In Vivo Mechanical Function of the Distal Radial Ulnar Ligaments During Rotation of the Wrist

Abstract

Purpose

The purpose of this study was to investigate changes in length of the volar and dorsal radioulnar ligaments (VRULs and DRULs), and the distal radioulnar joint (DRUJ) space during unweighted and weighted rotation of the wrist using magnetic resonance imaging and biplanar fluoroscopy.

Methods

Fourteen wrists in 7 normal adult volunteers were imaged to define the 3-dimensional geometry of the DRUJ and the insertion sites of the superficial and deep bundles of the VRULs and DRULs. Subjects were imaged at 10 positions of forearm rotation ranging from full pronation to full supination, with or without a 5-pound weight. Lengths of the superficial and deep VRUL and DRUL bundles and DRUJ space were measured (in millimeters) at each position to evaluate ligament function and DRUJ stability.

Results

In the unweighted and weighted trials, maximal elongation of both deep and superficial VRUL bundles occurred in supination and maximal lengths of the deep and superficial DRUL bundles occurred in pronation. Maximum DRUJ space occurred during pronation and a minimum occurred in 30° of supination. In weighted trials, there was a significant increase in deep and superficial VRUL bundle length at positions between 30° of pronation and 30° of supination; however, there was no effect of weight on DRULs length. In weighted trials, there was a significant increase in DRUJ space at positions between full pronation and 15° of supination.

Conclusions

This study demonstrates elongation of the VRULs in supination and the DRULs in pronation. There was no evidence of reciprocal loading of superficial/deep ligament bundles on either the dorsal or the volar aspects of the DRUJ. The effect of loading the wrist during rotation was apparent primarily in the VRULs, but not the DRULs. The DRUJ space was lowest at approximately 30° of supination.

Clinical relevance

These results add information to the literature regarding the complicated biomechanics of the triangular fibrocartilage complex and DRUJ. Future work should evaluate changes in biomechanics caused by triangular fibrocartilage complex tears to determine how tear severity and location relate to clinical symptoms.

Ulnar-sided wrist pain remains a frequent and functionally limiting complaint after both chronic and acute injuries of the wrist. Advances in imaging and arthroscopy have shown ligamentous injury to the triangular fibrocartilage complex (TFCC) to be a major cause of ulnar-sided wrist pain.^1,2 Despite improvements in the understanding of TFCC anatomy and function, there remains debate about the in vivo function of distal radial ulnar joint (DRUJ) ligaments in pronation and supination.^3–7

Anatomically, the TFCC is composed of the discus articularis, dorsal radioulnar ligaments (DRULs), volar radioulnar ligaments (VRULs), volar ulnocarpal ligaments, and extensor carpi ulnaris subsheath.^8,9 The primary functions of the TFCC are transmission of compressive forces across the ulnocarpal joint during axial loading, and stabilization of the DRUJ during wrist rotation.^10–12 The role of the TFCC in axial loading is well described because 16% to 40% of axial loads are transmitted through the TFCC depending on ulnar deviation and variance.^13,14 During wrist rotation, however, the functions of the TFCC and specifically the VRULs and DRULs are less clearly defined because the interosseous membrane and joint capsule also play a role in rotational stability. Cadaveric studies suggest i that both the VRULs and the DRULs are composed of superficial fibers that connect the radial sigmoid notch to the ulnar styloid and deep fibers, constraining the sigmoid notch to the rotational center of the distal forearm, the ulnar fovea.^4,19–21

A consensus has not been established in the literature regarding the relative contributions of the deep and superficial bundles of the DRULs and VRULs during wrist rotation.^5,6,12,22 Some studies report a reciprocal function of these bundles, where in supination the superficial VRULs and deep DRULs are loaded and in pronation the superficial DRULs and deep VRULs are loaded.^7,10 Other studies have reported no reciprocal function of the bundles, and ex vivo cadaveric studies suggest that the majority of DRUJ stability comes from the deep foveal fibers.^5,12 Overall, there remain limited data on the function of these ligaments during in vivo, unconstrained loading conditions.⁷ A better understanding of TFCC and DRUJ function in vivo might refine the diagnosis of joint disease, inform surgical techniques to stabilize traumatic injuries, and ultimately improve patient outcomes.

The purpose of this study was to characterize the deformations of the superficial and deep VRUL and DRUL bundles and to measure changes in DRUJ space during wrist rotation.

MATERIALS AND METHODS

Patient selection

After institutional review board approval, 7 subjects (6 men, 1 woman; age, 24−38 years) were recruited via promotional fliers posted at our medical center and adjacent university campus. subjects consented to participate in the study after a thorough discussion regarding its relevant risks. The exclusion criteria included a history of forearm fracture or injury of the TFCC, DRUJ, or elbow. Participants were imaged bilaterally to yield a total of 14 wrists in the study group. The study was performed at the Department of Orthopaedic surgery, Duke University Medical Center, Durham, NC.

Imaging and data processing

The wrists of all subjects were first imaged using a 3.0-T magnetic resonance imaging (MRI) (TrioTim, Siemens, Germany). Each wrist was imaged separately with the subject supine, the wrist at the subject’s side, and the palm facing down in mild pronation. images were acquired using a dedicated wrist coil (Invivo, Orlando, FL) and a coronal SPACE sequence (flip angle 40°; echo time 15 ms; repetition time 5 ms; slice thickness 0.3 mm). stacks of MRI scans were imported into a solid modeling program (Rhino; McNeel, Seattle, WA), and radius and ulna boundaries were manually segmented to create 3-dimensional models (Fig. 1, top). For each model, the ulnar and radial attachment sites for the DRULs and VRULs were manually segmented and divided into superficial and deep bundles using an approach similar to 2 previous studies^7,23 (Fig. 1, bottom). First, individual coordinate systems for the radius and ulna were defined for each subject. For the radius, the z axis was defined along the longitudinal axis, the y axis was defined perpendicular to the line connecting the radius sigmoid edges, and the x axis was defined perpendicular to the y and z axes in the dorsal direction. For the ulna, the z axis was defined along the longitudinal axis, the y axis was defined through the center of the ulnar styloid, and the x axis was defined perpendicular to the y and z axes in the dorsal direction. To ensure that each coordinate system was oriented consistently on both the left and the right wrist models, the left wrist models were imported to the same workspace as the right, mirrored, and an iterative closest point technique was used to align the left and right radius/ulna pairs. The right radius/ulna coordinate systems were then transferred to the left radius/ulna models. With the axes created, the ligament geometry was defined by visualizing the radial and ulnar insertion sites on the MRI scans and then the insertion boundaries were demarcated by manual segmentation. On the radial side of the DRUJ, common dorsal and common volar attachment sites were created because there was no separation between superficial and deep ligament bundles. On the ulnar side, 4 attachment sites were modeled, 1 for each bundle. The centroids of the 2 radial attachment sites and 4 ulnar attachment sites were then calculated to define the end points each ligament bundle (Fig. 1, right).

FIGURE 1:

Left 3-Dimensional MRI scans were manually segmented to define the geometry of the distal radius and ulna and the insertion sites of the VRULs and DRULs. Middle Coordinate systems for each bone were defined using anatomical references. Ligament insertions were mapped to the bone models. Right After calculating centroids of the ligament insertions, a wire model was generated to define the span of the superficial and deep VRUL and DRUL bundles.

The left and right wrists of each subject were then imaged using 2 orthogonal fluoroscopes (Pulsera; Philips, The Netherlands). To do so, participants were directed to hold their elbow to their side and flexed 90°, to hold their wrist in a neutral position, and to close their hand in a fist without straining (Fig. 2, top). Subjects then rotated each wrist from a maximally supinated to a maximally pronated position and static images were recorded at regular intervals defined by a goniometer. In total, 10 positions were imaged and analyzed for each wrist (the angle of maximum pronation, 45° pronation, 30° pronation, 15° pronation, 0°, 15° supination, 30° supination, 45° supination, 60° supination, the angle of maximum supination). To simulate a load encountered during routine daily activities, subjects repeated this imaging sequence while holding a 5-pound dumbbell asymmetrically (Fig. 2, top).

FIGURE 2:

Top Wrists were positioned within the fields of view of 2 orthogonal fluoroscopes and imaged in either unweighted or weighted conditions. Bottom MRI models were mapped to the fluoroscopic images to define the 3-dimensional positions of the radius and ulna. Representative image in bottom left pronation and bottom right neutral positions. The length of each ligament (blue line) was calculated in these positions.

Model matching and measurements

Volumetric MRI models of the radius and ulna were imported into a digital model of the laboratory and manually manipulated to match 2-dimensional projections of the 3-dimensional MRI models with the 2-dimensional biplanar fluoroscopic images (Fig. 2, bottom). The laboratory model was developed using 3 spatial calibration tools as previously described²⁴: the first to determine the position of the fluoroscopes relative to each other, a second to determine the x-ray beam centers on their opposing intensifiers, and a third to correct the intrinsic image distortion. Within this framework, the straight line distance between the modeled ligament end points and the angle of DRUJ rotation were measured. To enable comparison across subjects at exact angles of DRUJ rotation, radioulnar ligament lengths were calculated by linear interpolation at the 2 nearest DRUJ angle measurements. We also measured the DRUJ space. To do so, points on the articulating face of the radius were selected and matched with their nearest point on the ulna (Fig. 3). The average distance between matched points was calculated to determine DRUJ space at each rotation angle. We have measured the precision of each aspect of this analysis in previous studies (manual segmentation of MRI scans, 0.03 mm,²⁵ 0.04 mm²⁶; manual positioning of MRI models in biplanar fluoroscopy environment, 0.1 mm,²⁷ 0.15 mm²⁸; identifying ligament footprints, 0.3 mm,²⁹ 0.34 mm³⁰).

FIGURE 3:

Identification of the DRUJ space. Points on the radial surface adjacent to the ulna were demarcated and the nearest neighbors on the ulnar surface were identified. DRUJ spacing was defined as the average distance between these points.

Statistical analysis

The VRULs length, DRULs length, and DRUJ space at each angle of wrist rotation were compared by a repeated measures analysis of variance. Fisher least significant difference post hoc test was used to determine whether differences at each position in unweighted and weighted scenarios were statistically significant with P less than .05. All values are reported as mean ± 95% confidence intervals and are included in Table E1 (available on the Journal’s Web site at www.jhandsurg.org). A post hoc power analysis confirmed that β greater than 0.9 for the analysis of VRULs length, DRULs length, and DRUJ space.

TABLE E1.

		Superficial VRULs Length (mm)		Deep VRULs Length (mm)		Superficial DRULs Length (mm)		Deep DRULs Length (mm)		DRUJ Spacing (mm)
	Unweighted	Mean	95% CI	Mean	95% CI	Mean	95% CI	Mean	95% CI	Mean	95% CI
Supination	83	17.17	0.79	15.79	0.53	9.78	0.94	10.93	0.73	3.59	0.50
	60	17.66	0.69	15.81	0.59	11.53	0.73	11.82	0.61	3.24	0.38
	45	17.85	0.68	15.68	0.64	12.56	0.76	12.25	0.65	3.08	0.33
	30	17.77	0.73	15.37	0.69	13.42	0.72	12.61	0.60	2.98	0.27
	15	17.46	0.78	14.91	0.74	14.21	0.72	12.98	0.61	3.02	0.26
	0	17.03	0.88	14.39	0.83	14.99	0.64	13.38	0.56	3.12	0.26
Pronation	−15	16.21	0.93	13.73	0.89	15.17	0.69	13.41	0.57	3.54	0.34
	−30	15.36	0.99	13.04	0.97	15.33	0.78	13.43	0.68	3.95	0.41
	−45	14.07	0.98	12.03	0.95	15.23	0.88	13.28	0.77	4.22	0.41
	−63	11.90	0.88	10.28	0.86	14.80	0.74	12.97	0.76	4.24	0.43
	Weighted	Mean	95% CI	Mean	95% CI	Mean	95% CI	Mean	95% CI	Mean	95% CI
Supination	78	17.35	0.72	15.87	0.66	10.48	1.06	11.39	0.80	3.71	0.39
	60	17.71	0.62	15.83	0.60	11.73	0.82	11.87	0.67	3.45	0.35
	45	18.02	0.67	15.81	0.66	12.71	0.74	12.24	0.59	3.22	0.30
	30	18.26	0.83	15.77	0.83	13.63	0.73	12.59	0.54	3.14	0.28
	15	18.07	0.91	15.43	0.87	14.33	0.79	12.89	0.61	3.18	0.29
	0	17.45	1.00	14.79	0.93	14.92	0.79	13.22	0.64	3.36	0.37
Pronation	−15	16.85	1.02	14.29	0.96	15.20	0.76	13.28	0.64	3.82	0.50
	−30	15.93	0.97	13.53	0.91	15.38	0.73	13.37	0.61	4.31	0.50
	−45	14.39	1.00	12.34	0.95	15.16	0.69	13.20	0.58	4.52	0.41
	−64	12.20	0.98	10.66	1.00	14.48	0.80	12.66	0.73	4.53	0.44

95% CI, 95% confidence interval.

RESULTS

The lengths of the superficial and deep VRUL and DRUL bundles were measured at each angle of DRUJ rotation (Fig. 4). Forearm rotation had a significant effect on the lengths of both the superficial and deep DRUL and VRUL bundles in both the unweighted and the weighted conditions (P < .05). In unweighted trials, the superficial and deep VRUL bundles followed similar trends in elongation, with a minimum length in full pronation (superficial, 11.9 ± 0.5 mm; deep, 10.3 ± 0.4 mm) and a maximum length in supination (superficial at 45°, 17.9 ± 0.4 mm; deep at 60°, 15.8 ± 0.3 mm). The DRUL bundles exhibited a reciprocal behavior to the VRUL bundles, with maximum length in 30° pronation (superficial, 15.3 ± 0.4 mm; deep, 13.4 ± 0.4 mm) and minimum length in full supination (superficial, 9.8 ± 0.5 mm; deep: 10.9 ± 0.4 mm). In the weighted trials, the superficial VRUL bundle was elongated at positions between the 45° pronation and the 30° supination in comparison to the unweighted trials (average elongation, 0.5 mm), and the deep bundle was elongated at positions between 30° pronation and 30° supination (average elongation, 0.5 mm). Conversely, the superficial and deep DRULs lengths measured during the weighted trials were not significantly different from the unweighted condition at any angle of rotation.

FIGURE 4:

Lengths of superficial and deep left VRUL and right DRUL bundles in weighted and unweighted conditions. All data points are represented as mean ± 95% confidence interval; +, P < .1 and *, P < .05.

Rotation of the wrist caused the DRUJ space to increase in both pronation and supination, for both the unweighted and the weighted trials (Fig. 5). From the neutral position to the maximum pronation position, joint space increased, where it reached a maximum in full pronation (4.2 ± 0.2 mm). From the neutral position to the maximum supination position, joint space decreased until approximately 30° of supination where it reached a minimum (3.1 ± 0.2 mm), and then increased from 30° of supination to the maximum supination angle (at maximum supination, 3.6 ± 0.3 mm). In comparison with the unweighted trials, the measured DRUJ space during the weighted trials was significantly greater at angles between 45° pronation and 15° supination (average difference, 0.3 mm). The maximum difference in DRUJ space occurred at 30° of pronation (0.4 mm).

FIGURE 5:

DRUJ space during unweighted and weighted trials. All data points are represented as mean ± 95% confidence interval; +, P < .1 and *, P < .05.

DISCUSSION

Ulnar-sided wrist pain remains a challenging problem to diagnose and treat despite improvements in both open and minimally invasive arthroscopic techniques. This is in part due to differing opinions regarding the biomechanical role of the specific portions of the TFCC.^1,2,10,31,32 In this study, MRI, 3-dimensional joint modeling, and biplanar fluoroscopy were used to analyze in vivo VRULs and DRULs deformations during unweighted and weighted rotation of the wrist. The superficial and deep bundles of the VRULs reached maximum length at mid-supination in unweighted and weighted trials, whereas the superficial and deep bundles of the DRULs reached maximum length at mid-pronation in both trials. Because this study is one of few to evaluate in vivo deformations of the TFCC, it contributes valuable information on the ligamentous constraints during unconstrained rotation of the DRUJ.⁷

We defined a reciprocal function of the VRULs and DRULs, where during supination, both bundles of the VRULs are maximally elongated whereas both bundles of the DRULs are minimally elongated, and in pronation, both bundles of the VRULs are minimally elongated whereas both bundles of the DRULs are maximally elongated (Fig. 6). These findings are in conflict with the concept of reciprocal loading of deep and superficial bundles of the VRULs and DRULs described by Hagert in 1994,¹¹ who hypothesized that in supination the superficial VRULs and deep DRULs are taut, whereas in pronation the superficial DRULs and deep VRULs ligaments are taut. This concept was promoted by others, and was further elaborated upon, suggesting that superficial fibers resist extremes of rotation whereas deep bundles resist subluxation.^10,32,33 The findings of Stuart et al¹⁵ in cadaver experiments supported that the deep bundles prevent subluxation of the DRUJ. Here, however, normal rotation of the wrist did not cause sufficient translation at the DRUJ to necessitate support of the deep bundles. That our results are in conflict with the ex vivo work of Stuart et al¹⁵ suggests that passive and active muscular forces present in vivo may contribute to DRUJ stability and affect the function of the radioulnar ligaments. Furthermore, as ex vivo studies demonstrated that the superficial and deep bundles of the VRULs and DRULs are loaded similarly (not reciprocally) in pronation and supination, ex vivo mechanical conditions must be carefully controlled to properly mimic physiological loading.^6,34,35 The trends and magnitudes of deformation described here for the superficial VRULs and DRULs confirm the in vivo work of Xu and Tang.⁷ Contrary to our findings, however, Xu and Tang⁷ measured reciprocal motions in the superficial and deep bundles. In their study, the ligament attachment sites were approximated using published histological results, whereas with our MRI method, we can directly identify ligamentous insertions on an individual-by-individual basis and account for biological variability between volunteers. This may account for the discrepancy between the results.

FIGURE 6:

VRULs and DRULs tension at full supination, neutral, and full pronation positions (in reference to the minimum length of each ligament bundle measured throughout the full range of wrist motion). Superficial and deep bundles of the VRULs and DRULs have similar, not reciprocal, patterns of elongation throughout wrist rotation.

The effect of loading the wrist during rotation was apparent primarily in the VRULs and observed changes in DRUJ space. There were no differences in the overall shape of the length versus rotation angle curve between unweighted and weighted trials for either the VRULs or the DRULs. However, when loading the wrist, there was a significant increase in length of both VRUL bundles between 30° of pronation and 30° of supination that was not apparent in the DRULs. Other stabilizers of the DRUJ, such as the joint capsule or ulnocarpal collateral ligament, may also be resisting motions of the radius and ulna that increase dorsal DRUJ space, specifically in pronation when the DRUJ space is at its maximum. Contributions from the palmar osteocartilaginous lip, which may be present in up to 80% of individuals, may also affect volar translation of the distal ulna at positions beyond 30° of supination and, thus, have an effect on observed DRUJ space.³⁶ It has previously been observed that greater dorsopalmar translation exists during maximum isometric pronation than in maximum isometric supination (3.1 mm dorsal and 2.2 mm palmar, respectively).³⁷ Similarly, in vivo results put forth by Hojo et al³ demonstrate that loading of the wrist in a position of maximal pronation and hyperextension, as may be seen at the time of a fall onto an outstretched hand, results in an increase in palmar translation of the ulna compared with unweighted trials. These authors additionally noted significant elongation of both the superficial and the deep VRULs during weighted trials, with no change in either the superficial or the deep DRULs bundles, which is consistent with our findings with respect to weighted wrist rotation. Based on the results observed here, this may indicate that as the wrist moves from relative pronation to supination, the VRULs continue to elongate until maximal engagement of the ulnar head on the osteocartilagenous lip occurs. Following this maximal engagement of the ulnar head on the osteocartilagenous lip, the VRULs are no longer conferring a significant restraint to palmar translation of the ulnar head relative to that of the osteofibrocartilagenous lip and, therefore, do not further elongate. As the forearm rotation continues from 30° of supination to maximal supination, the ulnar head may remain relatively palmar while the radius continues to translate resulting in an increase in DRUJ space in maximal supination as was observed here.

In the setting of suspected DRUJ injuries with instability on examination, many patients are immobilized in near-full supination because this has been hypothesized to fully seat the ulna in the sigmoid notch of the radius and confer maximal stability to the DRUJ. However, we have demonstrated that DRUJ space is minimized at approximately 30° of supination. Because minimal joint space may be the position of maximal joint stability, immobilization of the wrist in 30° of supination following injury may confer improved stability. Further investigation is warranted to determine whether this biomechanical finding translates to a difference in clinical outcomes. Changes in mean DRUJ spacing may be attributed to gapping in one aspect of the articulation, while another aspect of the articulation remains in contact. In addition, cartilage compression, particularly in supination where the minimum DRUJ spacing was measured, may contribute to an overall decrease in joint width. These remain areas for future work.

Because muscle forces and geometric constraints likely cause stress in ligaments throughout the full range of wrist motion, it was not possible to measure the stress-free length of the VRULs and DRULs to estimate the tensile load in each. Evaluating reciprocal function as defined by the tensile forces in each ligament remains an area for future investigation. Furthermore, we measured the straight line distance between ligament insertions to determine ligament length. This provides an estimate of the true ligament length in positions when ligaments likely wrap around the radial aspect of the ulna. This represents a limitation, but likely affects only the magnitude of ligament length, not the overall trends in ligament elongation throughout wrist rotation. Although patients with a history of forearm fracture and injury of the TFCC, DRUJ, or elbow were excluded from the study, there was no formal prestudy physical examination to rule out any other subtle upper extremity pathology that may have affected the results of the study protocol completed by each participant. In addition, formal Beighton scoring for each participant was not performed prior to study inclusion. Although relatively uncommon, the presence of a connective tissue disorder may have had an impact on the results of the study performed here and future investigations should seek to control for these conditions.

In conclusion, this study presents an in vivo analysis of the DRUJ ligaments, directly measuring VRULs and DRULs length through the full range of forearm rotation. Our results suggest that both the superficial and the deep VRUL bundles are maximally elongated in supination whereas both bundles of the DRULs are maximally elongated in pronation. These findings conflict with the concept of the reciprocal function of the superficial and deep bundles. Loading of the wrist during rotation caused elongation of both VRUL bundles in pronation and supination, but did not affect the dorsal bundles. i Furthermore, wrist loading increased the DRUJ space primarily in pronation, and only mildly in supination.