Sage Journals: Discover world-class research

Abstract

Human wrist proprioception is particularly important due to its role in manual dexterity and associated tasks of daily living. Most studies have focused on testing single degree of freedom joints or were only capable of displacing or moving a joint in a single plane, such as flexion/extension of the wrist. The purpose of this study was to examine the effects of both direction and angular level on the accuracy of human wrist position reproduction error. Sixty subjects (all males) without a history of wrist pathology were recruited from a university campus. Subjects performed a position reproduction task in eight directions at three angular levels. The results showed that wrist position reproduction error depends on direction and angular level. Comparable reliability for the intra-observer measurements for wrist range of motion and joint position sense. The orientation of the joint position sense production ellipse is similar to the orientation of the range of motion ellipse, indicating that subjects generated the most accurate in directions where it is easy to generate more range of motion and the lowest accurate in directions where it is not easy to generate more range of motion. Joint position sense decreased in accuracy as the joint angle increased. The position reproduction error depends on the angular level, and subjects overshoot the target angle for the low angular levels (25% range of motion) and undershoot the target angle for high angular levels (75% range of motion). Mapping the human wrist joint position sense ellipse contributes to our understanding of the comprehensive proprioceptive function of the wrist. This technique offers the opportunity to assess all of the directions of proprioceptive function, which in turn may aid in improving therapeutic approaches.

Keywords

Proprioception joint position sense range of motion inertial and magnetic measurement systems wrist

Introduction

Proprioception is critical for accurate movement, allowing the body to deliver afferent information from its periphery to the central nervous system (CNS) in order to maintain postural status, joint position sense (JPS), and overall position in space. Conscious proprioceptive senses can be subdivided into three categories: kinesthesia, JPS, and sense of resistance or force.^1,2 JPS is defined as the ability to accurately reproduce a specific joint angle and is influenced by both muscle command³ and muscle conditioning.⁴

Human wrist proprioception is particularly important due to the role it plays in manual dexterity and associated tasks of daily living.⁵ Accurate assessment of wrist joint function is very important for establishing strategies to maximize functional potential and evaluating the treatment and progression of disease.⁶ The functional consequences of impaired wrist proprioceptive ability relate to the precise dynamic or static control of goal-directed arm movements.⁷

At present, there is no established objective method for clinical assessment, and even though clinical rating scales such as the static and dynamic up–down test,^8,9 a similar dual-joint position test,¹⁰ and the Rivermead Assessment of Somatosensory Performance¹¹ are available and currently in use, they can provide only qualitative information and have low resolution.^12,13

Over the past few years, several models of quantitative assessment have been proposed to investigate different aspects of proprioception. Some use the combinations of simple passive apparatuses restraining the movements to specific planes and matching paradigms with protractor scales^14–16 to provide quantitative outcome measures. The robotic approaches to assess proprioception have also increased over the last few years, as they can take advantage of the control and sensing capabilities of robotic technology¹⁷ to address requirements for an optimal assessment, such as high resolution, high reproducibility, and good control over stimuli. But expensive robotic devices are difficult to use in clinical settings.¹⁸

Inertial and magnetic measurement systems (IMMS) are a new generation of motion analysis systems that have been used in studies to measure the three-dimensional (3D) orientation of body segments and joint angles. An IMMS consists of one or more sensing units (SUs), which integrate an inertial measurement system composed of one 3D accelerometer, one 3D gyroscope, and a 3D magnetometer. Data supplied by the accelerometer, gyroscope, and magnetometer are combined through sensor-fusion algorithms^19,20 to measure the 3D orientation of the SU’s system of reference (SoR)—defined based on the sensing axes of the inertial and magnetic sensors—with respect to an earth-based SoR.²¹ Some studies performed with IMMS have shown good intra- and inter-observer agreement.^21–26

Up to this point, many studies have investigated the JPS of the wrist. However, most of these studies have focused on testing single DoF joints or were only capable of displacing or moving a joint in a single plane, such as flexion/extension of the wrist.^27–33 There are some studies that have assessed proprioceptive function across the three degrees of freedom (DoF) of the wrist/hand complex.^12,25,34 This is restrictive in that it only provides partial information on the proprioceptive status of a joint. However, it is clinically relevant to examine proprioceptive function for all of the directions of a given joint or limb system.^21,35 Due to the lack of research focusing on the variety of testing protocols used, further study is needed to better understand JPS for all of the directions of the wrist.

To our knowledge, no research reported the relationship between human wrist JPS and directional and angular levels. It is necessary to understand what factors affect the accuracy of position sense and how it changes in response to conditions such as direction and angular level. The aims of the study were to examine wrist joint proprioceptive acuity using IMMS. Specifically, we applied a wrist joint position matching task to determine JPS for different directions and angular levels.

Methods

Subjects

Sixty subjects (all males, age 19.7 ± 0.6 years, 64.5 ± 5.8 kg, 169.8 ± 3.1 cm, all right-handed) were tested. Subjects self-reported their hand dominance by indicating with which hand they write. Exclusion criteria included the following: (1) a long-standing history of highly skilled motor activity, such as playing a musical instrument or basketball;²⁸ (2) prior wrist surgery; and (3) the presence of wrist pain or pathology. Subjects were briefed on the purpose of the study and experimental procedure, and completed the informed consent form. This study was approved by the institutional ethics review board of Renmin University of China.

Apparatus

The joint position sense measuring system (JPSS) (Kjyl Technologies, Beijing Kangjian Medical Devices Co., Ltd., CHN) is an IMMS consisting of one SU connected via Bluetooth to a computer for data processing and data storage that is worn on the belt. The sensor is small (38 mm × 53 mm × 21 mm), lightweight (30 g), and ambulatory (i.e. they can be used everywhere at any time and are not restricted to a certain measurement volume in a laboratory). In Figure 1, given this 3D orientation, a JPSS, which is attached to the wrist, was used to estimate the joint angles of the wrist. For correct positioning, a line was positioned from the midpoint of the wrist to the fingertip of the index finger. The base of the wireless JPSS sensor was positioned over the flat surface of the wrist dorsum, with the lower edge along this line and away from the wrist. For the present study, the JPSS sampling frequency was fixed to 100 Hz. Based on the JPSS, we developed a protocol that measures the JPS of the wrist.

Figure 1.

Orientation of the SU’s SoR with respect to an earth-based SoR is provided as an output.

Protocol

The study was performed in a room that provided a quiet environment to assure proper concentration.³⁶ Subjects sat in a chair approximately 60 cm in front of a 16-in. LCD monitor secured with restraining straps across the midriff. Elbows were then positioned at shoulder height on an adjustable padded brace. The brace was secured in the optimum height position to produce a 90 degree angle between the upper arm and the forearm. The forearm was firmly strapped to the brace to ensure repeatability of wrist positioning across different trials and to avoid joint misalignment (slippage) during the experiment.²⁴ The JPSS was attached to the subject’s hand via a padded, non-elastic strap. The test hand was placed in a pronated position, and participants were asked to maintain this same arm configuration throughout the testing period. Participants were able to view their wrist position output on a 16-in. computer monitor.

The JPSS was used to record data during experimental trials. The range of motion (ROM) test program and JPS test program were coded using C++ software.³⁷

ROM test

Subjects moved their wrist in flexion until reaching their full ROM and then rotated their wrist counterclockwise in a circle, always maintaining the full ROM.

Data (ψ, θ) were recorded by the JPSS during experimental trials, where ψ is the angle in the ulnar/radial deviation axis, and θ is the angle in the flexion/extension axis. (Mψ₂, 0) was recorded as the maximal angle on a radial deviation axis (direction 2) when ψ < 0, and |θ| was the minimum value. (0, Mθ₄) was recorded as the maximal angle on the flexion axis (direction 4) when θ < 0, and |ψ| was the minimum value. (Mψ₆, 0) was recorded as the maximal angle on the ulnar deviation axis (direction 6) when ψ > 0, and |θ| was the minimum value. (0, Mθ₈) was recorded as the maximal angle on the extension axis (direction 8) when θ > 0, and |ψ| was the minimum value. (Mψ₁, Mθ₁) was recorded as the maximal angle in direction 1, which forms an angle of 45° with the extension axis when ψ < 0 and θ > 0, and ||ψ| – |θ|| was the minimum value. (Mψ₃, Mθ₃) was recorded as the maximal angle in direction 3, which forms an angle of 135° with the extension axis when ψ < 0 and θ_i < 0, and||ψ| – |θ|| was the minimum value. (Mψ₅, Mθ₅) was recorded as the maximal angle in direction 5, which forms an angle of −135° with the extension axis when ψ > 0 and θ_i < 0, and||ψ| – |θ|| was the minimum value. (Mψ₇, Mθ₇) was recorded as the maximal angle in direction 7, which forms an angle of −45° with the extension axis when ψ > 0 and θ > 0, and||ψ| – |θ|| was the minimum value. (Mψ₉, Mθ₉) was recorded as the maximal angle in the radial deviation axis when ψ was the maximum value. (Mψ₁₀, Mθ₁₀) was recorded as the maximal angle in the ulnar deviation axis when ψ was the minimum value. (Mψ₁₁, Mθ₁₁) was recorded as the maximal angle in the extension axis when θ was the maximum value. (Mψ₁₂, Mθ₁₂) was recorded as the maximal angle in the flexion axis when θ was the minimum value (Figure 2). This trial was repeated three times. $R_{k}$ is the ROM of the k-direction and was calculated using the following equation

R_{k} = \sqrt{M {ψ_{k}}^{2} + M {θ_{k}}^{2}}, (k = 1, 2, \dots, 12)

(1)

where $M ψ_{k}$ is the ROM in the ulnar/radial deviation axis of the k-direction for all the data together, and $M θ_{k}$ is the ROM in the flexion/extension axis of the k-direction for all the data together.

Figure 2.

A schematic representation of an ellipse that represents movement tracking of the wrist, shown in red. The black triangles indicate ROM values in that direction. Directions 2, 4, 6, and 8 are the radial deviation, flexion, ulnar, and extension deviation, respectively. The angle between every two adjacent directions is 45°.

To describe the ROM in all of the directions, ellipses were fitted through ROM measurements for all data together. To fit the ellipse to the data, we used “Direct Ellipse Fit”³⁸ fitting to find a set of parameters for the general quadratic equation to minimize the distance between the data points and the ellipse (see equation (5))

A \cdot X^{2} + B \cdot X \cdot Y + C \cdot Y^{2} + D \cdot X + E \cdot Y + 1 = 0

(2)

Orientation of the major axis

θ = \frac{1}{2} \cdot \arctan \frac{B}{A - C}

(3)

Location of the center of the ellipse

x_{c} = \frac{B \cdot E - 2 \cdot C \cdot D}{4 \cdot A \cdot C - B^{2}}

(4)

y_{c} = \frac{B \cdot D - 2 \cdot A \cdot E}{4 \cdot A \cdot C - B^{2}}

(5)

Semi-major axis

a^{2} = \frac{2 \cdot (A \cdot x_{c} + C \cdot {y_{c}}^{2} + B \cdot x_{c} \cdot y_{c} - 1)}{A + C + \sqrt{{(A - C)}^{2} + B^{2}}}

(6)

Semi-minor axis

b^{2} = \frac{2 \cdot (A \cdot x_{c} + C \cdot {y_{c}}^{2} + B \cdot x_{c} \cdot y_{c} - 1)}{A + C - \sqrt{{(A - C)}^{2} + B^{2}}}

(7)

JPS test

The testing protocol was thoroughly explained to subjects while they watched the visual output on a computer monitor.³⁹ A screen with a black circle was presented to subjects, via custom-written C++ software. The black circle represented the target position for a given trial. On the screen, an arrow appeared to prompt subjects as to which direction to move their arm to arrive at the target position. A red dot then appeared on the screen, representing the instantaneous wrist position (Figure 3). All trials began with the wrist in the neutral position. Subjects were instructed to move their wrist until the red dot on the screen was inside the black circle (when the actual wrist position was within 0.5° of the target position), indicating that the wrist was in the targeted position. Once the wrist was in the target position for 1 s, an audible beep was heard, and the screen turned black and remained so for the remainder of the trial. Subjects were instructed to maintain their wrist in the target position for 3 s, during which time they were to concentrate only on the position of the wrist. After the subject maintained the target position for 3 s, a computer-generated voice instructed subjects to relax, at which time the subject moved the wrist back to a neutral position.

Figure 3.

A schematic representation of the computer output seen through the monitor display guiding the subject to the target position. The red dot indicates the instantaneous wrist position, and the black circles indicate the targeted position in eight directions at three angular levels.

Three seconds later, subjects attempted to replicate the target position. When subjects perceived that the wrist was at the target position, they used the contralateral hand to push a trigger button interfaced with the computer to time-stamp the reproduced position. Subjects were instructed to maintain their wrist in the reproduced position for 1 s after pushing the trigger button, while data for the i-position and j-trial were recorded as $R_{m} (ε_{m}, γ_{m}) (m = 1, 2, 3, \dots, 24)$ , at which time the trial ended (where $ε_{m}$ is the reproduction position in the ulnar/radial deviation axis of the m-position, and $γ_{m}$ is the reproduction position in the flexion/extension axis of the m-position). This trial was performed three times. The procedure was explained and demonstrated to subjects until they felt comfortable with the process. Prior to the start of testing, subjects performed at least one practice trial. Practice trials were repeated until subjects felt confident in performing the task.

Twenty-four target positions $(T_{m} (α_{m}, β_{m})$ $(m = 1, 2, 3, \dots, 24))$ were presented as illustrated in Figure 3. These 24 trials were automated via the software and separated by a 30-s rest interval. Target positions were presented in random order. To facilitate attention to the task being maintained throughout testing, subjects were allowed to rest for 3 to 5 min after 10 trials.²⁵ To establish reliability, the 24 trials were performed in a randomized order, unlike that used for the first 24-trial sequence.

There were three dependent variables for JPS errors: absolute error (AE),^40–44 constant error (CE),^4,45–48 and variable error (VE).^49–51 While AE is an assessment of overall error, CE reflects the directionality of errors and VE is the variability of the error over a series of trials and represents the accuracy of the performance.

$A E_{m}$ was calculated to obtain the AE between the target position and reproduced position of the m-position for all the data together. $C E_{ψ m}$ was calculated to obtain the CE between the target position and reproduced position in the ulnar/radial deviation axis of the m-position for all the data together. $C E_{θ m}$ was calculated to obtain the CE between the target position and reproduced position in the flexion/extension axis of the m-position for all the data together. $DA E_{l}$ is the mean AE of the l-direction. VE_m was calculated as the overall standard deviation (SD) of CE from three trials of the m-position, irrespective of the target range. These parameters were calculated using the following equation

A E_{m} = \sqrt{{(ε_{m} - α_{m})}^{2} + {(γ_{m} - β_{m})}^{2}}

(8)

C E_{ψ m} = ε_{m} - α_{m}

(9)

C E_{θ m} = γ_{m} - β_{m}

(10)

V E_{m} = \sqrt{{(ε_{m} - {\bar{ε}}_{m})}^{2} + {(γ_{m} - {\bar{γ}}_{m})}^{2}}, (m = 1, 2, \dots, 24)

(11)

\begin{matrix} DA E_{l} = \frac{A E_{3 \cdot (l - 1) + 1} + A E_{3 \cdot (l - 1) + 2} + A E_{3 \cdot (l - 1) + 3}}{3}, \\ (l = 1, 2, \dots, 8) \end{matrix}

(12)

where $α_{m}$ is the target position in the ulnar/radial deviation axis of the m-position, and $β_{m}$ is the target position in the flexion/extension axis of the m-position.

To fit the ellipse to the data ( $(- DA E_{1} / \sqrt{2},$ $(- DA E_{1} / \sqrt{2})$ , $(- DA E_{2}, 0)$ , $(- DA E_{3} / \sqrt{2},$ $- DA E_{3} / \sqrt{2})$ , $(0, - DA E_{4})$ , $(DA E_{5} / \sqrt{2}, - DA E_{5} / \sqrt{2})$ , $(DA E_{6}, 0)$ , $(DA E_{7} / \sqrt{2}, DA E_{7} / \sqrt{2})$ , $(0, DA E_{8})$ ), we also used “Direct Ellipse Fit.”

To evaluate the reliability and measurement precision of the JPSS, measurements were performed on two different days (session 1 and session 2) by two different experienced testers.²⁵

Data analysis

To quantify test–retest reliability, the intraclass correlation coefficient (ICC) was calculated for two sessions.^52,53 ICC estimates were based on a single-measurement, absolute-agreement, two-way mixed-effects model.⁵⁴ The evaluation criteria and accepted standards for ICC values were outlined as follows: 0.00 to 0.39, poor; 0.40 to 0.74, fair; and 0.75 to 1.00, good.⁵⁵ A paired t-test was performed to identify systematic differences between the two sessions.⁵⁶

Two 8 × 3 repeated-measures analyses of variance were used to assess the effect of direction (directions 1, 2, 3, 4, 5, 6, 7, and 8) and angular level (25% ROM, 50% ROM, and 75% ROM) on AE, CE, and VE.⁵⁷ Follow-up comparisons were performed when appropriate using a Bonferroni adjustment for multiple comparisons.

Correlations between the ROM in different directions were assessed by Pearson’s correlation analysis. Statistical analysis was performed using IBM SPSS Statistics, version 20, software (IBM Corporation, Armonk, NY, USA). All data are provided as the M±SD. The level of statistical significance was set at p < 0.05.

Results

Reliability of the JPSS

Table 1 shows the test–retest reliability results for the 12 variables in the ROM test. Paired t-test results found no significant differences between the two sessions. Test–retest reliability between the two sessions exhibited a fair to excellent correlation, with ICCs ranging from 0.80 to 0.93.

Table 1.

Test–retest reliability results of the wrist ROM test (°).

	Session 1	Session 2	ICC	p
$R_{1}$	49.7 ± 5.7	49.8 ± 5.0	0.81	0.002**
$R_{2}$	26.2 ± 5.8	26.5 ± 6.4	0.80	0.005**
$R_{3}$	29.0 ± 5.9	29.3 ± 6.5	0.84	0.003**
$R_{4}$	55.6 ± 5.3	54.0 ± 5.5	0.84	0.012*
$R_{5}$	68.6 ± 6.1	69.6 ± 4.9	0.85	0.021*
$R_{6}$	41.4 ± 5.1	41.7 ± 5.5	0.83	0.004**
$R_{7}$	42.0 ± 5.7	41.9 ± 6.2	0.89	0.007**
$R_{8}$	58.7 ± 4.4	58.1 ± 4.6	0.90	0.008**
$R_{9}$	55.1 ± 5.9	54.4 ± 6.1	0.93	0.021*
$R_{10}$	64.9 ± 5.6	66.0 ± 5.7	0.86	0.032*
$R_{11}$	61.9 ± 7.1	61.7 ± 6.0	0.81	0.006**
$R_{12}$	64.9 ± 6.0	65.2 ± 5.1	0.89	0.015*

ROM: range of motion; ICC: intraclass correlation coefficient.

p < 0.05; **p < 0.01.

Table 2 shows the test–retest reliability results for the 24 variables in the JPS test. Paired t-test results found no significant differences between the two sessions. The test–retest reliability between the two sessions exhibited a fair to excellent correlation, with ICCs ranging from 0.76 to 0.91.

Table 2.

Test–retest reliability results of the wrist joint position sense test (°).

$i$	Session 1	Session 2	ICC	p
1	2.3 ± 0.6	2.4 ± 0.7	0.85	0.004**
2	3.2 ± 0.8	3.2 ± 0.7	0.79	0.001**
3	4.2 ± 0.9	4.1 ± 0.6	0.78	0.003**
4	2.7 ± 0.8	2.5 ± 0.6	0.86	0.011*
5	3.2 ± 0.6	2.9 ± 0.5	0.76	0.021*
6	3.0 ± 0.6	3.1 ± 0.7	0.77	0.009**
7	2.8 ± 0.8	2.8 ± 0.7	0.91	0.001**
8	3.7 ± 0.8	3.6 ± 0.5	0.78	0.003**
9	4.7 ± 0.4	4.7 ± 0.5	0.81	0.001**
10	3.4 ± 0.5	3.4 ± 0.6	0.91	0.001**
11	4.2 ± 0.6	4.0 ± 0.7	0.76	0.010*
12	5.1 ± 0.4	4.8 ± 0.5	0.76	0.021*
13	3.3 ± 0.5	3.8 ± 0.4	0.84	0.031*
14	3.9 ± 0.8	3.8 ± 0.5	0.79	0.012*
15	4.0 ± 0.7	4.1 ± 0.4	0.79	0.009**
16	3.0 ± 0.7	3.2 ± 0.5	0.86	0.020*
17	3.7 ± 0.6	3.2 ± 0.5	0.80	0.032*
18	3.4 ± 0.6	3.0 ± 0.6	0.78	0.035*
19	3.9 ± 0.5	3.6 ± 0.7	0.82	0.036*
20	4.4 ± 0.7	4.1 ± 0.6	0.77	0.031*
21	4.8 ± 0.7	4.8 ± 0.7	0.81	0.001**
22	2.5 ± 0.6	2.7 ± 0.5	0.85	0.023*
23	3.7 ± 0.7	3.9 ± 0.4	0.84	0.025*
24	4.2 ± 0.8	4.2 ± 0.6	0.76	0.001**

ICC: intraclass correlation coefficient.

p < 0.05; **p < 0.01.

ROM of the wrist

The maximum ROM was 68.6°± 6.1°, and the minimum ROM was 26.2°± 5.8° (see Table 3).

Table 3.

ROM of the wrist in different directions (°).

$R_{1}$	$R_{2}$	$R_{3}$	$R_{4}$	$R_{5}$	$R_{6}$	$R_{7}$	$R_{8}$	$R_{9}$	$R_{10}$	$R_{11}$	$R_{12}$
49.7 ± 5.7	26.2 ± 5.8	29.0 ± 5.9	55.6 ± 5.3	68.6 ± 6.1	41.4 ± 5.1	42.0 ± 5.7	58.7 ± 4.4	55.1 ± 5.9	64.9 ± 5.6	61.9 ± 7.1	64.9 ± 6.0

ROM: range of motion.

For the ROM test trials, an ellipse was found, indicating that the ROM differs in all of the directions, with the major axis of the ellipse as the maximum ROM direction. In Figure 4, the greater the distance from the center of the graph, the greater is the ROM. Black triangles show mean ROM values in that direction over subjects. An ellipse was fitted through these data, shown in red. The general elliptical shape of the data is very apparent.

Figure 4.

ROM data with fitted ellipses, showing a polar view of the ROM at different directions. Black triangles indicate mean ROM values in that direction across subjects.

The parameters of the ellipse fitted to the complete dataset were as follows: A = −0.0007727, B = −0.0005416, C = −0.0003450, D = 0.0108721, and E = 0.0027005. The orientation of the major axis (least accurate direction) was as follows: θ = 25.8°. The location of the center of the ellipse was as follows: x_c = 7.8° and y_c = −2.2°; semi-major axis: a = 68.3°; and semi-minor axis: b = 41.0°.

JPS of the wrist

AE

There was no significant interaction observed between direction and angular levels: F = 1.26 and p = 0.23. A significant main effect was found for direction, F = 6.21 and p = 0.01, indicating that the JPS differs between direction 1 (M = 3.2°, SD = 0.6), direction 2 (M = 3.0°, SD = 0.5), direction 3 (M = 3.7°, SD = 0.7), direction 4 (M = 4.2°, SD = 0.7), direction 5 (M = 3.7°, SD = 0.7), direction 6 (M = 3.4°, SD = 0.7), direction 7 (M = 4.4°, SD = 0.9), and direction 8 (M = 3.5°, SD = 0.6). The results showed the JPS at directions 1, 2, and 6 to be significantly more accurate than for other directions, p < 0.05, and the JPS at directions 3, 4, and 7 to be significantly less accurate than for other directions, p < 0.05. The major axis of the ellipse was the least accurate direction. In Figure 5, the greater the distance from the center of the graph, the less accurate are the results. Black X signs indicate mean JPS values in that direction over subjects. An ellipse was fitted through these data, shown in blue. The general elliptical shape of the data is very apparent.

Figure 5.

Position reproduction error with fitted ellipses, showing a polar view of the JPS at different directions. Black X signs indicate mean JPS values in that direction across subjects.

The parameters of the ellipse fitted to the complete dataset were as follows: A = −0.0918048, B = 0.0266083, C = −0.0624640, D = 0.0531735, and E = −0.0227221; the orientation of the major axis (least accurate direction) was as follows: θ = −21.1°; the location of the center of the ellipse was as follows: x_c = 0.3° and y_c = −0.1°; semi-major axis: a = 4.2°; and semi-minor axis: b = 3.8°.

The orientation of the JPS production ellipse (blue) was similar to the orientation of the ROM ellipse (red) (the direction of the major axis of the JPS production ellipse is similar to the direction of the minor axis of the ROM ellipse) (Figure 6).

Figure 6.

Ellipses fitted to normalized ROM (red) and JPS (blue) data across subjects. Black triangles indicate mean ROM values in that direction across subjects. Black X signs indicate mean JPS values in that direction across subjects.

There was a significant main effect found for angular level, F = 26.97 and p = 0.01. Follow-up t-tests were performed with a Bonferroni adjustment for multiple comparisons. The results showed the JPS at 25% ROM (M = 3.0°, SD = 0.8) to be significantly more accurate than 50% ROM (M = 3.8°, SD = 0.5) and 75% ROM (M = 4.2°, SD = 0.5), p = 0.01. AE decreased in accuracy as the angular level increased (Figure 7).

Figure 7.

AE, an assessment of the position reproduction overall error, as a function of angular levels. The red squares indicate mean AE values across subjects.

CE

CE in the ulnar/radial deviation axis

The interaction between direction and angular level was not significant, F = 1.28 and p = 0.21. There was no significant main effect found for direction, F = 1.27 and p = 0.27. However, there was a significant main effect found for angular level, F = 8.27 and p = 0.01. A significant difference was found between 25% ROM (M = 0.4°, SD = 0.1), 50% ROM (M = −0.1°, SD = 0.1), and 75% ROM (M = −0.4°, SD = 0.2), p = 0.01.

CE in the flexion/extension axis

The interaction between direction and angular level was not significant, F = 1.07 and p = 0.38. There was also no significant main effect found for direction, F = 1.907 and p = 0.069. There was a significant main effect found for angular level, F = 3.98 and p = 0.02. A significant difference was found between 25% ROM (M = 0.5°, SD = 0.2), 50% ROM (M = −0.1°, SD = 0.2), and 75% ROM (M = −0.2°, SD = 0.1), p = 0.01. In Figure 8, the blue triangles indicate mean CE values in the ulnar/radial deviation axis across subjects, and the red circles indicate the mean CE values in the flexion/extension axis across subjects.

Figure 8.

CE, the position of reproduction directionality of error, as a function of angular levels.

VE

There was no significant interaction observed between direction and angular levels, F = 1.56 and p = 0.08. A significant main effect was found for direction, F = 2.51 and p = 0.02, indicating that the JPS differs between direction 1 (M = 2.5°, SD = 0.7), direction 2 (M = 2.4°, SD = 0.8), direction 3 (M = 2.6°, SD = 0.6), direction 4 (M = 2.9°, SD = 0.8), direction 5 (M = 2.9°, SD = 0.7), direction 6 (M = 2.7°, SD = 0.8), direction 7 (M = 3.1°, SD = 0.8), and direction 8 (M = 2.5°, SD = 0.7). The results showed the JPS at directions 1, 2, and 8 to be significantly more precisely estimated than for other directions, p < 0.05, and the JPS at direction 7 to be significantly less precisely estimated than for other directions, p < 0.05 (Figure 9).

Figure 9.

VE, the variability of the error over a series of trials, represents the precision of the performance, as a function of direction.

A significant main effect was found for angular level, F = 17.65 and p = 0.01. Follow-up t-tests were performed with a Bonferroni adjustment for multiple comparisons. The results showed the JPS at 25% ROM (M = 2.2°, SD = 0.6) to be significantly more precisely estimated than 50% ROM (M = 2.9°, SD = 0.8) and 75% ROM (M = 2.9°, SD = 0.7), p = 0.01. AE decreased in accuracy as the angular level increased (Figure 10).

Figure 10.

VE, the variability of the error over a series of trials, represents the precision of the performance, as a function of angular levels.

For this method to have any clinical usefulness, individual data were presented and compared against mean data (Figure 11).

Figure 11.

(a) AE, (b, c) CE, and (d) VE of individual data and mean data, which represent the accuracy and precision of the performance, as a function of direction and angular levels.

Discussion

Reliability and precision of the JPSS

ICCs exhibited comparable reliability for both intra- and inter-observer measurements for wrist ROM and JPS, indicating that the reliability and precision of the wrist ROM and JPS measurement with the JPSS, as well as the current protocol, are not dependent on either the observer or day of measurement. Similar to our results, van den Noort found high ICCs within (intra) and between (inter) observers, especially for scapular retraction/protraction (ICC = 0.65–0.85) and medio/lateral rotation (ICC = 0.56–240.91).

JPS in different directions (human wrist JPS ellipse)

We have found the ROM for flexion/extension DoF to be 55.6°± 5.3°/58.7°± 4.4°, ulnar/radial deviation DoF is 41.1°± 5.1°/26.2°± 5.8°, and an average AE of reposition of a value from 3.0° to 4.2°. These results are similar to these data from other works of related topics (the ROM for flexion/extension DoF is 67.6°± 8.1°/71.0°± 9.6°, ulnar/radial deviation DoF: 44.3°± 8.2°/21.3°± 6.3°,⁵⁸ and the error of acuity for flexion/extension is 4.64°± 0.24°; abduction/adduction: 3.68°± 0.32°).¹²

Both AE and VE depend on the directions. The direction of the major axis of the JPS reproduction ellipse was the least accurate direction, while the direction of the minor axis of the ROM ellipse was the least ROM direction. In addition, the direction of the minor axis of the JPS reproduction ellipse was the most accurate direction, and the direction of the major axis of the ROM ellipse was the greatest ROM direction, indicating that subjects generated the most accuracy in directions in which it was easy to generate more ROM and the lowest accuracy in directions where it was not easy to generate more ROM (Figure 6). VEs were calculated to represent the precision of JPS. Subjects generated more precise estimates in the radial deviation, extension, and the direction between radial deviation and extension, and less precise estimates in the direction between ulnar deviation and extension than for other directions.

It has been postulated that such directional differences may be due to asymmetries in motor performance, structural differences in wrist joints, antagonistic muscle mass, and the density and distribution of receptors.⁵⁹

It is possible that, for different directions, the combination of the muscles used may change or muscles may react differently, which could explain the effect on ROM and position reproductions in different directions. Immunohistochemical studies of wrist anatomy revealed a high distribution density of mechanoreceptors in the dorso-radial ligaments, such as the dorsal radiocarpal, dorsal intercarpal, and scapholunate interosseous; medium density in the volar and volar-triquetral; and a long radiolunate ligament nearly devoid of mechanoreceptors. The wrist ligaments were found to vary in innervation—the dorsal intercarpal (DIC), dorsal radiocarpal (DRC), and scapholunate interosseous (SLI) being richly innervated and the long radiolunate (LRL) being almost without innervation. The difference in innervation between the ligaments might indicate differences in function. Ligaments without innervation might act as structures of passive restraint, whereas ligaments with rich innervation are proposed to also provide proprioceptive information. Wrist ligaments vary with regard to sensory and biomechanical functions. Rather, based on the differences found in structural composition and innervation, wrist ligaments are regarded as either mechanically important ligaments or sensory ligaments. The mechanically important ligaments are ligaments with densely packed collagen bundles and limited innervation. They are located primarily in the radial, force-bearing column of the wrist. The sensory ligaments, by contrast, are richly innervated although less dense in connective tissue composition and are related to the triquetrum. The triquetrum and its ligamentous attachments are regarded as key elements in the generation of the proprioceptive information necessary for adequate neuromuscular wrist stabilization.^60,61 There are more muscle spindles in muscles with larger cross-sectional areas,^62,63 and wrist flexors have twice the volume of wrist extensors.⁶⁴ It has been hypothesized that wrist extension–directed movements stretch the wrist flexors, thereby activating muscle spindles, rendering them more easily detected than flexion-directed movements.⁷ These differences in mechanoreceptor density and innervation might be responsible for the proprioceptive anisotropy reported in our previous investigation.²⁵

JPS in different angular levels

AE, CE, and VE depend on the angular level. JPS decreased in accuracy and precision as the angular level increased, and subjects overshot the target angle for the low angular levels (25% ROM) and undershot the target angle for high angular levels (75% ROM).

The means by which the reference joint angle is established also is an important factor to consider in tests of position matching ability. In one study, 20 healthy young adults were asked to match targets that were either 20° or 40° from the start elbow joint position. AEs were found to be, on average, 66% greater for the larger target amplitude.⁶⁵ One hypothesis that has been put forth to explain this finding is that the increased errors seen with larger target amplitudes reflect increased sensorimotor noise when movement requires an increased neural control signal.⁶⁶

It is possible that with increasing angular level, the contribution of joint receptors and skin receptors may change, which could explain the changing effect of angular level on position reproductions. Ferrell and Smith found that joint receptors provided positional information principally at the extremes of the normal range of joint movements, perhaps acting as limit detectors.⁶⁷ At more proximal joints, it appears that muscle afferents provide the major proprioceptive signals. It may be that joint input is more important than muscle afferent input at the extremes of the normal range of joint movements.⁶⁸ Joint rotation causes the skin on one side of the joint to be stretched while it is slackened or even folded on the other side. Such deformations stimulate skin mechanoreceptors. The sensitivity of human skin stretch receptors when expressed as impulses per degree of joint motion is similar to that of muscle spindle afferents.^69,70 Skin receptors become activated by skin stretching as the angular level increases. It has been postulated that as joints approach the limits of their ROM, joint receptor and skin receptor information may bias perception of joint angles.⁷¹

Conclusion

We here experimentally analyzed the influence of different directions and angular levels on the perception of human wrist position. The orientation of the JPS production ellipse is similar to the orientation of the ROM ellipse, indicating that subjects generated the most accuracy in directions in which it is easy to generate more ROM and the lowest accuracy in directions in which it is not easy to generate more ROM. JPS is one of the key components of proprioception that decreases in accuracy as the joint angle increases. The position reproduction error depends on the angular level, and subjects overshoot the target angle for the low angular levels (25% ROM) and undershoot the target angle for high angular levels (75% ROM). While mapping the human wrist joint position, sense ellipse contributes to our understanding of the comprehensive proprioceptive function of the wrist. This technique offers the opportunity to assess all of the directions of proprioceptive function, which in turn may aid in improving therapeutic approaches.

Footnotes

Handling Editor: Joel Rodrigues

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to acknowledge the support of the National Social Science Fund of China (BLA160071).

ORCID iD

Lin Li

References

Riemann

Lephart

SM.

The sensorimotor system, part I: the physiologic basis of functional joint stability. J Athl Train 2002; 37(1): 71–79.

Cappello

Contu

Elangovan

et al . Evaluation of wrist joint proprioception by means of a robotic device. Int Conf Ubiquit Robot Ambient Intel 1997; 1997: 531–534.

Smith

Crawford

Proske

et al . Signals of motor command bias joint position sense in the presence of feedback from proprioceptors. J Appl Physiol 2009; 106(3): 950–958.

Winter

Allen

Proske

Muscle spindle signals combine with the sense of effort to indicate limb position. J Physiol 2005; 568(Pt. 3): 1035–1046.

Hoseini

Sexton

Kurtz

et al . Adaptive staircase measurement of hand proprioception. PLoS ONE 2015; 10(8): 1–16.

Stambolieva

Fractal properties of postural sway during quiet stance with changed visual and proprioceptive inputs. J Physiol Sci 2011; 61(2): 123–130.

Wright

Adamo

Brown

SH.

Age-related declines in the detection of passive wrist movement. Neurosci Lett 2011; 500(2): 108–112.

Lincoln

Crow

Jackson

et al . The unreliability of sensory assessment. Clin Rehabil 1991; 5(4): 273–282.

Gilman

Joint position sense and vibration sense: anatomical organisation and assessment. J Neurol Neurosurg Psychiatry 2002; 73(5): 473–477.

10.

Beckmann

Ciftci

Ertekin

The detection of sensitivity of proprioception by a new clinical test: the dual joint position test. Clin Neurol Neurosurg 2013; 115(7): 1023–1027.

11.

Winward

Halligan

Wade

DT.

The Rivermead Assessment of Somatosensory Performance (RASP): standardization and reliability data. Clin Rehabil 2002; 16(5): 523–533.

12.

Marini

Squeri

Morasso

et al . Robot-aided mapping of wrist proprioceptive acuity across a 3D workspace. PLoS ONE 2016; 11(8): e0161155.

13.

Rinderknecht

Popp

Lambercy

et al . Reliable and rapid robotic assessment of wrist proprioception using a gauge position matching paradigm. Front Hum Neurosci 2016; 10: 316.

14.

Carey

Oke

Matyas

TA.

Impaired limb position sense after stroke: a quantitative test for clinical use. Arch Phys Med Rehabil 1996; 77(12): 1271–1278.

15.

Wycherley

Helliwell

Bird

HA.

A novel device for the measurement of proprioception in the hand. Rheumatology 2005; 44(5): 638–641.

16.

Schmidt

Depper

Kerkhoff

Effects of age, sex and arm on the precision of arm position sense-left-arm superiority in healthy right-handers. Front Hum Neurosci 2013; 7: 915.

17.

Scott

Dukelow

SP.

Potential of robots as next-generation technology for clinical assessment of neurological disorders and upper-limb therapy. J Rehabil Res Dev 2011; 48(4): 335–353.

18.

Hillier

Immink

Thewlis

Assessing proprioception: a systematic review of possibilities. Neurorehabil Neural Repair 2015; 29(10): 933–949.

19.

van Andel

Wolterbeek

Doorenbosch

et al . Complete 3D kinematics of upper extremity functional tasks. Gait Posture 2008; 27(1): 120–127.

20.

Karduna

McClure

Michener

et al . Dynamic measurements of three-dimensional scapular kinematics: a validation study. J Biomech Eng 2001; 123(2): 184–190.

21.

Cutti

Giovanardi

Rocchi

et al . Ambulatory measurement of shoulder and elbow kinematics through inertial and magnetic sensors. Med Biol Eng Comput 2008; 46(2): 169–178.

22.

Campolo

Formica

Guglielmelli

et al . Kinematic analysis of the human wrist during pointing tasks. Exp Brain Res 2010; 201(3): 561–573.

23.

Parel

Cutti

Fiumana

et al . Ambulatory measurement of the scapulohumeral rhythm: intra- and inter-operator agreement of a protocol based on inertial and magnetic sensors. Gait Posture 2012; 35(4): 636–640.

24.

van den Noort

Wiertsema

Hekman

et al . Reliability and precision of 3D wireless measurement of scapular kinematics. Med Biol Eng Comput 2014; 52(11): 921–931.

25.

Marini

Squeri

Morasso

et al . Robot-aided developmental assessment of wrist proprioception in children. J Neuroeng Rehabil 2017; 14(1): 3.

26.

Xia

et al . Development and validation of a portable human body joint power test system. Int J Distrib Sens Netw 2015; 2015(4): 8.

27.

Pilbeam

Hood-Moore

Test–retest reliability of wrist joint position sense in healthy adults in a clinical setting. Hand Therapy 2018; 23: 100–109.

28.

Adamo

Alexander

Brown

SH.

The influence of age and physical activity on upper limb proprioceptive ability. J Aging Phys Act 2009; 17(3): 272–293.

29.

Feczer

Wrist proprioceptive assessment in expert violin players. PhD Thesis, Minneapolis, MN, 2018.

30.

Adamo

Martin

BJ.

Position sense asymmetry. Exp Brain Res 2009; 192(1): 87–95.

31.

Wonhwee

Ohyun

Chunghwi

et al . Effects of taping on wrist extensor force and joint position reproduction sense of subjects with and without lateral epicondylitis. J Phys Therapy Sci 2011; 23(4): 629–634.

32.

Gay

Harbst

Hansen

et al . Effect of partial wrist denervation on wrist kinesthesia: wrist denervation does not impair proprioception. J Hand Surg Am 2011; 36(11): 1774–1779.

33.

Smitt

Bird

HA.

Measuring and enhancing proprioception in musicians and dancers. Clin Rheumatol 2013; 32(4): 469–473.

34.

Contu

Cappello

Konczak

et al . Preliminary analysis of non-dominant proprioceptive acuity and interlimb asymmetry in the human wrist. Conf Proc IEEE Eng Med Biol Soc 2015; 2015: 3598–3601.

35.

Cappello

Elangovan

Contu

et al . Robot-aided assessment of wrist proprioception. Front Hum Neurosci 2015; 9: 198.

36.

Lubiatowski

Ogrodowicz

Wojtaszek

et al . Measurement of active shoulder proprioception: dedicated system and device. Eur J Orthop Surg Traumatol 2013; 23(2): 177–183.

37.

Lafe

Pacheco

Newell

KM.

Bimanual coordination and the intermittency of visual information in isometric force tracking. Exp Brain Res 2016; 234(7): 2025–2034.

38.

Fitzgibbon

Pilu

Fisher

RB.

Direct least squares fitting of ellipses. Int Conf Pattern Recogn 2002; 251: 253–257.

39.

Suprak

Osternig

van Donkelaar

et al . Shoulder joint position sense improves with elevation angle in a novel, unconstrained task. J Orthop Res 2006; 24(3): 559–568.

40.

Baray

Philippot

Farizon

et al . Assessment of joint position sense deficit, muscular impairment and postural disorder following hemi-Castaing ankle ligamentoplasty. Orthop Traumatol Surg Res 2014; 100(6 Suppl.): S271–S274.

41.

Bang

Shin

Choi

et al . Comparison of the effect of weight-bearing and non-weight-bearing positions on knee position sense in patients with chronic stroke. J Phys Ther Sci 2015; 27(4): 1203–1206.

42.

Duzgun

Kanbur

Baltaci

et al . Effect of Tanner stage on proprioception accuracy. J Foot Ankle Surg 2011; 50(1): 11–15.

43.

Vila-Cha

Riis

Lund

et al . Effect of unaccustomed eccentric exercise on proprioception of the knee in weight and non-weight bearing tasks. J Electromyogr Kinesiol 2011; 21(1): 141–147.

44.

Han

Lee

JH.

Effects of kinesiology taping on repositioning error of the knee joint after quadriceps muscle fatigue. J Phys Ther Sci 2014; 26(6): 921–923.

45.

Knox

Hodges

PW.

Changes in head and neck position affect elbow joint position sense. Exp Brain Res 2005; 165(1): 107–113.

46.

Salgado

Ribeiro

Oliveira

Joint-position sense is altered by football pre-participation warm-up exercise and match induced fatigue. Knee 2015; 22(3): 243–248.

47.

Bennell

Wee

Crossley

et al . Effects of experimentally-induced anterior knee pain on knee joint position sense in healthy individuals. J Orthop Res 2005; 23(1): 46–53.

48.

Lonn

Crenshaw

Djupsjobacka

et al . Position sense testing: influence of starting position and type of displacement. Arch Phys Med Rehabil 2000; 81(5): 592–597.

49.

Vafadar

Cote

Archambault

PS.

Sex differences in the shoulder joint position sense acuity: a cross-sectional study. BMC Musculoskelet Disord 2015; 16: 273.

50.

Ettinger

Shapiro

Karduna

Subacromial anesthetics increase proprioceptive deficit in the shoulder and elbow in patients with subacromial impingement syndrome. Clin Med Insights Arthritis Musculoskelet Disord 2017; 10: 713198.

51.

Mazuchi

FAS

Bigongiari

Francica

et al . Aerobic training in aquatic environment improves the position sense of stroke patients: a randomized clinical trial. Motriz Rev Educ Fis 2018; 24(1): 1590.

52.

Drouin

Valovich-mcLeod

Shultz

et al . Reliability and validity of the Biodex system 3 pro isokinetic dynamometer velocity, torque and position measurements. Eur J Appl Physiol 2004; 91(1): 22–29.

53.

You

SH.

Joint position sense in elderly fallers: a preliminary investigation of the validity and reliability of the SENSERite measure. Arch Phys Med Rehabil 2005; 86(2): 346–352.

54.

Koo

MY.

A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 2016; 15(2): 155–163.

55.

Shrout

Fleiss

JL.

Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979; 86(2): 420–428.

56.

Anderson

Wee

Impaired joint proprioception at higher shoulder elevations in chronic rotator cuff pathology. Arch Phys Med Rehabil 2011; 92(7): 1146–1151.

57.

Phillips

Karduna

No relationship between joint position sense and force sense at the shoulder. J Motor Behav 2017(3): 1–7.

58.

Marshall

Mozrall

Shealy

JE.

The effects of complex wrist and forearm posture on wrist range of motion. Hum Factors 1999; 41(2): 205–213.

59.

Weiler

Awiszus

Differences between motion-direction perception and unspecific motion perception in the human knee joint. Exp Brain Res 2000; 132(4): 523–530.

60.

Hagert

Forsgren

Ljung

BO.

Differences in the presence of mechanoreceptors and nerve structures between wrist ligaments may imply differential roles in wrist stabilization. J Orthop Res 2005; 23(4): 757–763.

61.

Hagert

Garcia-Elias

Forsgren

et al . Immunohistochemical analysis of wrist ligament innervation in relation to their structural composition. J Hand Surg Am 2007; 32(1): 30–36.

62.

Banks

RW.

An allometric analysis of the number of muscle spindles in mammalian skeletal muscles. J Anat 2006; 208(6): 753–768.

63.

Kokkorogiannis

Somatic and intramuscular distribution of muscle spindles and their relation to muscular angiotypes. J Theor Biol 2004; 229(2): 263–280.

64.

Holzbaur

Murray

Gold

et al . Upper limb muscle volumes in adult subjects. J Biomech 2007; 40(4): 742–749.

65.

Goble

DJ.

Proprioceptive acuity assessment via joint position matching: from basic science to general practice. Phys Ther 2010; 90(8): 1176–1184.

66.

Harris

Wolpert

DM.

Signal-dependent noise determines motor planning. Nature 1998; 394(6695): 780–784.

67.

Ferrell

Smith

Position sense at the proximal interphalangeal joint of the human index finger. J Physiol 1988; 399: 49–61.

68.

Clark

Burgess

Chapin

et al . Role of intramuscular receptors in the awareness of limb position. J Neurophysiol 1985; 54(6): 1529–1540.

69.

Edin

BB.

Quantitative analysis of static strain sensitivity in human mechanoreceptors from hairy skin. J Neurophysiol 1992; 67(5): 1105–1113.

70.

Grill

Hallett

Marcus

et al . Disturbances of kinaesthesia in patients with cerebellar disorders. Brain 1994; 117(Pt. 6): 1433–1447.

71.

Proske

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.

Wrist joint proprioceptive acuity assessment using inertial and magnetic measurement systems

Abstract

Keywords

Introduction

Methods

Subjects

Apparatus

Protocol

ROM test

JPS test

Data analysis

Results

Reliability of the JPSS

ROM of the wrist

JPS of the wrist

AE

CE

CE in the ulnar/radial deviation axis

CE in the flexion/extension axis

VE

Discussion

Reliability and precision of the JPSS

JPS in different directions (human wrist JPS ellipse)

JPS in different angular levels

Conclusion

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

References