strength of our model was that our dependent variable used to measure upper extremity function was a composite score derived from the principal components analysis on a battery of clinical tests. Using this methodology, we were able to sensitively quantify upper extremity function by avoiding ceiling and floor effects for any one test and by assessing a range of behaviors/movements across tests (e.g. fine dexterity with the 9 Hole Peg Test, functional capacity at home with the Stroke Impact Scale Hand Function subscale). Thus, we were able to avoid generating and interpreting multiple models where contributions from independent variables may not have been consistent 1. Another methodological difference is we did not make distinctions of what constituted better or worse AROM, rather, we used the measured value of AROM at each of the segments as our independent variables. We also investigated AROM of the entire upper extremity, not just proximally or distally1, 17, allowing for middle segments to contribute to the prediction of upper extremity function. Nonetheless, it is interesting that our results support those found by both Katrak et al. (1998), and Smania et al. (2007), and is remarkable that AROM measurement, a basic clinical skill taught to many healthcare professionals, can provide such a large amount of information about future upper extremity function.

AROM of the shoulder and the middle finger were the two variables that predicted upper extremity function in our model. Along with their strong correlations with upper extremity function, AROM values at all 9 segments were strongly correlated with each other at the 1 month time point 22. The shoulder and middle finger AROM variables that entered the stepwise model can therefore be interpreted as representative of AROM at the other segments as well. Because of the inter-correlations between shoulder AROM and AROM at other segments, the information contributed by shoulder AROM likely reflects the ability of the proximal arm to translate and orient the hand to functionally interact with objects in the environment. Similarly, information contributed by middle finger AROM likely reflects the ability of the distal segments to manipulate objects. These results suggest that measuring AROM at only two segments, shoulder flexion and middle finger flexion early after stroke, is useful for determining prognosis of upper extremity function at 3 months after stroke.

While the present study does not include data greater than 3 months post stroke, the question does arise as to whether the predictive ability of AROM would change at a more chronic phase post stroke. We found only one study1 in the literature that has attempted to investigate this question. Their data showed mixed results (59%, 72%, 64% of the variance) when predicting function out to 6 months post stroke, depending on the dependent variable used to measure upper extremity function1. These authors did not make conclusions based on how the variance changed over time, but by interpreting their results, it would appear that AROM continues to be a good predictor beyond 3 months. Also, we recently reported in a cross-sectional study of individuals with chronic stroke (21 months) that AROM explained 73% of the variance in upper extremity function using very similar methodology to the present study 42. Because the similar amount of explained variance in the present study and in both of the above referenced articles it would be reasonable to conclude that AROM would be a good predictor of upper extremity function at a more chronic phase post stroke.

A secondary purpose of our study was to examine how AROM at various upper extremity segments recovered in this time period. Our subjects recovered, on average, between 72 – 99% of normal AROM by 3 months. This amount of recovery may be a function of initial NIHSS – motor arm scores indicating that the subjects were only moderately affected. We recently and unexpectedly found that the nine segments of the upper extremity were similarly affected at one month after stroke 22. This finding is in conflict with common clinical perceptions 47 but is consistent with earlier, qualitative work 48. Here, our results indicate that the ability to move each segment showed a trend toward recovering to a similar degree as demonstrated by a significant main effect of time and no interactions between time and segment (Figure 1). Our longitudinal data therefore indicate that upper extremity paresis post stroke is not more severe distally than proximally at 1 month and that a proximal to distal gradient in movement deficits does not manifest over time. One explanation for this may be that the lesions disrupt similar proportions of proximal and distal inputs. A similar proportion of inputs could be disrupted following stroke because motor cortical territory for proximal and distal muscles is strongly overlapping 49 and as the axons descend through the subcortical structures, they are densely packed together 50. In our sample with heterogeneous lesions, damage to many locations within the motor system may result in a similar disruption to proximal and distal inputs.

Four limitations with this study are important to address when interpreting the results. First, the sample size was small and a portion of subjects was lost to attrition. Statistical analyses indicated that those subjects lost to attrition did not differ from the subjects who remained. Further studies are needed to confirm the predictive value of our model in a larger sample. Second, subjects in this sample had primarily motor deficits. Similar to other studies of motor function post stroke, our inclusion/exclusion restricted our sample by eliminating individuals who had severe aphasia, neglect, and/or who could not give informed consent. Our results may be less applicable to patients with other, substantial deficits beyond the motor domain. Third, it was surprising that normalized forearm AROM values at both time points (Figure 1B) were significantly greater than normalized AROM values of other segments. This result may be because the start position for forearm AROM was tested in a gravity eliminated position. We find this possibility unlikely however because of the aforementioned negligible moment arm, and the normal group was tested in the same manner. Another reason that this result might have arisen is that our control sample from which the normalizing values were taken 42, had smaller than expected forearm motion. In looking at the raw values (Figure 1A), it appears that the average change (recovery) in forearm AROM values is similar to the change in the other segments. Fourth, we did not collect information on factors such as motivation, depression, participation in rehabilitation, etc. It is possible that these data could have predicted additional variance in upper extremity function beyond that which AROM predicted.

In conclusion, we found that AROM measurements taken an average of three weeks post stroke can be used to predict upper extremity function at 3 months. AROM can be measured quickly at bedside or in the clinic with a goniometer by a variety of healthcare practitioners. Taking these inexpensive, simple measurements at only 2 segments provides a significant amount of information about eventual upper extremity function. More accurate prognosis about upper extremity function is important for determining appropriate rehabilitation strategies and discussing expectations of functional recovery with patients and their families. Future studies are needed to test the predictive value of our model in larger, more heterogeneous samples and at earlier time points than those evaluated here.


We thank Lucy Morris, MD, MPH, for her assistance with lesion data analysis, and Dustin Hardwick, DPT, for his assistance with data collection. Subject recruitment was facilitated by the Stroke Registry supported by the James S. McDonnell Foundation. Supported by National Institutes of Health, and the National Institute of Child Health and Human Development HD047669, HD007434, and Foundation for Physical Therapy Promotion of Doctoral Studies II Scholarship.


Publisher's Disclaimer: This is an un-copyedited author manuscript that was accepted for publication in Stroke, copyright The American Heart Association. This may not be duplicated or reproduced, other than for personal use or within the “Fair Use of Copyrighted Materials” (section 107, title 17, U.S. Code) without prior permission of the copyright owner, The American Heart Association. The final copyedited article, which is the version of record, can be found at Stroke. The American Heart Association disclaims any responsibility or liability for errors or omissions in this version of the manuscript or in any version derived from it by the National Institutes of Health or other parties.

Reference List

1. Smania N, Paolucci S, Tinazzi M, Borghero A, Manganotti P, Fiaschi A, Moretto G, Bovi P, Gambarin M. Active finger extension: a simple movement predicting recovery of arm function in patients with acute stroke. Stroke. 2007 March;38(3):1088–90. [PubMed]
2. Parker VM, Wade DT, Langton HR. Loss of arm function after stroke: measurement, frequency, and recovery. Int Rehabil Med. 1986;8(2):69–73. [PubMed]
3. Wade DT, Langton-Hewer R, Wood VA, Skilbeck CE, Ismail HM. The hemiplegic arm after stroke: measurement and recovery. J Neurol Neurosurg Psychiatry. 1983 June;46(6):521–4. [PMC free article] [PubMed]
4. Olsen TS. Arm and leg paresis as outcome predictors in stroke rehabilitation. Stroke. 1990 February;21(2):247–51. [PubMed]
5. Konig IR, Ziegler A, Bluhmki E, Hacke W, Bath PM, Sacco RL, Diener HC, Weimar C. Predicting long-term outcome after acute ischemic stroke: a simple index works in patients from controlled clinical trials. Stroke. 2008 June;39(6):1821–6. [PubMed]
6. Weimar C, Ziegler A, Konig IR, Diener HC. Predicting functional outcome and survival after acute ischemic stroke. Journal of Neurology. 2002 July;249(7):888–95. [PubMed]
7. Weimar C, Konig IR, Kraywinkel K, Ziegler A, Diener HC. Age and National Institutes of Health Stroke Scale Score within 6 hours after onset are accurate predictors of outcome after cerebral ischemia: development and external validation of prognostic models. Stroke. 2004 January;35(1):158–62. [PubMed]
8. Edwards DF, Hahn M, Baum C, Dromerick AW. The impact of mild stroke on meaningful activity and life satisfaction. J Stroke Cerebrovasc Dis. 2006 July;15(4):151–7. [PubMed]
9. Kwakkel G, van Dijk GM, Wagenaar RC. Accuracy of physical and occupational therapists' early predictions of recovery after severe middle cerebral artery stroke. Clinical Rehabilitation. 2000 February;14(1):28–41. [PubMed]
10. Woldag H, Gerhold LL, de GM, Wohlfart K, Wagner A, Hummelsheim H. Early prediction of functional outcome after stroke. Brain Injury. 2006 September;20(10):1047–52. [PubMed]
11. Hsieh CL, Sheu CF, Hsueh IP, Wang CH. Trunk control as an early predictor of comprehensive activities of daily living function in stroke patients. Stroke. 2002 November;33(11):2626–30. [PubMed]
12. Chester CS, McLaren CE. Somatosensory evoked response and recovery from stroke. Archives of Physical Medicine and Rehabilitation. 1989 July;70(7):520–5. [PubMed]
13. Loewen SC, Anderson BA. Predictors of stroke outcome using objective measurement scales. Stroke. 1990 January;21(1):78–81. [PubMed]
14. Studenski SA, Wallace D, Duncan PW, Rymer M, Lai SM. Predicting stroke recovery: three- and six-month rates of patient-centered functional outcomes based on the orpington prognostic scale. J Am Geriatr Soc. 2001 March;49(3):308–12. [PubMed]
15. Johnston KC, Connors AF, Jr., Wagner DP, Knaus WA, Wang X, Haley EC., Jr A predictive risk model for outcomes of ischemic stroke. Stroke. 2000 February;31(2):448–55. [PubMed]
16. Tilling K, Sterne JA, Rudd AG, Glass TA, Wityk RJ, Wolfe CD. A new method for predicting recovery after stroke. Stroke. 2001 December 1;32(12):2867–73. [PubMed]
17. Katrak P, Bowring G, Conroy P, Chilvers M, Poulos R, McNeil D. Predicting upper limb recovery after stroke: the place of early shoulder and hand movement. Archives of Physical Medicine and Rehabilitation. 1998 July;79(7):758–61. [PubMed]
18. De WW, Lincoln NB, Harrison MA. Prediction of arm and hand function recovery in stroke patients. International Journal of Rehabilitation Research. 1987;10(4 Suppl 5):110–2. [PubMed]
19. Kwakkel G, Kollen BJ, van der GJ, Prevo AJ. Probability of regaining dexterity in the flaccid upper limb: impact of severity of paresis and time since onset in acute stroke. Stroke. 2003 September;34(9):2181–6. [PubMed]
20. Feys H, De WW, Nuyens G, van de WA, Selz B, Kiekens C. Predicting motor recovery of the upper limb after stroke rehabilitation: value of a clinical examination. Physiother Res Int. 2000;5(1):1–18. [PubMed]
21. Feys H, Hetebrij J, Wilms G, Dom R, De WW. Predicting arm recovery following stroke: value of site of lesion. Acta Neurolo Scand. 2000 December;102(6):371–7. [PubMed]
22. Beebe JA, Lang CE. Absence of a proximal to distal gradient of motor deficits in the upper extremity early after stroke. Clin Neurophysiol. 2008;119(9):2074–85. [PMC free article] [PubMed]
23. Duncan PW, Lai SM, Keighley J. Defining post-stroke recovery: implications for design and interpretation of drug trials. Neuropharmacology. 2000 March 3;39(5):835–41. [PubMed]
24. Schmidt RT, Toews JV. Grip strength as measured by the Jamar dynamometer. Arch Phys Med Rehabil. 1970 June;51(6):321–7. [PubMed]
25. Boissy P, Bourbonnais D, Carlotti MM, Gravel D, Arsenault BA. Maximal grip force in chronic stroke subjects and its relationship to global upper extremity function. Clin Rehabil. 1999 August;13(4):354–62. [PubMed]
26. Mathiowetz V, Kashman N, Volland G, Weber K, Dowe M, Rogers S. Grip and pinch strength: normative data for adults. Arch Phys Med Rehabil. 1985 February;66(2):69–74. [PubMed]
27. Mathiowetz V, Weber K, Volland G, Kashman N. Reliability and validity of grip and pinch strength evaluations. J Hand Surg. 1984 March;9(2):222–6. [PubMed]
28. Lyle RC. A performance test for assessment of upper limb function in physical rehabilitation treatment and research. Int J Rehabil Res. 1981;4(4):483–92. [PubMed]
29. Hsieh CL, Hsueh IP, Chiang FM, Lin PH. Inter-rater reliability and validity of the action research arm test in stroke patients. Age Ageing. 1998 March;27(2):107–13. [PubMed]
30. Van der Lee JH, De G,V, Beckerman H, Wagenaar RC, Lankhorst GJ, Bouter LM. The intra- and interrater reliability of the action research arm test: a practical test of upper extremity function in patients with stroke. Arch Phys Med Rehabil. 2001 January;82(1):14–9. [PubMed]
31. Van der Lee JH, Beckerman H, Lankhorst GJ, Bouter LM. The responsiveness of the Action Research Arm test and the Fugl-Meyer Assessment scale in chronic stroke patients. J Rehabil Med. 2001 March;33(3):110–3. [PubMed]
32. Lang CE, Wagner JM, Dromerick AW, Edwards DF. Measurement of upper-extremity function early after stroke: properties of the action research arm test. Arch Phys Med Rehabil. 2006 December;87(12):1605–10. [PubMed]
33. Lang CE, Edwards DF, Birkenmeier RL, Dromerick AW. Estimating minimal clinically important differences of upper extremity measures early after stroke. Arch Phys Med Rehabil. 2008 In Press. [PMC free article] [PubMed]
34. Jebsen RH, Taylor N, Trieschmann RB, Trotter MJ, Howard LA. An objective and standardized test of hand function. Arch Phys Med Rehabil. 1969 June;50(6):311–9. [PubMed]
35. Mathiowetz V, Weber K, Kashman N, Volland G. Adult norms for the nine hole peg test of finger dexterity. OTJR. 1985;5(1):24–38.
36. Duncan PW, Wallace D, Studenski S, Lai SM, Johnson D. Conceptualization of a new stroke-specific outcome measure: the stroke impact scale. Top Stroke Rehabil. 2001;8(2):19–33. [PubMed]
37. Duncan PW, Wallace D, Lai SM, Johnson D, Embretson S, Laster LJ. The stroke impact scale version 2.0. Evaluation of reliability, validity, and sensitivity to change. Stroke. 1999 October;30(10):2131–40. [PubMed]
38. Duncan P, Reker D, Kwon S, Lai SM, Studenski S, Perera S, Alfrey C, Marquez J. Measuring stroke impact with the stroke impact scale: telephone versus mail administration in veterans with stroke. Med Care. 2005 May;43(5):507–15. [PubMed]
39. Hummel F, Celnik P, Giraux P, Floel A, Wu WH, Gerloff C, Cohen LG. Effects of non-invasive cortical stimulation on skilled motor function in chronic stroke. Brain. 2005 March;128(Pt 3):490–9. [PubMed]
40. Ward NS, Brown MM, Thompson AJ, Frackowiak RS. Neural correlates of motor recovery after stroke: a longitudinal fMRI study. Brain. 2003 November;126(Pt 11):2476–96. [PMC free article] [PubMed]
41. Ward NS, Brown MM, Thompson AJ, Frackowiak RS. Neural correlates of outcome after stroke: a cross-sectional fMRI study. Brain. 2003 June;126(Pt 6):1430–48. [PMC free article] [PubMed]
42. Lang CE, Beebe JA. Relating movement control at 9 upper extremity segments to loss of hand function in people with chronic hemiparesis. Neurorehabil Neural Repair. 2007 May;21(3):279–91. [PubMed]
43. Theverapperuma LS, Hendrix CM, Mason CR, Ebner TJ. Finger movements during reach-to-grasp in the monkey: amplitude scaling of a temporal synergy. Exp Brain Res. 2006 March;169(4):433–48. [PubMed]
44. Andrews AW, Thomas MW, Bohannon RW. Normative values for isometric muscle force measurements obtained with hand-held dynamometers. Phys Ther. 1996 March;76(3):248–59. [PubMed]
45. Wagner JM, Lang CE, Sahrmann SA, Edwards DF, Dromerick AW. Sensorimotor impairments and reaching performance in subjects with poststroke hemiparesis during the first few months of recovery. Phys Ther. 2007 June;87(6):751–65. [PubMed]
46. Feys H, Van HJ, Bruyninckx F, Mercelis R, De WW. Value of somatosensory and motor evoked potentials in predicting arm recovery after a stroke. Journal of Neurology, Neurosurgery, and Psychiatry. 2000 March;68(3):323–31. [PMC free article] [PubMed]
47. Colebatch JG, Gandevia SC. The distribution of muscular weakness in upper motor neuron lesions affecting the arm. Brain. 1989 June;112(Pt 3):749–63. [PubMed]
48. Bard G, Hirschberg GG. Recovery of voluntary motionin upper extremity following hemiplegia. Arch Phys Med Rehabil. 1965 August;46:567–72. [PubMed]
49. Park MC, Belhaj-Saif A, Gordon M, Cheney PD. Consistent features in the forelimb representation of primary motor cortex in rhesus macaques. J Neurosci. 2001 April 15;21(8):2784–92. [PubMed]
50. Morecraft RJ, McNeal DW, Stilwell-Morecraft KS, Dvanajscak Z, Ge J, Schneider P. Localization of arm representation in the cerebral peduncle of the non-human primate. J Comp Neurol. 2007 September 10;504(2):149–67. [PubMed]