Median nerve travel and deformation in the transverse carpal tunnel increases with chuck grip force and deviated wrist position

Participants

Fourteen healthy right-hand dominant participants (7 men and 7 women) completed this study (Table 1), which was approved by the Nipissing University Research Ethics Board (File #101506). All participants provided written informed consent before completing a health screening questionnaire. Exclusionary criteria included diabetes mellitus, thyroid condition, gout, amyloidosis or sarcoidosis, renal failure or hemodialysis, degenerative joint disease, arthritis of the wrist/hand, corticosteroid injection, Colles fracture, radial malunion, peripheral neuropathy, carpal tunnel syndrome, flexor tendinopathy, wrist/hand surgery or injuries, or symptoms of pain, tingling, or numbness of the wrist/hand in the past year.

Experimental setup

Each participant was seated with their right arm supinated in a custom testing apparatus (Fig. 1), which was constructed from slotted aluminum extrusion and linear motion components to provide participant specific adjustability (80/20 Inc., Columbia City, IN). The apparatus included a padded splint to support the forearm. At the level of the metacarpal bones, two adjustable supports fixed the hand to maintain the wrist in either radial or ulnar deviation (depending on the experimental condition). The ultrasound probe was fixed in place with a custom holding device consisting of a locking ball-and-socket joint and three-prong clamp. The probe was positioned parallel to the distal wrist crease, at the base of the thenar eminence with a generous amount of gel to optimize acoustic coupling and minimize contact pressure during ultrasound imaging. Participants interacted with a custom chuck grip dynamometer, which consisted of a single-axis load cell (MLP-75, Transducer Techniques, Temecula, CA) and two pushbuttons attached on each side (1 for the thumb and 1 for the opposing index and middle fingers; 1.13 cm² and 5.73 cm², respectively) with a fixed grip span of 4.5 cm (Fig. 1).

Experimental protocol

Participants were asked to perform a chuck grip, which involved placing the thumb on the smaller push button of the custom dynamometer and placing the index and long fingers on the larger push button of the dynamometer in opposition to the thumb (Fig. 1). Participants performed the chuck grip in both a low force and a high force condition, set relative to their individual strength at 10% and 40% of their maximum voluntary effort (MVE). A chuck grip at low and high force levels was selected since this grip type is well documented within field-based studies in ergonomics (Gerhardsson & Hagberg, 2014; MacDonald & Keir, 2018). Also, previous studies have observed changes when chuck gripping forcefully (25% and 50% compared to 0% MVE) with the wrist in 30° flexion (Cowley et al., 2017). We selected 10% MVE to represent the low force condition since the task required the participants to hold the chuck grip dynamometer and it would have been difficult to maintain the chuck grip with absolutely no force (i.e., 0% MVE). 40% MVE was selected as the forceful condition with careful consideration for the loss of mechanical advantage when the wrist was deviated (Li, 2002; Vanswearingen, 1983). The chuck gripping task was performed in a neutral wrist position (0° neutral) as well as 2 deviated wrist positions, including 15° radial deviation and 30° ulnar deviation (Fig. 2). The deviated wrist positions were confirmed with a handheld goniometer on the palmar side with the forearm in supination. Specifically, the fixed arm of the goniometer lined up in the middle of the forearm with the fulcrum positioned in the middle of the radial and ulnar styloid processes, while the moveable arm was aligned with the third metacarpal to measure the deviated wrist position. The custom supports on the testing apparatus were locked in place to ensure deviated wrist positions were maintained throughout the experimental gripping trials. Since this study focused on investigating the effects of radioulnar deviated wrist positions, the wrist flexion-extension angle was maintained at 0° throughout the testing protocol in all experimental conditions.

Maximum Voluntary Efforts (MVEs)

Before the experimental trials, participants performed 3 chuck grip MVEs using the grip dynamometer with the forearm supinated and wrist in a neutral position (0° radioulnar deviation; 0° flexion-extension). Participants were provided with real-time force feedback on a visual display monitor and the researchers also gave verbal encouragement to elicit maximal responses. MVEs were held for approximately 3–5 s each; 60 s of rest was provided between each successive MVE to minimize neuromuscular fatigue. Maximum chuck grip force was computed as the mean of ±50 ms surrounding the peak during each trial. An average of all 3 MVEs was computed in order to scale the chuck gripping task during the experimental trials (10% or 40% MVE).

Chuck gripping trials

During the experimental chuck gripping trials, participants were asked to complete a force matching task with real-time visual force feedback provided in a custom program (LabVIEW 2014, National Instruments, Austin, TX). Specifically, participants were asked to match a trapezoidal force matching template, which consisted of ramping force up for 2 s, holding the force level (plateau) for 3 s, and ramping force down for 2 s. The force matching template was completed 3 times in succession with 3 s of rest between each trapezoidal profile (Fig. 3). The force ramps were completed at 10% and 40% of MVE within each of the 3 static wrist deviation positions, for a total of 18 static chuck gripping trials (2 chuck grip forces × 3 wrist deviation positions × 3 trials). Prior to the experimental trials, participants were provided with at least 2 practice trials at both force levels with the opportunity for additional practice if needed. The 3 wrist deviation positions were block randomized, and the 2 force levels were counter-balanced within each wrist deviation position to mitigate any potential fatigue or learning effects during the experimental trials.

Data collection

30-second cine-loops of the transverse carpal tunnel were captured with a linear array ultrasound scanner (L7, Clarius Mobile Health, Burnaby, BC). Carpal tunnel structures identified before collecting each trial included the MN and the lunate carpal bone. The MN was easily identifiable by its hypoechoic area and surrounding hyperechoic boundary near the border of the transverse carpal ligament (Fig. 4). The lunate was located dorsal to the MN and was identifiable by its highly hyperechoic appearance and morphology. All cine-loops were captured in the musculoskeletal (“MSK”) mode to maximize the acquisition frequency of 12 MHz at a scanning depth of 15 mm. Each cine-loop was recorded at 30 frames per second for a total of 900 frames per recording. Both MVE and experimental gripping forces were collected in the custom LabVIEW program. Grip forces were amplified and digitally sampled via a USB high speed analog-to-digital converter at 1,000 Hz (USB-2533, Measurement Computing, Norton, MA). Grip force and ultrasound collections were initiated simultaneously for synchronization.

Data analysis

Images were extracted from the captured cine-loops at the mid-points of the force plateaus from each set of the 3 successive force ramp profiles performed for each chuck gripping condition, corresponding to time points 6.5, 16.5, and 26.5 s within the 30-second recordings (Fig. 3). Images were also extracted at the mid-points of the rest periods that preceded each chuck gripping ramp (corresponding to time points 1.5, 11.5, and 21.5 s). The extracted static images of the transverse carpal tunnel were imported into ImageJ (1.52a, National Institutes of Health, Bethesda, MD) to analyze travel and deformation of the MN. The polygon selection tool within ImageJ was used to trace the inner hyperechoic border of the MN (Alemán et al., 2008; Filius et al., 2013). The multi-point tool was also used to determine the coordinate of the most anterior peak of the lunate bone. MN deformation was assessed during each gripping trial, which included the medial-lateral (X) width and anterior-posterior (Y) height of the bounding box that enclosed the nerve, cross-sectional area (CSA), and circularity (Eq. (1)): (1)${\displaystyle C=\frac{4\pi A}{P^2}}$ where, C is the dimensionless circularity coefficient ranging from 0 to 1 (1 = perfect circle);

A = area (mm²);

p = perimeter (mm).

MN position in both the X and Y directions were expressed relative to the lunate bone in order to account for any potential minor shifts of the ultrasound probe on the wrist during the 30-second cine-loop recordings. MN position was defined by the centroid, which is the average of the X and Y coordinates of all pixels in the traced nerve. X and Y travel was then calculated as the difference between the position at the force gripping plateau relative to the preceding rest period. Since we noted that MN displacement patterns varied considerably between participants, the cumulative travel (i.e., absolute displacement) was calculated to describe the movement. For each participant, the 3 trials completed within each experimental condition were averaged together to characterize mean travel and deformation metrics, which were submitted for inferential testing. For completeness, movement patterns (i.e., ulnar versus radial displacement and palmar versus dorsal displacement) were also categorized for each condition in chi-squared tables.

All grip force data were de-biased, calibrated, and smoothed with a low-pass Butterworth filter at 30 Hz in MatLab (R2017a, Mathworks, Natick, MA). Maximum chuck grip forces from the MVE trials were computed as the ± 50 ms surrounding the peak force. The force levels corresponding to the extracted images at each force plateau (i.e., at 6.5, 16.5, and 26.5 s) were also computed to ensure that the participants achieved the 10% and 40% MVE target forces during the chuck gripping trials. As with the ultrasound metrics, the chuck grip forces of the 3 experimental trials performed within each condition were averaged together and submitted to inferential testing.

Statistical analyses

Two-way repeated measures analysis of variance (ANOVA) statistical models were used to test effects of force level (10% and 40% of MVE) and wrist deviation position (15° radial deviation, 0° neutral, and 30° ulnar deviation) on the grip force and all MN metrics (X width, Y height, cross-sectional area, circularity, X medial-lateral travel, and Y anterior-posterior travel). A separate two-way ANOVA was run for each metric. Significant main effects and interactions were further analyzed with Tukey’s Honest Significant Difference test to control familywise error (alpha set at 0.05).

Chuck grip forces during Maximum Voluntary Efforts (MVEs)

The maximum chuck grip force elicited over 3 consecutive MVE trials was a mean (±95% confidence interval) of 94.20 ± 11.47 N (Table 2). The coefficient of variation over the 3 MVE chuck gripping trials was 4.9 ± 1.4%, demonstrating good trial-to-trial repeatability. Participants successfully followed the force level templates during the experimental trials; the average chuck grip force at the midpoint of the plateaus, where ultrasound images were extracted for analysis, for the 10% MVE condition was 10.1 ± 0.2% and for the 40% MVE condition was 39.1% ± 0.4%. Not surprisingly, there was a main effect of force level (F_1,13 = 52009.10, p < 0.01), with significantly higher chuck grip forces in the 40% versus 10% MVE experimental trials. However, there was no effect of wrist deviation position on chuck grip forces during the experimental trials (F_2,26 = 1.81, p = 0.18).

MN cross-sectional area and deformation during chuck gripping

There was a significant wrist deviation position by force level interaction on MN CSA (F_2,26 = 3.75, p = 0.037). CSA was lower in the 40% vs 10% MVE chuck gripping condition when the wrist was in 30° ulnar deviation; however, there were no changes in 0° neutral and 15° radial deviation (Fig. 5; Table 3). We also found a wrist deviation position by force level interaction on circularity of the MN (F_2,26 = 3.77, p = 0.04). Circularity increased with greater chuck grip force, but only in the 0° neutral wrist position; while there was a similar trend with chuck grip force in 30° ulnar deviation, it did not achieve significance (Fig. 6). Although there was no significant interaction, there was a main effect of force level on MN width in the medial-lateral axis ( F_1,13 = 15.392, p < 0.01). MN width was smaller when chuck gripping at 40% MVE (5.26 ± 0.39 mm) compared to 10% MVE (5.65 ± 0.43 mm). Conversely, there were no significant differences on height of the MN in the anterior-posterior axis.

MN travel during chuck gripping

Although there were no significant interactions involving MN travel, there was a main effect of wrist deviation position in the medial-lateral axis (F_2,26 = 6.81, p = 0.004), with increased travel in 30° ulnar and 15° radial deviation than 0° neutral (Fig. 7). There was also a main effect of force level (F_1,13 = 4.75, p = 0.048), due to increased travel in 40% MVE compared to 10% MVE (Fig. 7). In the anterior-posterior axis, there was only a main effect of wrist deviation position on MN travel (F_2,26 = 8.37, p = 0.002). Greater MN travel occurred in 30° ulnar deviation relative to 15° radial deviation and 0° neutral (Fig. 8).