The proprioceptive senses, including static position sense and movement sense or kinesthesia,1 are critical for accurate movement,2,3 but are often impaired following stroke.2,4–9 Deficits in hand proprioception are associated with functional impairment,10–12 including poor motor recovery13–15 and reduced ability to complete tasks of daily living.2,16,17 Clinicians often need to assess proprioception to determine the extent of neurological injury, manage therapy plans, and evaluate progress.18,19 The detection of proprioception deficits in individuals after stroke has important clinical ramifications considering rehabilitation interventions are based on accurate identification of deficits19 and the ability to demonstrate rehabilitation progress. For example, proprioceptive deficits detected immediately after stroke can help predict arm function 1 year later,20 which may influence rehabilitation choices. Failure to detect proprioception deficits may leave a substantial number of patients who require treatment not receiving it, or not receiving appropriate treatments and thereby wasting treatment time.19 Current clinical methods, however, are too subjective and poorly standardized to detect subtle deficits or changes in proprioception.21,22
Current options for clinical testing of proprioception deficits following stroke are quite limited because there is no single standardized measure that is generally accepted.22,23 Options include the proprioception subsection of the Fugl-Meyer (F-M),24 passive movement direction discrimination (PMDD), and position matching, all of which entail subjective judgments by the clinician and poor standardization of the testing procedure. Even within standardized clinical assessments such as the Rivermead Assessment of Somatosensory Performance,23 proprioception is determined subjectively. The proprioception subsection of the F-M is arguably the most common clinical test, and could be considered the de facto gold standard; however, the sensory scales of the F-M perform poorly in the stroke population, with Lin et al25 reporting a significant ceiling effect and low to moderate validity and responsiveness. This and other subjective tests continue to be implemented clinically,26 but more objective tests have been developed in the research setting. Carey et al19 used an apparatus of protractors and forearm splints to conduct a wrist angle matching task,19 and more recent applications have made use of sophisticated robotics to generate quantifiable metrics for proprioception and kinesthesia.9,27–30 For example, Goble et al31 used a bimanual robotic system to accurately execute an elbow angle matching task. Semrau et al32 assessed upper limb kinesthesia with psychometric techniques implemented in a robotic system as well. These measures provide more sensitive and objective data regarding proprioception; indeed, motor control research has applied these methods to understand somatosensory contributions to movement in healthy subjects.33–36 Unfortunately, robotic methods have limited clinical utility due to costs associated with the equipment, limited portability, and the considerable training required.
As an alternative to both subjective clinical tests and costly robotic assessments of proprioception, we developed a tool that requires only a computer tablet and a hand positioning stand.37 Subjects make a series of judgments about the location of their unseen stationary index finger in relation to elements displayed on the tablet.37 This allows us to determine proprioceptive bias (perceptual boundary) at the metacarpophalangeal (MP) joint of the index finger in the plane of adduction/abduction, chosen because of its importance for fine motor skill such as a key grip/pinch of small objects between the thumb and the index finger.
In healthy adults, the tablet-based measure had better test-retest reliability, interrater reliability, and construct validity than 2 common clinical methods, PMDD and matching, which we customized to reduce subjectivity.37 While the tablet-based measure performed well in healthy adults, its utility in stroke survivors is unknown. After a unilateral cortical or subcortical stroke, proprioception in the contralateral (herein referred to as the “affected”) hand may be more impaired than in the ipsilateral (herein referred to as the “unaffected”) hand. Here, we asked whether the tablet-based measure could detect proprioceptive differences between the hands in chronic stroke survivors that were more extreme than between the hands of healthy age-matched controls. We compared the tablet-based measure with the custom PMDD37 and with a proprioception subsection of the F-M24 modified to assess the index finger along with the thumb. Because even the unaffected hand often experiences deficits after stroke,38 we also compared the affected hand to the control group directly. Finally, we compared the tablet-based measure and the PMDD to deficits in primary tactile function and manual dexterity.39
A total of 32 subjects completed the study, 16 with stroke and 16 without. All reported they were free of visual impairments and severe or uncontrolled cardiovascular, pulmonary, renal, neurological, or psychiatric conditions. Subjects were screened for serious cognitive deficits with the Mini-Mental State Examination (MMSE).40 Participants with stroke were screened for hemispatial neglect with the star cancellation task.41 Procedures were approved by the institutional review board of Indiana University Bloomington. All subjects gave written informed consent.
Subjects with stroke completed the MMSE, star cancellation test, and Medical Research Council (MRC) muscle testing42 of an index finger muscle, first dorsal interosseous, in that order. Proprioception testing was performed next, with the custom tablet task,37 PMDD,37 and manual assessment based on the F-M24 administered in random order. We then obtained participants' fine-touch (Baseline Evaluation Instruments, White Plains, New York) and 2-point discrimination thresholds (Baseline Evaluation Instruments, White Plains, New York) and performed the Box and Block test of manual dexterity39 (BBT; Model 7531, Sammons Preston, Bolingbrook, IL), followed by the F-M upper extremity motor section.24 Most subjects completed the session in approximately 90 minutes.
Age-matched control subjects completed the MMSE and Edinburgh handedness inventory,43 then the proprioception tablet task and PMDD in random order, and finally the Purdue Pegboard Test (PPT; Lafayette Instrument, Lafayette, IN) of manual dexterity.44 The session required approximately 45 minutes to complete.
With eyes closed, the subject's hand was positioned in the stand by the experimenter (Figure 1A). Subjects were asked to press down firmly with their index finger and then relax, to control for muscle thixotropy.1 After the experimenter positioned the tablet (Samsung Galaxy Tab Pro 12.2, Suwon, South Korea) over the hand layer, subjects were asked to open their eyes. Subjects were familiarized with the task by a training sequence of the custom application. Subjects were instructed not to move their hand during the test.
For each trial, the tablet displayed 2 colored regions, with 1 end of the dividing line fixed over the MP joint (Figure 1Aii). Subjects reported which color they felt was over the center of their index fingertip. The angle of the dividing line changed every trial in an adaptive staircase algorithm based on Parameter Estimation by Sequential Testing,46 which has been implemented in robotic measures of proprioception.33 The test consisted of 2 staircases, with line angle beginning at 30° left and right of true index finger angle. Initial step size was 10°. In a staircase beginning 30° to the left of true finger position, most subjects perceive that the dividing line is left of their index finger and choose the color to the right of the line. The line would then move 10° to the right, and many subjects would still choose the color on the right. However, as the line approaches true finger position, subjects become more uncertain and eventually choose the color to the left of the line. Whenever the subject's choice of color reversed, the line reversed direction and step size decreased by half to yield more measurements near the subject's perceptual boundary (the angle at which the subject is equally likely to choose either color). Each staircase terminated after 4 reversals. For most subjects, this resulted in approximately 10 trials per staircase. The complete test, including positioning the subject in the apparatus and training, took 2 to 3 minutes.
Each test resulted in 8 reversal angles (4 per staircase). We averaged these 8 angles to obtain an estimate of perceptual boundary, analogous to spatial bias in perceived joint angle. We previously found this corresponds closely to the perceptual boundary determined by fitting a logistic model, and that 2 staircases yield data nearly as reliable as 6 staircases.37 Because we have no evidence to suggest a rightward bias in proprioception is better or worse than a leftward bias, group analysis used the absolute value.
For detailed testing protocol, see Supplemental Digital Content 2 (http://links.lww.com/JNPT/A264).
PMDD was tested in both groups to assess dynamic proprioception at the MP joint of the index finger. Subject positioning was the same as the tablet task, except the eyes were closed throughout and the hand was pronated on a paper template (Figure 1B). Once correctly positioned, subjects were asked to press down firmly with their index finger and then relax, to control for muscle thixotropy.1 The initial passive movement amplitude tested was 5°. Each amplitude was tested 6 times, with 3 rightward and 3 leftward movements in random order. If a subject was correct on all 6 movements of 5° amplitude, the movement amplitude was reduced to 2.5°. If a subject made an error at 5°, amplitude increased to 10°. The procedure took approximately 5 minutes per hand.
PMDD threshold was defined as the smallest of 1.25°, 2.5°, 5°, 10°, or 15° movement amplitudes that the subject responded correctly for all 6 movements. PMDD data were right-censored; that is, many subjects made errors even on 15° movements, especially on the affected hand in the stroke group, but larger movements were not biomechanically feasible. PMDD threshold was recorded as “15+” (Table 1) for these subjects, but a value of 15° was substituted for group analyses.
Manual Assessment (Stroke Group only)
Following the standard F-M proprioception subsection procedure, we tested thumb flexion-extension at the interphalangeal joint. In addition, we tested right and left movements of the index finger (abduction/adduction at the MP joint) to be consistent with the PMDD and tablet tasks, using the same sequence of steps as the thumb test. Standard F-M scoring of 0, 1, or 2 was applied for both versions.24
Wilcoxon rank sum tests were performed to test whether median between-hand differences in the stroke group were mo-re extreme than in the control group on the tablet-based measure or PMDD. The differences analyzed were affected minus unaffected hand for the stroke group, and nondominant minus dominant hand for the control group. Wilcoxon signed rank tests were performed to test whether median 2-point discrimination, fine-touch, or BBT scores were worse in the affected than in the unaffected hands in the stroke group, whether median PPT scores differed between dominant and nondominant hands in the control group, and whether the tablet-based measure and the PMDD differed between the stroke group's affected or unaffected hand and the control group. For the latter stroke-control comparisons, we used the average of control subjects' dominant and nondominant hands, as we found no between-hand difference and conflicting literature as to whether such a difference should be expected.47–54
For both the tablet-based method and the PMDD, we computed sensitivity, specificity, and positive and negative predictive values, treating the manual assessment based on the F-M proprioception subsection as the gold standard. Because 2 of the participants with stroke lacked the manual assessment, only 14 of the participants with stroke were included in the sensitivity analyses. We identified participants with stroke who had proprioceptive deficits (“positives”) in 2 ways. First, for participants with stroke we compared tablet-based or PMDD value in their affected hand versus their unaffected hand: if the affected hand had a worse (higher) value than the unaffected hand, that subject was considered a positive. Second, in participants with stroke we compared the tablet-based or PMDD measurement of each affected hand with the mean control group value. Participants with stroke having values that exceeded the control group mean plus 95% confidence interval were considered positives.23,27,32,55
Spearman rank correlations were computed to assess the relationship between values obtained from the tablet or PMDD and each of 2-point discrimination threshold, fine-touch threshold, and BBT score in the stroke group. Affected hand values were divided by unaffected hand values to normalize for individual variation that may be unrelated to the stroke. Bonferroni corrections were performed to compensate for multiple comparisons. All hypothesis tests were performed 2-tailed with α of 0.05.
The 16 participants with stroke (8 female, age 64.1 ± 15.2 years) had experienced unilateral ischemic or hemorrhagic cortical or subcortical stroke with motor deficits at least 6 months earlier (Table 1).
Average time since stroke was 7.4 ± 7.6 years (mean ± standard deviation). Motor function of the impaired hand averaged an MRC42 grade of 3.6 ± 2.0, indicating moderate impairment. Subjects S-01 and S-02 completed the experiment before the manual assessment of proprioception was added, so analysis of this parameter excluded these subjects. Subjects S-05 and S-15 lack data for the motor assessments (BBT and F-M upper limb) because of limited shoulder mobility due to arthritis and history of a torn rotator cuff, respectively. Analysis of the motor variables therefore excluded these 2 subjects. The 16 age-matched controls had no history of stroke (9 female, age 65.7 ± 14.4 years); each control subject was within 5 years of the age of a participant with stroke (Table 2).
Comparing Proprioception Across Hands
Stroke group: Proprioceptive bias in the affected hand was worse than in the unaffected hand (Figure 2) for 12/16 of the participants with stroke. Median bias was 15.5° and 9.2° in the affected and unaffected hands, respectively.
Control group: Proprioceptive bias on the tablet-based measure was worse in the nondominant hand compared with the dominant hand for 5 of 16 subjects. At the group level, median bias was 9.2° and 10.7° on the tablet-based measure in the nondominant and dominant hands, respectively. The between-hand difference was greater in the stroke group than in the control group (W = 203.5, Z = −2.26, P = 0.024). This suggests that between-hand differences are greater in the stroke group (median difference 6.3°) than in controls (1.5°) (Figure 3A).
Stroke group: The PMDD showed a higher threshold (worse proprioception) in the affected than in the unaffected hand for 9 of 16 participants with stroke. All 9 were among those found by the tablet-based measure to have worse proprioception in their affected hand. Because we dealt with the PMDD ceiling effect by substituting 15 for “15+”, group PMDD values likely underestimate true threshold.56 Median values were 15° and 5° in the affected and unaffected hands, respectively.
Control group: The PMDD showed worse proprioception in the nondominant hand for 3 of the 16 control subjects; median PMDD threshold was 5° for each hand. The between-hand difference in PMDD was greater in the stroke group than in the control group (W = 206, Z = −2.20, P = 0.028), suggesting that between-hand differences are greater in the stroke group (median difference of 10°) than in controls (0°) (Figure 3B).
Manual Assessment of Proprioception (Participants With Stroke Only)
The manual assessment showed worse proprioception on the affected side in only 2 of 14 participants with stroke at the thumb, and in 4 of 14 participants with stroke at the index finger. Manual assessment score did not differ significantly across hands for either the thumb (W = 3, Z = 1.41, P = 0.31) or index finger (W = 10, Z = 2.0, P = 0.09). If we consider the manual assessment to be the de facto gold standard measure of proprioception, and define tablet and PMDD deficits by comparing the affected to the unaffected hand, then sensitivity of both the tablet and the PMDD is 0.75, with 7 and 4 false positives, respectively (Table 3A).
Comparing Proprioception in the Affected Hand With the Control Group
Comparing the affected hand of the 16 participants with stroke to the control group, we found that 9 of 16 participants with stroke had impaired proprioception according to the tablet test, and 12 had impaired proprioception according to the PMDD. Eight of the 9 impairments identified with the tablet were also identified by the PMDD. Proprioception was significantly worse in the stroke group's affected hand compared with the control group, according to both the tablet (W = 209, Z = −2.05, P = 0.040) and the PMDD (W = 207.5, Z = −2.15, P = 0.032). If we consider manual assessment the de facto gold standard measure of proprioception, and define tablet and PMDD deficits by comparing the affected hand to controls, then sensitivity of the tablet is 0.75 with 5 false positives, and sensitivity of the PMDD is 1 with 6 false positives (Table 3B).
Comparing the unaffected hand of the 16 participants with stroke with the control group, we found that 4 of 16 of the participants with stroke had impaired proprioception according to the tablet, and 5 of 16 had impaired proprioception according to the PMDD. However, as a group, the unaffected hands of the participants with stroke were not significantly different than controls for either the tablet (W = 257, Z = −0.25, P = 0.81) or PMDD (W = 255.5, Z = −0.31, P = 0.76). This suggests that some individuals in the stroke group had proprioceptive deficits in their unaffected hand, but the group overall did not.
Primary Tactile and Manual Dexterity Measures
Median 2-point discrimination threshold in the stroke group was 7 and 2.5 mm in the affected and unaffected hands, representing a significant difference (W = 6, Z = −2.59, P = 0.01; Figure 3Bi). Median fine-touch threshold did not differ across hands (W = 9, Z = −1.26, P = 0.21); median value was 3.61 (0.4g of force) in both (Figure 3Bii). Participants with stroke achieved a median BBT score of 50.5 and 23.5 in the unaffected and affected hands, respectively (Figure 3Ci), representing a significant difference (W = 105, Z = 3.30, P < 0.001). Control subjects achieved a median PPT score of 12.5 and 13 in the nondominant and dominant hands, respectively (Figure 3Cii), representing a nonsignificant difference (W = 96.5, Z = 1.48, P = 0.14).
For participants with stroke, between-hand percent difference in tablet bias was positively correlated with 2-point discrimination threshold (r = 0.55, P = 0.04; Figure 4Ai). However, percent differences in tablet bias were not significantly correlated with either fine-touch threshold or BBT (r = 0.29, P > 0.1; r = −0.37, P > 0.1, respectively; Figure 4Aii-iii). Between-hand percent differences in PMDD threshold were positively correlated with 2-point discrimination (r = 0.68, P = 0.006; Figure 4Bi) and fine touch (r = 0.88, P < 0.001; Figure 4Bii), but not with BBT (r = −0.038, P > 0.1; Figure 4Biii).
Computer tablets are increasingly being used in rehabilitation.57 In this study we compared 3 proprioception measures in chronic stroke survivors. Only the tablet and the PMDD found median differences between the hands relative to controls, although the PMDD had a ceiling effect. Whether comparing the affected hand to the unaffected hand or to controls, the tablet and the PMDD identified more proprioception deficits in participants with stroke than the manual assessment did. PMDD and tablet measures were each correlated with primary tactile sensation, but not with manual dexterity.
Detecting Proprioceptive Deficits
Assessment is complicated by proprioception's multiple submodalities (eg, static position sense, passive motion sense, and sense of effort).1 The PMDD and the proprioceptive subsection of the F-M measure passive movement sense; matching tests static position sense after active movement, or in some cases subjects are asked to actively match a passive movement.58 Common clinical versions of these tests, such as the manual assessment performed here, are known to be coarse, subjective, and poorly standardized22,59,60; they fail to control for muscle thixotropy, the contraction history of the muscle, or movement speed and amplitude, all of which affect proprioception.1,22,61
Proprioceptive assessments that require movement may be confounded by pain,62 spasticity, and in the case of tests requiring active movement, motor deficits. For example, an individual with pain may tense up with movement, providing extra stimulation to spindles and leading to overestimation of proprioceptive function. The individual may have less pain on a return visit, yielding a more accurate proprioceptive estimate but erroneously indicating a worsening of proprioception since the first visit. Tests involving movement also carry the risk of a ceiling effect, as the PMDD showed in the present study. In an individual who makes errors at the largest feasible movement amplitude, slight improvements due to rehabilitation cannot be detected. For these reasons, a static test of proprioception may be a useful supplement or replacement for movement-based tests. The brain continues to integrate background spindle activity, joint receptors, and skin stretch input even when the body is stationary1; indeed, static proprioception is critical for accurate motor control, as it allows the brain to estimate the limb's starting position for motor planning.1 Animal research indicates static and dynamic position information depends on the same neural networks.63,64
Between-hand differences in the stroke group relative to controls were detected by the tablet and the PMDD, but not the manual assessment. The tablet and the PMDD both identified more participants with stroke as having an impairment relative to the unaffected hand than the manual assessment did. Consistent with the literature,38 some participants with stroke had proprioceptive deficits in their unaffected hand according to both the tablet and the PMDD, but not the manual assessment. This may have caused the between-hand approach to underestimate the true number of participants with stroke with proprioceptive deficits, although deficits in the affected hand can generally be expected to be more severe than in the unaffected hand. The stroke-control comparison supplements the between-hand analysis. Both the tablet and the PMDD detected significant differences between the affected hand and the control group, and both identified more deficits in participants with stroke than the manual assessment did. However, the stroke-control comparison also has limitations. Namely, neither the tablet nor PMDD has large normative data sets for comparison, and using a small age-matched control group to define deficits must be interpreted with caution.
These results, combined with the sensitivity analysis, suggest the tablet has greater sensitivity to detect proprioception deficits compared with the manual assessment. Although these initially appear as false-positives, many participants who showed deficits with the tablet or the PMDD scored full points in both hands on the manual assessment, both the standard F-M subsection version at the thumb, and the alternate version at the index finger. This suggests manual assessment lacks sensitivity to detect more subtle proprioceptive deficits even though it is often considered the “gold standard” clinical assessment. As a result, in the clinical setting, manual assessment may not detect subtle proprioceptive deficits or changes due to rehabilitation. These findings are consistent with Carey et al,19 who demonstrated the lack of abnormal outcomes on clinical measures compared with newer standardized tests. Given the increased demands to evaluate progress with rehabilitation interventions, a more quantifiable measure of proprioception including the tablet measure may overcome these current clinical challenges.
The tablet offers an advantage over the manual assessment or other PMDD tests in terms of clinician training time. To use the tablet test, the clinician only needs to know how to place the hand on the stand and run the app. The test itself is automated from that point, which likely contributed to the greater reliability we previously found relative to the PMDD or matching.37 In contrast, to use the PMDD or manual assessment correctly, the clinician needs skill at grasping the thumb or finger without giving pressure cues, and moving the thumb or finger at a slow but consistent speed, all while following a random sequence of movement directions and recording subject response.
Relationship of Proprioception to Primary Tactile and Motor Assessments
Two-point discrimination and manual dexterity in the stroke group showed clear impairments in the affected hand, while fine-touch threshold did not. The PMDD and tablet measurements were both correlated with 2-point discrimination, but only PMDD was related to fine-touch threshold. The tablet impairment was relatively clustered, which may have limited our ability to detect correlations. Future studies will need to examine tablet and PMDD in larger samples to determine how strongly these measures are related to tactile assessments.
F-M upper limb scores ranged from 8 to 64 in the stroke group, suggesting considerable variation in motor impairment. The lack of correlation between proprioception and BBT was not necessarily surprising. This is consistent with several previous studies that found no relationship between proprioceptive and motor deficits after stroke.10,65
To simplify comparisons, both the tablet and PMDD measurements were applied with the hand pronated and fingers slightly spread. Measurements in these conditions may not generalize to other postures1 or to extreme finger angles.66 Also unknown is how proprioceptive deficits measured at the index finger would compare to other joints such as wrist and elbow. While the hand-positioning stand could be modified to apply the tablet test at other finger joints and even the wrist, the need for the participant to gaze directly at the tablet when placed over the joint being tested precludes the use of this method at the shoulder or elbow. However, this limitation does not take away from the clinical utility of the test when manual assessment is insufficiently sensitive or functional deficits are most noticeable in the hand. In other words, if an individual has such severe proprioceptive deficits that the shoulder and elbow are substantially affected, then manual assessment is likely sufficient to detect the problem. The greatest utility of the tablet test is in cases where the proprioceptive deficit is more subtle and could be overlooked, which would result in the individual not receiving needed treatment.
While the absence of movement in the tablet test clearly offers advantages, the dependence on vision creates a limitation. The test would not be valid for subjects with visual impairments or neglect. Subjects were not tested for color vision; some subjects called the yellow color green, or the pink color purple, but yellow and green were never paired with each other, nor were pink and purple, so there was no potential for the experimenter to press the wrong button. It is possible that slight luminance differences between the colors, or differences in subjects' color perception, could make one color easier to see than the other of a given pair, which might influence the subject's choice. However, this influence would be limited to increasing noise, not proprioceptive bias, as the placement of each color in the left versus right position was randomized.
Both the tablet and the custom PMDD performed better than manual assessment in detecting proprioceptive deficits. The PMDD may be useful when the deficit is mild or assessment of dynamic proprioception is desired. The tablet, lacking the PMDD's ceiling effect, could be useful with patients having any degree of severity of proprioceptive impairment, and may be preferable if testing or clinician training time needs to be minimized, or pain or spasticity is present.
1. Proske U, Gandevia SC. The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. Physiol Rev. 2012;92(4):1651–1697. doi:10.1152/physrev.00048.2011.
2. Smith DL, Akhtar AJ, Garraway WM. Proprioception and spatial neglect after stroke. Age Ageing. 1983;12(1):63–69.
3. Reding MJ, Potes E. Rehabilitation outcome following initial unilateral hemispheric stroke. Life table analysis approach. Stroke. 1988;19(11):1354–1358.
4. Tyson SF, Hanley M, Chillala J, Selley AB, Tallis RC. Sensory loss in hospital-admitted people with stroke: characteristics, associated factors, and relationship with function. Neurorehabil Neural Repair. 2008;22(2):166–172. doi:10.1177/1545968307305523.
5. Twitchell TE. The restoration of motor function following hemiplegia in man. Brain J Neurol. 1951;74(4):443–480.
6. Rand D. Proprioception deficits in chronic stroke—upper extremity function and daily living. PLoS One. 2018;13(3):e0195043. doi:10.1371/journal.pone.0195043.
7. Connell L, Lincoln N, Radford K. Somatosensory
impairment after stroke: frequency of different deficits and their recovery. Clin Rehabil. 2008;22(8):758–767. doi:10.1177/0269215508090674.
8. Carey LM. Somatosensory
loss after stroke. Crit Rev Phys Rehabil Med. 1995;7(1). doi:10.1615/CritRevPhysRehabilMed.v7.i1.40.
9. Dukelow SP, Herter TM, Moore KD, et al Quantitative assessment of limb position sense following stroke. Neurorehabil Neural Repair. 2010;24(2):178–187. doi:10.1177/1545968309345267.
10. Dukelow SP, Herter TM, Bagg SD, Scott SH. The independence of deficits in position sense and visually guided reaching following stroke. J Neuroengineering Rehabil. 2012;9(1):72. doi:10.1186/1743-0003-9-72.
11. Stern PH, McDowell F, Miller JM, Robinson M. Factors influencing stroke rehabilitation. Stroke. 1971;2(3):213–218.
12. Prescott RJ, Garraway WM, Akhtar AJ. Predicting functional outcome following acute stroke using a standard clinical examination. Stroke. 1982;13(5):641–647.
13. Campfens SF, Zandvliet SB, Meskers CGM, Schouten AC, van Putten MJAM, van der Kooij H. Poor motor function is associated with reduced sensory processing after stroke. Exp Brain Res. 2015;233(4):1339–1349. doi:10.1007/s00221-015-4206-z.
14. Vidoni ED, Boyd LA. Preserved motor learning after stroke is related to the degree of proprioceptive deficit. Behav Brain Funct BBF. 2009;5:36. doi:10.1186/1744-9081-5-36.
15. Wade DT, Langton-Hewer R, Wood VA, Skilbeck CE, Ismail HM. The hemiplegic arm after stroke: measurement and recovery. J Neurol Neurosurg Psychiatry. 1983;46(6):521–524.
16. Reding MJ. A model stroke classification scheme and its use in outcome research. Stroke. 1990;21(9 suppl):II35–II37.
17. Gottlieb D, Kipnis M, Sister E, Medvedev V, Brill S, Vardi Y. Classification of stroke rehabilitation patients with a simple impairment scale. J Neurol Rehabil. 1997;11(4):239–243. doi:10.1177/154596839701100406.
18. Li K, Wu Y. Clinical evaluation of motion and position sense in the upper extremities of the elderly using motion analysis system. Clin Interv Aging. 2014;9:1123–1131. doi:10.2147/CIA.S62037.
19. Carey LM, Matyas TA, Oke LE. Evaluation of impaired fingertip texture discrimination and wrist position sense in patients affected by stroke: comparison of clinical and new quantitative measures. J Hand Ther Off J Am Soc Hand Ther. 2002;15(1):71–82.
20. Rand D, Eng JJ. Predicting daily use of the affected upper extremity 1 year after stroke. J Stroke Cerebrovasc Dis Off J Natl Stroke Assoc. 2015;24(2):274–283. doi:10.1016/j.jstrokecerebrovasdis.2014.07.039.
21. Lincoln NB, Crow JL, Jackson JM, Waters GR, Adams SA, Hodgson P. The unreliability of sensory assessments. Clin Rehabil. 1991;5(4):273–282. doi:10.1177/026921559100500403.
22. Hillier S, Immink M, Thewlis D. Assessing proprioception: a systematic review of possibilities. Neurorehabil Neural Repair. 2015;29(10):933–949. doi:10.1177/1545968315573055.
23. Winward CE, Halligan PW, Wade DT. The Rivermead Assessment of Somatosensory
Performance (RASP): standardization and reliability data. Clin Rehabil. 2002;16(5):523–533. doi:10.1191/0269215502cr522oa.
24. Fugl-Meyer AR, Jääskö L, Leyman I, Olsson S, Steglind S. The post-stroke hemiplegic patient 1. A method for evaluation of physical performance. Scand J Rehabil Med. 1975;7(1):13–31.
25. Lin JH, Hsueh IP, Sheu CF, Hsieh CL. Psychometric properties of the sensory scale of the Fugl-Meyer Assessment in stroke patients. Clin Rehabil. 2004;18(4):391–397. doi:10.1191/0269215504cr737oa.
26. Pumpa LU, Cahill LS, Carey LM. Somatosensory
assessment and treatment after stroke: an evidence-practice gap. Aust Occup Ther J. 2015;62(2):93–104. doi:10.1111/1440-1630.12170.
27. Simo L, Botzer L, Ghez C, Scheidt RA. A robotic test of proprioception within the hemiparetic arm post-stroke. J Neuroengineering Rehabil. 2014;11:77. doi:10.1186/1743-0003-11-77.
28. Findlater SE, Dukelow SP. Upper extremity proprioception after stroke: bridging the gap between neuroscience and rehabilitation. J Mot Behav. 2017;49(1):27–34. doi:10.1080/00222895.2016.1219303.
29. Mrotek LA, Bengtson M, Stoeckmann T, et al The Arm Movement Detection (AMD) test: a fast robotic test of proprioceptive acuity in the arm. J Neuroeng Rehabil. 2017;14(1):64. doi:10.1186/s12984-017-0269-3.
30. Simo LS, Ghez C, Botzer L, Scheidt RA. A quantitative and standardized robotic method for the evaluation of arm proprioception after stroke. Conf Proc IEEE Eng Med Biol Soc. 2011;2011:8227–8230. doi:10.1109/IEMBS.2011.6092029.
31. Goble DJ, Hurvitz EA, Brown SH. Deficits in the ability to use proprioceptive feedback in children with hemiplegic cerebral palsy. Int J Rehabil Res. 2009;32(3):267–269. doi:10.1097/MRR.0b013e32832a62d5.
32. Semrau JA, Herter TM, Scott SH, Dukelow SP. Robotic identification of kinesthetic deficits after stroke. Stroke. 2013;44(12):3414–3421. doi:10.1161/STROKEAHA.113.002058.
33. Ostry DJ, Darainy M, Mattar AA, Wong J, Gribble PL. Somatosensory
plasticity and motor learning. J Neurosci. 2010;30(15):5384–5393.
34. Vahdat S, Darainy M, Ostry DJ. Structure of plasticity in human sensory and motor networks due to perceptual learning. J Neurosci. 2014;34(7):2451–2463. doi:10.1523/JNEUROSCI.4291-13.2014.
35. Cressman EK, Henriques DY. Sensory recalibration of hand position following visuomotor adaptation. J Neurophysiol. 2009;102(6):3505–3518.
36. Henriques DYP, Cressman EK. Visuomotor adaptation and proprioceptive recalibration. J Mot Behav. 2012;44(6):435–444. doi:10.1080/00222895.2012.659232.
37. Hoseini N, Sexton BM, Kurtz K, Liu Y, Block HJ. Adaptive staircase measurement of hand proprioception. PLoS One. 2015;10(8):e0135757. doi:10.1371/journal.pone.0135757.
38. Desrosiers J, Bourbonnais D, Bravo G, Roy PM, Guay M. Performance of the “unaffected” upper extremity of elderly stroke patients. Stroke. 1996;27(9):1564–1570.
39. Mathiowetz V, Volland G, Kashman N, Weber K. Adult norms for the Box and Block Test of manual dexterity. Am J Occup Ther. 1985;39(6):386–391.
40. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–198.
41. Halligan P, Wilson B, Cockburn J. A short screening test for visual neglect in stroke patients. Int Disabil Stud. 1990;12(3):95–99.
42. John J. Grading of muscle power: comparison of MRC and analogue scales by physiotherapists. Int J Rehabil Res. 1984;7(2):173–181.
43. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia. 1971;9(1):97–113.
44. Desrosiers J, Hébert R, Bravo G, Dutil E. The Purdue Pegboard Test: normative data for people aged 60 and over. Disabil Rehabil. 1995;17(5):217–224.
45. Grob KR, Kuster MS, Higgins SA, Lloyd DG, Yata H. Lack of correlation between different measurements of proprioception in the knee. J Bone Joint Surg Br. 2002;84(4):614–618.
46. Taylor MM, Creelman CD. PEST: efficient estimates on probability functions. J Acoust Soc Am. 1967;41(4A):782–787. doi:10.1121/1.1910407.
47. Schaap TS, Gonzales TI, Janssen TWJ, Brown SH. Proprioceptively guided reaching movements in 3D space: effects of age, task complexity and handedness. Exp Brain Res. 2015;233(2):631–639. doi:10.1007/s00221-014-4142-3.
48. Chapman CD, Heath MD, Westwood DA, Roy EA. Memory for kinesthetically defined target location: evidence for manual asymmetries. Brain Cogn. 2001;46(1/2):62–66.
49. Carson RG, Elliott D, Goodman D, Dickinson J. Manual asymmetries in the reproduction of a 3-dimensional spatial location. Neuropsychologia. 1990;28(1):99–103.
50. King J, Harding E, Karduna A. The shoulder and elbow joints and right and left sides demonstrate similar joint position sense. J Mot Behav. 2013;45(6):479–486. doi:10.1080/00222895.2013.832136.
51. Goble DJ, Lewis CA, Brown SH. Upper limb asymmetries in the utilization of proprioceptive feedback. Exp Brain Res. 2006;168(1-2):307–311. doi:10.1007/s00221-005-0280-y.
52. Goble DJ, Brown SH. Task-dependent asymmetries in the utilization of proprioceptive feedback for goal-directed movement. Exp Brain Res. 2007;180(4):693–704. doi:10.1007/s00221-007-0890-7.
53. Colley A. Spatial location judgements by right and left-handers. Cortex J Devoted Study Nerv Syst Behav. 1984;20(1):47–53.
54. Roy EA, MacKenzie C. Handedness effects in kinesthetic spatial location judgements. Cortex J Devoted Study Nerv Syst Behav. 1978;14(2):250–258.
55. Niessen MH, Veeger DH, Koppe PA, Konijnenbelt MH, van Dieën J, Janssen TW. Proprioception of the shoulder after stroke. Arch Phys Med Rehabil. 2008;89(2):333–338. doi:10.1016/j.apmr.2007.08.157.
56. Hewett P, Ganser GH. A comparison of several methods for analyzing censored data. Ann Occup Hyg. 2007;51(7):611–632. doi:10.1093/annhyg/mem045.
57. Kizony R, Zeilig G, Dudkiewicz I, Schejter-Margalit T, Rand D. Tablet apps and dexterity: comparison between 3 age groups and proof of concept for stroke rehabilitation. J Neurol Phys Ther. 2016;40(1):31–39. doi:10.1097/NPT.0000000000000110.
58. Lincoln N, Jackson J, Adams S. Reliability and Revision of the Nottingham Sensory Assessment for Stroke Patients. Physiotherapy. 1998;84(8):358–365. doi:10.1016/S0031-9406(05)61454-X.
59. Wycherley AS, Helliwell PS, Bird HA. A novel device for the measurement of proprioception in the hand. Rheumatology. 2005;44(5):638–641. doi:10.1093/rheumatology/keh568.
60. Leibowitz N, Levy N, Weingarten S, et al Automated measurement of proprioception following stroke. Disabil Rehabil. 2008;30(24):1829–1836. doi:10.1080/09638280701640145.
61. Gokeler A, Benjaminse A, Hewett TE, et al Proprioceptive deficits after ACL injury: are they clinically relevant? Br J Sports Med. 2012;46(3):180–192.
62. Gill KP, Callaghan MJ. The measurement of lumbar proprioception in individuals with and without low back pain. Spine. 1998;23(3):371–377.
63. Vargas-Irwin CE, Shakhnarovich G, Yadollahpour P, Mislow JMK, Black MJ, Donoghue JP. Decoding complete reach and grasp actions from local primary motor cortex populations. J Neurosci. 2010;30(29):9659–9669. doi:10.1523/JNEUROSCI.5443-09.2010.
64. Bosco G, Poppele RE. Representation of multiple kinematic parameters of the cat hind limb in spinocerebellar activity. J Neurophysiol. 1997;78(3):1421–1432.
65. Semrau JA, Herter TM, Scott SH, Dukelow SP. Examining differences in patterns of sensory and motor recovery after stroke with robotics. Stroke J Cereb Circ. 2015;46(12):3459–3469. doi:10.1161/STROKEAHA.115.010750.
66. Fuentes CT, Bastian AJ. Where is your arm? Variations in proprioception across space and tasks. J Neurophysiol. 2010;103(1):164–171.
finger; kinesthesia; outcomes assessment; psychometrics; somatosensory
Supplemental Digital Content
© 2019 Academy of Neurologic Physical Therapy, APTA