Median Nerve Deformation and Displacement in the Carpal Tunnel during Index Finger and Thumb Motion

Abstract

The purpose of this study was to investigate the deformation and displacement of the normal median nerve in the carpal tunnel during index finger and thumb motion, using ultrasound. Thirty wrists from 15 asymptomatic volunteers were evaluated. Cross-sectional images during motion from full extension to flexion of the index finger and thumb were recorded. On the initial and final frames, the median nerve, flexor pollicis longus (FPL), and index finger flexor digitorum superficialis (FDS) tendons were outlined. Coordinate data were recorded and median nerve cross-sectional area, perimeter, aspect ratio of the minimal enclosing rectangle, and circularity in extension and flexion positions were calculated. During index finger flexion, the tendon moves volarly while the nerve moves radially. With thumb flexion, the tendon moves volarly, but the median nerve moves towards the ulnar side. In both motions, the area and perimeter of the median nerve in flexion were smaller than in extension. Thus, during index finger or thumb flexion, the median nerve in a healthy human subject shifts away from the index finger FDS and FPL tendons while being compressed between the tendons and the flexor retinaculum in the carpal tunnel. We are planning to compare these data with measurements in patients with carpal tunnel syndrome and believe that these parameters may be useful tools for the assessment of CTS and carpal tunnel mechanics with ultrasound in the future.

INTRODUCTION

The carpal tunnel contains nine flexor tendons and the median nerve. These structures are surrounded by the subsynovial connective tissue (SSCT), which functions as a sliding interface among these structures [1]. The major pathological finding in carpal tunnel syndrome (CTS) is fibrosis of the SSCT, which changes the motion characteristics of the SSCT, tendon excursion and median nerve, as noted during intraoperative inspection in cases of carpal tunnel release [1–4]. These changes may also cause elevated strain and pressure in the carpal tunnel, which ultimately can lead to CTS [2, 5].

We hypothesize that, due to fibrosis of the SSCT, the kinematics of the nerve and tendons in the carpal tunnel change in patients with CTS. We further hypothesize that these changes are associated with the evolution of CTS and that these changes can be monitored non-invasively by ultrasound. A first step in testing our hypotheses is to identify the normal motion pattern of the tendons and the median nerve in the carpal tunnel. These data can then be used as a baseline against which to compare CTS patients’ data. If, as we hypothesize, detectable differences exist in the SSCT and tendon and nerve kinematics in patients with CTS, then these differences could be sought in individuals at risk for CTS. If our hypotheses are supported, then ultrasound could be a useful non-invasive tool to study the genesis of CTS and to monitor at risk individuals.

Ultrasonography is a good imaging technique for the structures in the carpal tunnel. Several parameters within the carpal tunnel have been assessed using this technology, both clinically and in cadaver models [6–10]. Most studies focused on the longitudinal motion of the tendons and the median nerve. Although the ulnar-radial and dorsal-palmar movement of the median nerve were assessed, tendon movement in these directions has not been studied in depth [11–14]. The carpal tunnel is a 3D structure and ultimately 3D motion over time (i.e., 4D motion analysis) will be necessary to understand the kinematics within the tunnel. Recent research from our laboratory evaluated the transverse motion of the middle finger flexor digitorum superficialis (FDS) tendon, because it is superficial and positioned next to the median nerve in the carpal tunnel [15], thus facilitating image capture. The index finger and thumb, however, are most commonly used in activities like pinch, which can be impaired in patients with CTS [16]. Even though the index finger flexor tendons and the flexor pollicis longus (FPL) tendon are anatomically further away from the median nerve than the middle finger flexor tendon, they are directly posterior to the nerve. We believe that it would be useful to know how their motion normally affects the deformation and motion direction of the median nerve.

The purpose of this study was, therefore, to investigate the motion direction and deformation of the normal median nerve, the index finger FDS and FPL tendons in index finger and thumb movement using ultrasound.

METHODS

This study was approved by our Institutional Review Board. We recruited 15 healthy volunteers (9 men, 6 women, mean age = 36.3 ± 6.9 yrs). Participants were excluded if they had a history of wrist trauma, wrist surgery, or any symptoms related to, or which could mimic, CTS. After written consent was obtained, we proceeded with the ultrasound on both wrists.

The image acquisition procedure of the cross-sectional plane of the carpal tunnel was described previously [15]. In brief, the subjects were lying supine with their arm outstretched, and their lower arm and wrist fixed on a custom-made device. Image acquisition was done at a frame rate of 30Hz, using an Acuson Sequoia C512 ultrasound machine (Siemens Medical Solutions, Malvern, PA) with a 15L8 linear array transducer (Fig. 1). The resolution of the monitor is 640×480pixels (NTSC format). With this combination of transducer and monitor, the pixel size is ~0.04 mm². The transducer was held in a 90° angle to the wrist, without applying extra pressure to avoid compression of the carpal tunnel. Images were acquired at the wrist crease level, with the transducer parallel to the wrist crease. The participant was asked to fully flex and extend the index finger or the thumb to the sound of a metronome at a pace of 0.8Hz for half a cycle (flexion or extension). After a period of practice, 5 cycles of the flexion-extension motion were recorded. We found in initial analyses that during finger or thumb extension, the nerve and tendons were furthest away from each other compared to their position in flexion, so we choose to measure the full extension and full flexion positions of both.

Figure 1.

Custom-made table with transducer holder.

After obtaining the data, the images were evaluated using Analyze 8.1 software (Mayo Clinic, Rochester, MN). After the initial and final frames of the motion were selected and reviewed, the median nerve and the tendon were outlined (Fig. 2). Area, perimeter, displacement, circularity, and the aspect ratio of a minimum enclosing rectangle were measured by the software. The averages of the 5 cycles were calculated. The minimum enclosing rectangle was determined as the smallest possible enclosing rectangle to the image. The aspect ratio was defined as the ratio of the minor axis divided by the major axis of this rectangle. In addition, the deformation ratio for each parameter was calculated by dividing the flexion by the extension value. This ratio gives an indication of the deformation of the median nerve during full extension to flexion motion.

Figure 2.

Cross sectional ultrasound image with outlined median nerve and index finger FDS and FPL tendons.

The displacement of the nerve was defined as the difference between the centroid coordinates of the extension and flexion positions. This way, the displacement in ulnar-radial and palmar-dorsal direction could be calculated. The palmar and ulnar directions were defined as positive and the radial and dorsal as negative.

Statistical Analysis

All results were expressed in mean ± standard deviation. Since we evaluated flexion and extension in both left and right wrists, we used a mixed model approach where participants were treated as a repeated factor, wrists (left and right) as a random effect factor, and fingers (thumb and index) or motion direction (flexion/extension) as a fixed effect factor. A p-value of <0.05 was considered significant. The reliability of the 5 measurements was estimated by Intraclass correlation coefficient (ICC), and an ICC>0.75 was rated excellent. All statistical analyses were performed by SAS version 9.1.3 software (SAS institute Inc., Cary, NC).

RESULTS

We did no differences between left and right hands (p=0.96), within cycles (p=0.37), or among patients (p=0.98) for a worst case series of calculated ratios. Therefore, we used the average for further statistical analysis. Using the same method for a series of measurements of absolute data, we again found no difference between cycles (p=0.33) or left and right hands (p=0.17). We did find a difference among patients (p<0.0001), however, using a mixed model with patients as a repeated factor, this difference disappeared.

Figures 3 and 4 show the results of the direction and amount of movement of the median nerve and the tendons over the course of a full excursion, from full flexion to full extension. In the palmar-dorsal direction, the median nerve moved 0.041mm dorsally with index finger motion and 0.047mm volarly with thumb motion. The motion of the nerve in the palmar-dorsal direction was not significantly different in index FDS or FPL motion. From full extension to full flexion, the FPL moved more volarly (0.49 ± 0.44 mm) than did the index FDS (0.03 ± 0.97 mm) (p<0.05). The radial-ulnar direction of the tendon was not significantly different comparing the index FDS and the FPL, but the median nerve moved differently depending on which tendon was moving, in an ulnar direction with FPL motion (0.23 ± 0.43 mm) and in radial direction with index FDS motion (0.36 ± 1.08 mm) (p<0.05).

Figure 3.

Radial-ulnar motion direction and radial-ulnar nerve-tendon distance.

Figure 4.

Palmar-dorsal motion direction and palmar-dorsal nerve-tendon distance.

The distance between the nerve and the tendon in the radial-ulnar direction was significantly smaller with index FDS motion than with FPL motion (p<0.0001). The palmar-dorsal movement of the index FDS and FPL was not different, but the FPL was significantly closer to the nerve in flexion than it was in extension (p=0.0006). The median nerve parameter indices and deformation ratios are shown in Table 1.

Table 1.

Median Nerve Indices and Deformation Ratios

MEDIAN NERVE INDICES		Extension Average (SD)	Flexion Average (SD)	DEFORMATION RATIOS (SD)
Area (mm2)	Index	9.93 (1.56) *	9.55 (1.58)	0.961 (0.044)
	FPL	10.11(1.42) *	9.62(1.30)	0.955 (0.067)
Perimeter (mm)	Index	14.82(1.83) *	14.61(2.00)	0.985 (0.038)
	FPL	15.04(1.30) *	14.72(1.21)	0.980 (0.042)
Aspect Ratio	Index	0.37(0.08)	0.39(0.11)	1.060 (0.169)
	FPL	0.36(0.08)	0.37(0.09)	1.015 (0.102)
Circularity	Index	1.78(0.26)	1.81(0.31)	1.013 (0.078)
	FPL	1.81(0.28)	1.83(0.33)	1.008 (0.064)

^*p<0.05

No differences were found between the index FDS and FPL measurements. With respect to the flexion and extension positions within these motions, however, the cross-sectional area and the perimeter of the median nerve were larger in extension compared to flexion. Aspect ratio of the minimal enclosing rectangle and circularity were not significantly different, nor were the deformation ratios of the four parameters.

Tthe ICC from a worst case series of measurements was 0.812 (95% confidence interval = 0.60 to 0.93). We expect the ICC to be similar or even better for all measurements.

DISCUSSION

We showed that with the individual index finger and thumb motion, the tendons move towards the median nerve, thereby pushing the median nerve in either a radial or ulnar direction. As the FPL contracts, causing thumb interphalangeal flexion, its motion is in the palmar direction towards the median nerve, thereby compressing the nerve and pushing it in the ulnar direction. During index finger motion the tendon also moves in a palmar direction, while the nerve moves radially. The area of the median nerve is smaller in flexion than in extension with both index finger and thumb movements, which suggests that compression of the median nerve occurs between the tendons and the flexor retinaculum during these single digit motions. There was also no difference in the aspect ratio of the minimal enclosing rectangle and the circularity, indicating that the shape of the median nerve does not change. In longitudinal ultrasound images taken during the same exam, we did not see longitudinal motion of the median nerve in a proximal or distal direction. We do not, therefore. believe that we scanned a different part of the median nerve in extension than we did in flexion.

Our results show that in healthy subjects the median nerve not only undergoes compression during index finger and thumb motions, but can also ‘escape’ the most severe compression, because it can move from side to side to avoid the most direct contact with the underlying tendons. In CTS, the nerve is often constrained by SSCT fibrosis to the overlying flexor retinaculum and underlying tendons, making it liable to even greater compression.

Yoshii et al. showed that isolated motion of the middle finger affects median nerve deformation more than fist motion. They also found that in isolated middle finger motion, the FDS moved in the radial and dorsal direction, while in fist motion it moved in the ulnar and palmar direction. These findings are consistent with our findings and suggest that different hand activities, for example pinch versus grip, might quite differently affect median nerve compression, and thus might be helpful in better understanding the etiology of CTS, a condition that is extremely common and most often idiopathic [17,18].

While tendon motion in the carpal tunnel has not been commonly studied, several authors studied transverse displacement of the median nerve [13,19,20]. Nakamichi and Tachibana observed a transverse sliding of the median nerve beneath the flexor retinaculum in passive motion of the proximal and distal interphalangeal joints of the index finger [21]. They found that the median nerve slides 1.75 ± 0.49 mm in the ulnar direction. This is inconsistent with our findings. The difference probably relates to the fact that they measured passive flexion of the proximal and distal interphalangeal joints, while our subjects actively flexed the PIP, DIP, and metacarpophalangeal joints.

The strength of our study is the active in vivo measurements of the motion direction of the median nerve and the index finger and FPL. We believe that with these results, we have shown that in single digit motion of either the index finger or the thumb, compression of the median nerve occurs. Fibrosis of the surrounding SSCT might result in even more compression of the median nerve.

Our study also has several shortcomings. First, we did not evaluate intra- and interobserver differences. Ultrasound is known for having a significant variation among examiners. However, we found several ways to avoid differences in imaging. The subject’s arm and hand were tied to the custom-made table to prevent motion (Fig. 1). The transducer was held in place with an adjustable arm, allowing the examiner to focus on image acquisition. Finally, our study has a small sample size. But based on previous research, we believe that these results give a good indication of the motion direction and deformation of the carpal tunnel contents and were quite consistent among subjects.

Our results may be useful as baseline data for future studies of the motion direction and deformation of the carpal tunnel contents in both healthy human subjects and in patients with CTS. Based on our data, which shows that our measurement methods are feasible, we plan a future study to compare these data to measurements in patients with CTS.