Study of Ulnar Variance with High-Resolution MRI: Correlation with Triangular Fibrocartilage Complex and Cartilage of Ulnar Side of Wrist

Transcription

1 JOURNAL OF MAGNETIC RESONANCE IMAGING 26: (2007) Original Research Study of Ulnar Variance with High-Resolution MRI: Correlation with Triangular Fibrocartilage Complex and Cartilage of Ulnar Side of Wrist Hiroshi Yoshioka, MD, PhD, 1 * Toshikazu Tanaka, MD, 2 Teruko Ueno, MD, 3 John A. Carrino, MD, 1 Carl S. Winalski, MD, 1 Piran Aliabadi, MD, 1 Philipp Lang, MD, MBA, 1 and Barbara N. Weissman, MD 1 Purpose: To investigate correlations with ulnar variance and the triangular fibrocartilage complex (TFCC) or cartilage of ulnar side of the wrist on high-resolution MRI with a microscopy coil. Materials and Methods: We reviewed ulnar variance, TFCC, and cartilage of the ulnar side of the wrist in 93 subjects (29 asymptomatic volunteers and 64 patients with suspected TFCC injury) with high-resolution MRI using a 47-mm microscopy surface coil. All MR images were obtained with a 1.5 T scanner. Coronal 2D gradient recalled echo T 2 *-weighted images were used for analysis. For qualitative analysis we measured ulnar variance, TFCC angle, thickness in the central portion of TFCC disc proper, and cartilage thickness of the lunate and the ulnar head on MRI and calculated the correlation coefficient between measured values. We also examined the relationship between ulnar variance and age or sex. Results: High-resolution MR images clearly demonstrated TFCC and cartilage of the wrist and ulnar variance. The mean ulnar variance on MRI was 0.26 mm (range, 4.59 to 3.71 mm). The mean TFCC angle and TFCC thickness were 23.9 (range, 4.6 to 54.1 ) and 1.11 mm (range, 0.4 to 3.22 mm), respectively. Ulnar variance and TFCC angle were positively correlated (r 0.84), and ulnar variance and TFCC thickness were negatively correlated (r 0.71). However, ulnar variance and lunate or ulnar head cartilage thickness were not significantly correlated. Conclusion: High-resolution MRI with a microscopy coil is a useful tool for evaluating the relationship between ulnar variance and ulnar side structures. 1 Department of Radiology, Brigham and Women s Hospital, Boston, Massachusetts. 2 Department of Orthopedic Surgery, Kikkoman General Hospital, Noda, Japan. 3 Japan Department of Radiology, Iwate Medical School, Morioka, Iwate, Japan. *Address reprint requests to: H.Y., Department of Radiology, Brigham and Women s Hospital, 75 Francis St., Boston, MA hyoshioka@partners.org Received September 22, 2006; Accepted June 26, DOI /jmri Published online in Wiley InterScience ( Key Words: ulnar variance; high-resolution MRI; triangular fibrocartilage complex J. Magn. Reson. Imaging 2007;26: Wiley-Liss, Inc. ULNAR VARIANCE refers to the lengths of the distal articular surfaces of the radius and the ulna (1). This variance affects the force distribution across the wrist and is related to some ulnar pathologies. Several studies reported an association of negative ulnar variance with Kienböck s disease (2 4), while some studies reported that negative ulnar variance does not seem to be an important factor in the etiology of Kienböck s disease (5,6). The relationship between Kienböck s disease and negative ulnar variance remains controversial. On the other hand, positive ulnar variance plays a significant role in triangular fibrocartilage complex (TFCC) tears and ulnar impaction syndrome (1,7 10). Chronic impaction between the ulnar head and the TFCC and ulnar carpus results in ulnar impaction syndrome, where degenerative tear of the TFCC and chondromalacia of the lunate, triquetrum, and distal ulnar head are seen (1). Thus, an understanding of the relationship between ulnar variance and ulnar-sided structures of the wrist such as TFCC and cartilage of the lunate and ulnar head is clinically important. The TFCC is a fibrocartilage-ligament complex on the ulnar side of the wrist that is interposed between the ulna and the ulnar carpus and arises from the radius and inserts into the distal ulna and ulnar carpus (10). The disc proper of the TFCC forms the congruent surfaces for both the distal ulna and the ulnar carpus (9). The TFCC has three major functions: 1) it cushions the ulnar carpus, 2) is the major stabilizer of the distal radioulnar joint ulnar-sided carpal stability, and 3) is a stabilizer of the ulnar carpus (10). With positive ulnar variance the compressive load through the center of the disc to the ulnar head is greatly increased, whereas with negative ulnar variance probably more of the force is converted to tension (9). From these anatomic characteristics the disc proper injury is deeply related to ulnar variance Wiley-Liss, Inc. 714

2 Ulnar Variance with High-Resolution MRI 715 Figure 1. High-resolution T 2 *-weighted MR image demonstrating an example of measurement of (a) ulnar variance and (b) TFCC angle. Generally, measurements of ulnar variance are based on posteroanterior radiographs obtained while the patient forearm is in a neutral position (11,12). However, the soft tissue contrast in a plain radiograph is poor, whereas the contrast is excellent with MRI. The normal disc proper shows asymmetrical bowtie-like low signal intensity on coronal MRI with all pulse sequences. The carpal cartilage shows high signal intensity with T 2 *- weighted sequences. High-resolution MRI allows clear visualization of both TFCC and carpal cartilage (13). Accordingly, the purpose of this study was to investigate correlations with ulnar variance and the TFCC or cartilage of ulnar side of the wrist on high-resolution MRI with a microscopy coil. We also aimed to understand the relationship between progression of TFCC disc injury and TFCC or cartilage morphology in conjunction with ulnar variance. MATERIALS AND METHODS We examined the relationship between ulnar variance and TFCC or cartilage of the lunate and the distal ulnar head in 93 subjects (41 males and 52 females; 29 asymptomatic healthy volunteers and 64 patients with suspected TFCC injury) with high-resolution MRI using a receive-only 47-mm microscopy surface coil. Subjects ranged in age from 14 to 67 years, with a mean age of 32.6 years (27.0 years for volunteers and 35.2 years for patients; 32.2 years for males and 32.9 years for females) and a median age of 30 years. The patients had a positive sign in the painful click test and pain due to ulnar deviation and negative for traction and pronation/supination and were clinically suspected of having TFCC injury. Arthroscopy was performed in 11 patients and TFCC injury was confirmed. No subjects had a history of wrist fracture. MR images of the patients wrist were performed as a routine MR examination of the TFCC injury. All subjects gave written informed consent. All MR images were obtained with a 1.5 T scanner (Gyroscan NT Intera, Philips Medical Systems; Best, The Netherlands). All subjects were placed in the prone position with their arm extended over the head, simulating a swimming position (forearm pronation) without flexion/extension or ulnar/radial deviation, and each surface coil was positioned over the ulnar side of the wrist. Coronal 2D gradient recalled echo (GRE) T 2 *-weighted images (T 2 *WI) (TR/TE/flip angle /13-18/40, a mm slice thickness, a 50-mm field of view, an imaging matrix of using zero-fill interpolation, 72.4 khz receiver bandwidth, and three or four excitations) were used for analysis. For qualitative analysis one slice where the entire TFCC was best visualized including its radial and ulnar attachment was selected. Then the ulnar variance on MRI, TFCC angle, thickness in the central portion of TFCC disc proper, and thickness of cartilage of the lunate and the ulnar head adjacent to the central portion of TFCC were measured on that slice. In 10 patients TFCC thickness was not evaluated because they had a large disrupted tear in the disc proper and TFCC thickness was not measurable. These 10 patients were classified as having advanced disc injury. A peripheral TFCC tear such as the lunotriquetal ligament injury and a small disc perforation were included in the TFCC thickness measurement as long as TFCC thickness was measurable in the central portion of the disc. Among them, two had TFCC injury confirmed by arthroscopy and eight did not undergo arthroscopy. The ulnar variance on MRI was indicated by a difference between the line at the level of lunate fossa cartilage surface and the line at the level of the ulnar head cartilage surface (Fig. 1a). TFCC angle was defined as the angle between the long axis of TFCC and the above lines (Fig. 1b). The long axis of TFCC was drawn from the tip of the ulnar notch of the radius at the TFCC radial attachment toward ulnar direction along the midpoint of the disc substance, which is equal distance from the proximal surface and the distal surface of the disc. The correlation coefficient between measured values was calculated. The relationship between ulnar variance and age or sex of the subjects also was examined. The Mann Whitney U-test and the Kruskal-Wallis test were used to determine the significance of the differences in measured values between two samples and more than two samples, respectively. The chi-square test was used for a test of independence. In each analysis a P-value 0.05 was considered significant. RESULTS High-resolution MR images with a microscopy coil clearly demonstrated TFCC and cartilage of the wrist, as well as ulnar positive, neutral, and negative variance (Fig. 2). The mean ulnar variance on the MRIs of all subjects was 0.26 mm with a range of 4.59 to 3.71 mm (Table 1). The mean TFCC angle and the mean TFCC thickness were 23.9 (range, 4.57 to 54.1 ) and 1.11 mm (range, 0.40 to 3.22 mm), respectively. Figure 2. Ulnar variance. (a) Positive, (b) neutral, (c) negative ulnar variance can be clearly demonstrated with high-resolution MRI.

3 716 Yoshioka et al. Table 1 Mean SD, Maximum, and Minimum Values of Ulnar Variance and Ulnar-sided Structures of the Wrist on High-resolution MRI Measurement Patients with Advanced Disc Injury (n 10) Patients without Advanced Disc Injury (n 54) Volunteers (n 29) Ulnar variance (mm) Mean SD Maximum Minimum TFCC angle (degrees) Mean SD Maximum Minimum TFCC thickness (mm) Mean SD Maximum Minimum Lunate cartilage thickness (mm) Mean SD Maximum Minimum Ulnar head cartilage thickness (mm) Mean SD Maximum Minimum Total Ulnar variance and TFCC angle showed a positive correlation (r 0.84) (Fig. 3), whereas ulnar variance and TFCC thickness showed a negative correlation (r 0.71) (Fig. 4). The TFCC thickness decreased almost linearly below an ulnar variance of 1 mm. However, the thickness was constant above an ulnar variance of 1 mm because of limited cartilage space between the lunate and the ulnar head. The mean TFCC angle and TFCC thickness were slightly lower among volunteers than among patients; however, this difference was not significant (Table 1). The mean ulnar variance was 1.56 mm among the 10 patients whose TFCC thickness was not evaluated because of advanced disc injury, whereas it was 0.14 mm among patients without advanced disc injury and 0.05 in volunteers (Fig. 5; Table 1). The differences in ulnar variance between the three groups were statistically significant (P 0.05). The mean TFCC angle with advanced disc injury, without advanced disc injury, and in volunteers was 26.2, 24.2, and 22.5, respectively. The differences in TFCC angle between groups were not statistically significant. Table 2 shows the number of patients without advanced disc injury and volunteers classified by 1.5 mm of ulnar variance. Among these 54 patients, 13 (24.1%) showed ulnar variance of 1.5 mm or greater, whereas among the 29 volunteers only 1 (0.03%) did. The difference was statistically significant (P 0.05). The mean TFCC angle in these 13 patients was 37.8, which was significantly larger than that in volunteers and the rest of patients without advanced disc injury (P ). The mean cartilage thickness of the lunate and ulnar head was 1.10 mm and 0.80 mm, respectively, a statistically significant difference in thickness (P ). There was a slight negative correlation, but no significant correlation between ulnar variance and lunate cartilage thickness (r 0.10) or ulnar head cartilage thickness (r 0.12). The mean lunate/ulnar head cartilage thickness with advanced disc injury, without advanced disc injury, and in volunteers was 1.04 mm / 0.82 mm, 1.13 mm / 0.82 mm, and 1.05 mm / 0.76 Figure 3. Scattergram of ulnar variance versus TFCC angle. Figure 4. Scattergram of ulnar variance versus TFCC thickness.

4 Ulnar Variance with High-Resolution MRI 717 Figure 6. Graph shows correlation between ulnar variance and age. The equation of the regression line plotted in the graph is y x , and the correlation coefficient (r) is Figure 5. MR images of ulnar side of the wrist in a 54-year-old woman. Advanced disc injury does not allow measurement of TFCC thickness, but ulnar variance and cartilage thickness can be measured. The TFCC angle also can be measured from the ulnar notch of the radius and remaining TFCC. Note that this patient has positive ulnar variance. mm, respectively. The differences in cartilage thickness between groups were not statistically significant. The linear correlation coefficient between age and ulnar variance was 0.16 (r ), and the positive correlation was weak (Fig. 6). The mean and standard deviation (SD) of ulnar variance was for females and for males (Fig. 7). The difference in ulnar variance between females and males was statistically significant (P 0.05). Even among healthy volunteers, the mean ulnar variance was higher in males than in females (mean SD; for females and for males), but this difference was not statistically significant. DISCUSSION Accurate measurement of ulnar variance is important in evaluating unexplained ulnar-sided wrist pain. Ulnar variance is defined as neutral, positive (plus), or negative (minus) on the basis of whether the distal articular surface of the ulna is aligned with the distal articular surface of the radius on a neutral posteroanterior radiograph (12). A change in ulnar variance of 1 mm can alter mechanical transfer characteristics by 25% and is likely to have particular clinical significance in individuals who perform repetitive loaded rotational maneuvers (12). However, the mean ulnar variance ranges from 0.46 mm to 0.31 mm in published reports and differs by ethnicity, gender, and age (9,12,14,15). Ulnar variance also alters with forearm rotation, wrist deviation, and x-ray beam inclination (16). In our study the mean ulnar variance on MRI was 0.26 mm. This value was in the range of previously reported data (9,12,14,15) but was relatively large because the MR images were obtained with the subjects forearm in the pronated position. The TFCC thickness was measured at the central part of the disc, and the mean thickness was 1.11 mm, with a range of 0.40 to 3.22 mm in this study. This result is consistent with that of a cadaveric study of 109 wrist joints, in which the thickness of central area of the disc ranged from 0.5 to 3 mm and in most cases was 1 mm (9). Therefore, the measurement of ulnar variance and TFCC thickness with high-resolution MRI seems to be accurate. High-resolution MRI can be a useful tool for measuring ulnar side structures. This study, like an earlier study (17), showed a negative correlation between the ulnar variance and TFCC Table 2 Number of Patients without Advanced Disc Injury and Volunteers Classified by 1.5 mm of Ulnar Variance Group 1.5 mm 1.5 mm Total Patients without advanced disc injury Volunteers Total Figure 7. Scattergram shows distribution of ulnar variance in males and females.

5 718 Yoshioka et al. thickness. In addition, our study showed a positive correlation between ulnar variance and TFCC angle. These are reasonable results when one considers the space available between the lunate and the ulnar head (17). However, the mean cartilage thickness of the lunate and the ulnar head was not significantly correlated with ulnar variance, although a slight tendency toward a negative correlation was seen. This suggests that the thickness and angle of the TFCC is dependent on the available space between ulnar carpus and ulnar head in the development process, while the development of the bone and cartilage of the carpus and that of the ulnar head are relatively independent. Interestingly, the lunate cartilage was significantly thicker than the ulnar head cartilage. The cartilage thickness of proximal ulnar side of the lunate seems to be thicker than that of the rest of the lunate (Figs. 2, 5). A chemical-shift artifact might influence cartilage thickness, but the difference in cartilage thickness might be related to the fact that the ulnar side of the disc suffers much more severe degenerative changes than the carpal side of the disc (9). Even in patients with advanced disc injury, the cartilage thickness of the lunate and the ulnar head was not significantly different from that of other groups, probably because most of our patients had traumatic TFCC lesions. This result also means TFCC abnormality precedes cartilage abnormality. Unfortunately, with the GRE sequence in the present study we could not see the changes in signal intensity of the carpal cartilage, which might be an early sign of cartilage degeneration (18). We also analyzed the difference in measured values among patients with and without advanced disc injury and among healthy volunteers. Palmer (10) reported that positive and neutral variance is more commonly associated with abnormalities with the TFCC, and our study also showed that patients with advanced disc injury had a larger ulnar plus variance and TFCC angle as compared with other groups. Even patients without advanced disc injury tended to show larger ulnar plus variance and TFCC angle than did healthy volunteers. These findings support positive ulnar variance as a predisposing factor of TFCC disc injury. The TFCC angle also could be related to the disc injury. Negative ulnar variance has been reported to be uncommon in individuals with a TFCC abnormality (10), and we also showed that no patients with advanced disc injury had negative ulnar variance. Given that positive ulnar variance plays an important role in TFCC injury, it would be helpful to determine the threshold of positive ulnar variance at which TFCC injury develops. Because the mean ulnar variance in patients with advanced disc injury was 1.56 mm, we classified volunteers and patients without advanced disc injury into two groups, those with an ulnar variance below and those with a variance above 1.5 mm. Only one volunteer (.4%) had an ulnar variance larger than 1.5 mm, whereas 13 patients (24.1%) belonged to this category. These patients did not have advanced disc injury at the MR examination, but perhaps they will progress to advanced disc injury. It is interesting that these patients also had a significantly larger TFCC angle than did volunteers or patients without disc injury. The correlation between ulnar variance and age in this study was similar to that seen in previous studies (2,6). Nakamura et al (6) reported the correlation coefficient between ulnar variance and age in normal wrist was in males, in females, and overall. Bonzar et al (2) reported a correlation coefficient of 0.21 for the control subjects in their study; the value in our study was The slope of the regression line was in the present study, whereas it was in a previous study (6). However, the sex-related difference in ulnar variance in our study was the opposite that of other studies. Nakamura et al (6) reported that the mean value of ulnar variance was lower in males than in females in all four age groups: 14 30, 31 40, 41 50, and years. The mean ulnar variance was 0.14 in all males and 0.77 in all females in their study. Schuind et al (19) reported a mean ulnar variance of 2.4 mm in males and 1.9 mm in females. We do not know why the ulnar variance was higher in males than in females in our study ( 0.65 in males and 0.04 in females), even though other measured values were comparable to those in previous studies. However, only a few studies reported the relationship between ulnar variance and sex (6,19), and future studies may provide a more definitive answer. Our study has several limitations. First, the MR images were obtained with the forearm in the pronated position, although the standard radiographic view is generally with the forearm in the neutral position. Supination results in a relative ulnar minus variance and pronation, more relative ulnar plus variance (10). These changes in ulnar variance result in relatively small but potentially significant changes in the load borne by selected portions of the TFCC (10). However, great skill is necessary to measure ulnar variance accurately with plain radiography to avoid the effect of the beam incidence angle (20). Sugimoto et al (17) reported that the measurement of ulnar variance is more precise with MRI than with plain radiography because the MRI need not consider this angle and successfully measured ulnar variance on MRI with pronation in their study. The second limitation of our study is the lack of pathological confirmation. Therefore, our group of patients without advanced disc injury might have included patients with early-stage disc injury or disc degeneration. Third, the numbers of patients with injury to the large disc proper and healthy volunteers were small. The cases with more advanced disc injury are necessary to fully establish the relationship between ulnar variance and the development of disc injury. Fourth, although the ulnar variance changes depending on the amount of wrist flexionextension and deviation, we did not test reproducibility by doing repetitive measure on the same studies after repeated repositioning. Finally, all subjects in this study were Japanese. Since ulnar variance differs among different ethnic groups, the outcome may be different among different populations. However, this result from one ethnic group might be useful for future comparisons with other ethnic groups. Despite these limitations, we believe our study highlights the impor-

6 Ulnar Variance with High-Resolution MRI 719 tance of measuring TFCC thickness and TFCC angle, as well as ulnar variance, on high-resolution MRI. In conclusion, measurements of ulnar variance with MRI were correlated with TFCC thickness and TFCC angle but not with cartilage thickness of the lunate and ulnar head. In this study the correlation between ulnar variance and age was similar to that in earlier studies, whereas the relationship between ulnar variance and sex was the opposite of earlier reports. High-resolution MRI with a microscopy coil could be a useful tool for evaluating the ulnar side structures of the wrist, including TFCC and cartilage associated with ulnar variance. REFERENCES 1. Cerezal L, del Pinal F, Abascal F, Garcia-Valtuille R, Pereda T, Canga A. Imaging findings in ulnar-sided wrist impaction syndromes. Radiographics 2002:22: Bonzar M, Firrell JC, Hainer M, Mah ET, McCabe SJ. Kienbock disease and negative ulnar variance. J Bone Joint Surg Am 1998: 80: Gelberman RH, Salamon PB, Jurist JM, Posch JL. Ulnar variance in Kienbock s disease. J Bone Joint Surg Am 1975:57: Mirabello SC, Rosenthal DI, Smith RJ. Correlation of clinical and radiographic findings in Kienbock s disease. J Hand Surg [Am] 1987:12: D Hoore K, De Smet L, Verellen K, Vral J, Fabry G. Negative ulnar variance is not a risk factor for Kienbock s disease. J Hand Surg [Am] 1994:19: Nakamura R, Tanaka Y, Imaeda T, Miura T. The influence of age and sex on ulnar variance. J Hand Surg [Br] 1991:16: Mellado JM, Calmet J, Domenech S, Sauri A. Clinically significant skeletal variations of the shoulder and the wrist: role of MR imaging. Eur Radiol 2003:13: Imaeda T, Nakamura R, Shionoya K, Makino N. Ulnar impaction syndrome: MR imaging findings. Radiology 1996:201: Mikic ZD. Detailed anatomy of the articular disc of the distal radioulnar joint. Clin Orthop Relat Res 1989: Palmer AK. Triangular fibrocartilage complex lesions: a classification. J Hand Surg [Am] 1989:14: Tomaino MM. The importance of the pronated grip x-ray view in evaluating ulnar variance. J Hand Surg [Am] 2000:25: Mann FA, Wilson AJ, Gilula LA. Radiographic evaluation of the wrist: what does the hand surgeon want to know? Radiology 1992: 184: Yoshioka H, Ueno T, Tanaka T, Shindo M, Itai Y. High-resolution MR imaging of triangular fibrocartilage complex (TFCC): comparison of microscopy coils and a conventional small surface coil. Skeletal Radiol 2003:32: Kristensen SS, Thomassen E, Christensen F. Ulnar variance determination. J Hand Surg [Br] 1986:11: Chen WS, Shih CH. Ulnar variance and Kienbock s disease. An investigation in Taiwan. Clin Orthop Relat Res 1990: Schuurman AH, Maas M, Dijkstra PF, Kauer JM. Assessment of ulnar variance: a radiological investigation in a Dutch population. Skeletal Radiol 2001:30: Sugimoto H, Shinozaki T, Ohsawa T. Triangular fibrocartilage in asymptomatic subjects: investigation of abnormal MR signal intensity. Radiology 1994:191: Yoshioka H, Stevens K, Hargreaves BA, et al. Magnetic resonance imaging of articular cartilage of the knee: comparison between fat-suppressed three-dimensional SPGR imaging, fat-suppressed FSE imaging, and fat-suppressed three-dimensional DEFT imaging, and correlation with arthroscopy. J Magn Reson Imaging 2004: 20: Schuind F, Alemzadeh S, Stallenberg B, Burny F. Does the normal contralateral wrist provide the best reference for X-ray film measurements of the pathologic wrist? J Hand Surg [Am] 1996:21: Epner RA, Bowers WH, Guilford WB. Ulnar variance the effect of wrist positioning and roentgen filming technique. J Hand Surg [Am] 1982:7: