Validating Clinical Distal Radioulnar Joint Examination With Radiographic Parameters

Abstract

Background:

The purpose of this study was to quantify the in vivo displacement of bilateral distal radioulnar joints (DRUJs) in resisted pronosupination. We hypothesize that this will demonstrate no appreciable difference between the left and right DRUJ, thus validating the concept of using the uninjured wrist as a control for physical examination as well as dynamic imaging studies.

Methods:

Thirty-two participants without a history of wrist pathology underwent a dynamic computed tomography (CT) protocol evaluating bilateral forearm rotation in neutral forearm rotation, 60° pronation, and 60° supination, including maximal isometric muscle loading. The DRUJ alignment, specifically the absolute degree and direction of subluxation of the ulna relative to the sigmoid notch, was then assessed using a modification of the radioulnar line method.

Results:

There was no significant difference in the mean displacement when comparing the right and left sides in neutral, pronation, or supination. The mean displacement was also compared between male and female patients, and there was no statistically significant difference in absolute displacement in neutral (male 0.99 mm vs female 1.38 mm) or supination (male −0.57 mm vs female −0.23 mm). However, the difference in pronation was statistically significant (male 2.69 mm vs female 3.26 mm). Of the 192 sequences, the measurements of displacement of the authors were within 1 mm 86% of the time (166 of 192).

Conclusions:

Dynamic CT of bilateral DRUJs in resisted pronation, supination, and neutral demonstrated symmetry between the right and left DRUJ, supporting the concept of using the contralateral side as a control to identify instability in an injured wrist.

Introduction

The distal radioulnar joint (DRUJ) is critical for forearm and wrist function in load bearing and pronosupination. Symptomatic DRUJ pathology may result from derangements to several structures that include the orientation of the articulating surfaces, the cartilage of the DRUJ, the radioulnar or ulnocarpal joint ligaments, and dynamic stabilizers such as the extensor carpi ulnaris. ¹ As such, DRUJ instability and its cause are challenging to diagnose and requires a thorough history and physical examination in addition to appropriate imaging studies. Clinical examination of the DRUJ involves multiple provocative maneuvers and compares one side with the other. The premise is that the uninjured, contralateral wrist is normal and acts as a control to which the injured wrist is compared with and is the foundation on which diagnoses of DRUJ instability is made. To our knowledge, this has not been studied.

Traditional imaging to diagnose DRUJ pathology includes anteroposterior and lateral radiographs of the wrist. A true lateral radiograph of the wrist requires superimposition of the palmar cortex of the pisiform between volar cortex of the capitate and distal pole of the scaphoid. In this situation, subluxation of the ulna can be detected. ² However, this is unable to evaluate dynamic subluxation. Several studies have used computed tomography (CT) of bilateral wrists to discern differences in displacement that indicate DRUJ instability.^{3 -6} However, the current literature compares the imaging in a “normal” wrist with that of a wrist with prior trauma or DRUJ symptomatology. More novel approaches with dynamic CT protocols focus on imaging the DRUJ with various amount of torque to better simulate physiologic loading and identify dynamic instability. ⁷ To our knowledge, there are no studies in the literature evaluating the outcomes of dynamic CT protocols comparing bilateral “normal” wrists.

The purpose of this study was to quantify the in vivo displacement of bilateral DRUJs in resisted pronosupination. We hypothesize that this will demonstrate no appreciable difference between the left and right DRUJs, thus validating the concept of using the uninjured wrist as a control for physical examination as well as dynamic imaging studies.

Methods

Patients and Evaluation

An institutional review board–approved study was performed of 32 participants devoid of any history of wrist pathology. Healthy volunteers were recruited from the community. Volunteers were questioned if they had a history of wrist pain or trauma/disease affecting the upper extremities and were excluded if present. They underwent a dynamic CT protocol evaluating bilateral forearm rotation in a customized device that allows pronosupination against resistance. ⁷ In brief, the protocol is as follows. The patient lies in a prone position with both arms partially elevated above the head. The device allows CT imaging simultaneously of bilateral DRUJs in positions of neutral forearm rotation, 60° pronation, and 60° supination, including maximal isometric muscle loading. The forearms are supported and secured into well-padded channels such that the elbows and wrists are suspended, and shoulder motion is restrained to minimize internal and external rotations of the arm. The custom device with single grip is mounted on a Plexiglass frame and can be rotated and located into 3 positions corresponding to neutral forearm rotation, 60° pronation, and 60° supination. The patient holds onto the grips and maximally rotates both forearms according to the instruction of the technician (Figure 1).

Figure 1.

Photograph of the device allowing computed tomography imaging simultaneously of bilateral distal radioulnar joints in standard positions of neutral forearm rotation, 60° pronation, and 60° supination, including maximal isometric muscle loading.

The DRUJ alignment, specifically, the absolute degree and direction of subluxation of the ulna relative to the sigmoid notch, was then assessed. We used a modified radioulnar line method similar to that described by Nakamura et al ⁴ to quantify the amount of absolute displacement of the ulnar head relative to the sigmoid notch. A straight line (A) was drawn between the dorsal-radial corner of the radius and the dorsal-most aspect of the sigmoid notch articular surface followed by a second straight line (B) between the volar-radial corner of the radius and the volar-most aspect of the sigmoid notch. We recorded the amount of subluxation of the ulnar head relative to the sigmoid notch as the maximum distance of the ulnar head outside of and perpendicular to these lines (Figure 2). In neutral and pronation, we measured the amount of displacement of the ulnar head relative to the dorsal line, while in supination, we measured the amount of displacement of the ulnar head relative to the volar line. For supination, the displacement was recorded as a negative value if the ulnar head was dorsal to the line and a positive value if the ulnar head was volar to the line. A pilot measurement of the first 5 patients was performed by all 4 of the authors including selection of the CT series and image number, followed by calculation of the ulnar translation. To reduce variability, we then elected to have the 2 senior authors (K.K.A., S.K.) choose the CT series and image for the final analysis. The axial reformatted CT image of each wrist showing the largest area of the sigmoid notch, including Lister’s tubercle and the ulnar head, were selected by 2 senior authors (K.K.A., S.K.). When different images were selected, the final decision was left to the senior authors who reached consensus (K.K.A., S.K.). All CT images were then independently assessed by 2 authors (J.D.M., J.J.M.) for measurement of ulnar translation in neutral, pronation, and supination. Age, sex, and handedness were recorded for all participants.

Figure 2.

Axial computed tomography of the right distal radius including the measurements demonstrating our technique for measuring displacement of the ulnar head with respect to the distal radioulnar joint.

Statistical Analysis

Continuous variables were summarized using mean, range, and standard deviation. The α level was set at .05. Means comparison was performed with paired t test or analysis of variance.

A power analysis was performed for the paired data. Using the mean difference in displacement of 0.89 mm and a standard deviation of the difference of 0.65 mm, the study required a sample size of 8 to achieve a power of 0.80 at a significance level of .05.

Results

The study included 64 wrists in 32 patients. There were 19 male and 13 female patients with a mean age of 30 years (range = 22-47). Twelve patients were right-hand dominant, but handedness was not recorded for the remaining 20 patients.

For the final analysis, the axial images were selected independently by the 2 senior authors (S.K., K.K.A.). The initial agreement rate was 92% (177 of 192 images). The mean displacement values are recorded in Table 1. There was no significant difference in the mean displacement when comparing the right and left sides in neutral (P = .13), pronation (P = .36), or supination (P = .17). Neutral and pronation displacement values were measured relative to the same line. Therefore, we compared the degree of displacement between pronation and neutral, and there was a statistically significant greater displacement in pronation as compared with neutral on both the right (3 vs 1.3 mm, P < .01) and left (2.8 vs 1 mm, P < .01).

Table 1.

Author 1 + Author 2 Combined Displacement Data (mm).

	Neutral		Pronation		Supination
	R	L	R	L	R	L
Mean	1.30625	0.99375	3.00625	2.835938	–0.565625	–0.30156
Range—max	4.15	3	5.3	4.65	1.95	1.75
Range—min	–1.5	–3	0.75	0	–2.4	–2.7
SD	1.215077	1.181425	1.135764	0.967224	1.1379408	1.202315
	P = .13		P = .36		P = .17

The mean displacement was also compared between male and female patients. There was no statistically significant difference in absolute displacement in neutral (male 0.99 mm vs female 1.38 mm, P = .16) or supination (male −0.57 mm vs female −0.23 mm, P = .27). However, the difference in pronation was statistically significant (male 2.69 mm vs female 3.26 mm, P = .03). There was no significant difference in age between male and female patients in our cohort.

Of the 192 sequences, the measurements of displacement of the 2 authors (J.J.M., J.D.M.) were within 1 mm 86% of the time (166/192). A Bland-Altman plot was generated to assess the agreement between the 2 authors (Figure 3). The mean bias for all measurements of displacement was −0.34 mm, indicating that on average, the measurements of author J.D.M. were 0.34 mm greater than that of author J.J.M. The standard deviation was 0.6 and the correlation coefficient (r) was 0.95. The 95% confidence limit ranged from −1.51 to 0.82.

Figure 3.

Bland-Alman plot demonstrating the agreement between authors.

The sigmoid shape was characterized for each patient. The 2 sides were symmetric in shape in all cases. There were 15 (47%) flat face types, 9 (28%) “C” types, 4 (13%) “S” types, and 4 (13%) ski-slope types. There was no significant difference in mean displacement based upon sigmoid shape type in neutral (P = .10), pronation (P = .38), or supination (P = .13).

Discussion

When treating patients with DRUJ instability, traditional teaching advocates for the comparison of the injured wrist with the contralateral wrist in neutral, pronation, and supination. This concept is based on the uninjured wrist having similar stability to the affected wrist prior to injury. To our knowledge, this has not been validated. We therefore undertook this study to establish the basis of bilateral DRUJ kinematics in a normal population using a novel DRUJ CT protocol. The CT imaging provides a rapid and accurate method to assess bilateral DRUJs in multiple planes, thus allowing a comparison of the normal side with the injured side. Multiple methods of quantifying DRUJ translation have been developed. Mino et al ⁸ described the diagnosis of DRUJ instability on CT and provided the initial description of the radioulnar line method. Lines were drawn from the volar-radial and dorsal-radial borders of the distal radius to the volar and dorsal portions of the sigmoid notch, and any displacement of the ulnar head outside of these borders was defined as instability. We found that in neutral and pronation, the ulnar head often resides outside the borders of the sigmoid notch in normal patients. Nakamura et al ⁴ described a modified criterion to account for the high number of false-positive results. They examined 37 patients with suspected DRUJ subluxation with bilateral wrist CTs and imaged the DRUJ in pronation, supination, and neutral. They concluded that dorsal subluxation in pronation and volar subluxation in supination of less than 25% of sigmoid notch diameter is “normal,” because in all except 1 of the uninjured wrists, the maximum value of ulnar subluxation was less than 25% of the sigmoid notch diameter. Several other scoring methods have been devised including epicenter, radioulnar ratio, and subluxation ratio.^{9 -11} Park and Kim critically evaluated the various methods of CT for quantifying DRUJ translation. ¹⁰ They scanned 45 asymptomatic wrists in supination, neutral, and pronation and tried to establish normative values. However, they discovered significant variation in normal values using all these reported methods, supporting the need for bilateral, comparative studies to reliably diagnose DRUJ instability.

The significant patient-to-patient variation in DRUJ biomechanics and translation has thus far precluded the development of reliable normative values. Wijffels et al ⁶ evaluated bilateral wrist CTs in 46 patients treated conservatively for distal radius fractures. There were no differences in pronation and supination between the uninjured and injured wrists or when comparing injured wrists with and without pain in pronosupination. Pan et al ⁵ evaluated bilateral CT scans in pronation and supination in 40 patients with a history of unilateral distal radius fracture and a clinically asymptomatic contralateral wrist without prior trauma. They compared the data of the 40 uninjured, contralateral wrists with 25 injured wrists that had no pain (15 of the injured wrists were excluded as they had either pain or stiffness). The authors found that the measurements of the injured but asymptomatic wrists and the uninjured wrists were not statistically different, concluding that values in normal wrists could serve as a normative baseline. These findings largely agree with those reported in our study, where we were unable to identify differences between the right and left sides. However, the current literature is limited to comparisons of a previously injured wrist with an uninjured wrist and relies on the assumption that the left and right sides are symmetric at baseline and therefore that any detectable differences represent instability. This leaves an important question unanswered: Are the left and right sides symmetric in healthy participants? In addition, the described protocols and the above studies evaluated the wrist in various degrees of pronation, supination, and neutral but do not apply torque to the wrist and thus do not thoroughly evaluate the dynamic instability of the wrist under load.

To our knowledge, this is the first study in the literature comparing dynamic CT scans in bilateral asymptomatic wrists without a history of prior trauma or dysfunction. Our findings demonstrate that there are no appreciable differences between the right and left sides at baseline in pronation, supination, or neutral with torque. This is a critical finding as it supports the premise that one can use the contralateral wrist as a control for DRUJ kinematics when examining the injured side.

Our findings have been confirmed by others using a validated ultrasound technique.^12,13 Yuine and colleagues ¹⁴ used a force-monitor ultrasound system to quantitively assess DRUJ displacement in 60 patients devoid of any wrist pain or pathology and found that there were no significant differences between the dominant and nondominant side regarding radioulnar displacement, applied force to displacement, or displacement-to-force ratio. They concluded that “if there are differences in the numerical values of each parameter on the injured side compared with the healthy side with this measurement method, it can be considered as abnormal findings of DRUJ stability,” which is further supported by our findings. However, there are limitations to this ultrasound technique. First, the authors examined for volar displacement and did not consider dorsal displacement. In addition, the results of ultrasound are subject to the skill and expertise of the sonographer to obtain accurate measurements which can vary from person to person.

We also sought to evaluate whether there is a difference in displacement values between male and female patients and noted no significant differences in neutral or supination, but in pronation, there was more displacement in female patients (male 2.69 mm vs female 3.26 mm, P = .03). Yuine et al ¹⁴ found no significant difference between male and female patients regarding radioulnar displacement or displacement-to-force ratios, but the applied force-to-displacement was significantly greater in the male group, suggesting that women may have more DRUJ laxity than men.

There are several limitations with our study. The generalizability of the technique is dependent on the reproducible selection of appropriate axial images and anatomic landmarks to draw reference lines. We, however, demonstrated excellent concordance in the measurements between the 2 authors. Given the similarity of the trends measured within this cohort, we feel that the findings of concordance between both wrists is valid. All patients who were enrolled reported no history of wrist injury or trauma and as such, we did not collect information pertaining to range of motion, clinical evidence of instability, or joint laxity. Future applications include dynamic testing in the setting of DRUJ pathology to objectively identify DRUJ instability. However, patients with painful DRUJ instability may have a limited ability to fully participate in the examination. In these cases, the senior authors have had success performing intra-articular local anesthetic injections prior to the dynamic testing to improve patient comfort and optimize participation in the examination.

There have been several publications evaluating the different techniques for evaluating DRUJ translation with some conflicting data on which provides the highest intra-observer and inter-observer reliability.^6,13 Ultimately we selected the radioulnar line method because of: (1) its high intra-observer and inter-observer reliability as reported by Park and Kim; (2) the simplicity and ease of the technique with our imaging software; (3) the relatively few anatomic landmarks required to measure; and (4) the concordance between authors when completing the measurements. ¹⁰ Given these findings, we felt that the radioulnar line method is a proven method. The goal of this article was not to further evaluate which method was superior. Rather we chose to use a proven, reproducible, reliable technique that allowed us to complete the primary objective of the manuscript—to compare the displacement between sides.

Notwithstanding these limitations, this study demonstrates that dynamic CT of bilateral DRUJs in resisted pronation, supination, and neutral demonstrated symmetry between the right and left DRUJs, supporting the concept of using the contralateral side as a control to identify instability in an injured wrist. These findings have important clinical ramifications when the physician is using the contralateral, normal wrist as a basis to determine pathology within the injured DRUJ.