Anatomical Variations of the Extensor Carpi Ulnaris Groove: A New Computed Tomography-based Evaluation

Transcription

1 Original Article The Journal of Hand Surgery (Asian-Pacific Volume) 2017;22(3): DOI: /S Anatomical Variations of the Extensor Carpi Ulnaris Groove: A New Computed Tomography-based Evaluation Sho Kohyama, Toshikazu Tanaka, Eriko Okano, Ochiai Naoyuki Department of Orthopedic Surgery, Kikkoman General Hospital, Noda, Chiba, Japan Background: This study aimed to develop a classification that precisely describes the extensor carpi ulnaris (ECU) groove morphology. Methods: Reconstructed axial plane computed tomography scans of the wrists of 200 patients were reviewed. Three groups of groove shape were created based on the position of the deepest point: deviated to the ulnar side (Type U), in the middle of the groove (Type M), and deviated to the radial side (Type R). Groove depth, width, carrying angle, and radius of the curvature were measured using the Picture Archiving and Communication System in a slice in which the ulnar head was the largest. Results: Type U was present in 88 patients (44%), Type M in 74 patients (37%), and Type R in 38 patients (19%). The average depth, width, carrying angle, and radius of curvature were 2.2 mm, 9.2 mm, 135.8, and 7.0 mm, respectively. Depth, width, and carrying angle were normally distributed. Both depth and width were statistically correlated with the carrying angle; groove depth and width were not correlated. ECU groove shape and depth showed excellent intra- and inter-observer reliabilities; the reliabilities for the width were poor. Therefore, depth subgroups were defined using cutoffs of ±2 standard deviations (SD): d1 ( -2 SD), 1.0 mm; d2 (±2 SD), mm; and d3 ( +2 SD), 3.4 mm. The 200 wrists were classified as follows: 0.5% Ud1, 40.2% Ud2, 1.3% Ud3, 0.8% Md1, 38.3% Md2, 0.8% Md3, 1.5% Rd1, 16.4% Rd2, and 0.2% Rd3. Conclusions: The detailed morphology of the ECU groove was classified using three major types and depth subgroups. The Type R ECU groove, which lacks a medial bony buttress, might be more prone to ECU-related injuries. This classification helps to understand the ECU tendon-related injury pathologies and may provide valuable information for treatment decisions; however, further research is necessary. Keywords: Anatomy & histology, Classification, Pathology, Tendons, Wrist INTRODUCTION The morphology of the extensor carpi ulnaris (ECU) groove is likely related with the pathology of ECU-related injuries. The fibro-osseous ECU subsheath reportedly prevents subluxation of the ECU tendon and works with Received: Feb. 4, 2016; Revised: Jun. 2, 2016; Accepted: Jun. 28, 2016 Correspondence to: Toshikazu Tanaka Department of Orthopedic Surgery, Kikkoman General Hospital, Miyazaki 100, Noda, Chiba , Japan Tel: , Fax: tanaka1041@msn.com the extensor retinaculum to prevent tendon bowstring. 1,2) Volar subluxation of the tendon during supination is prevented by the transverse fibers of the medial wall of the subsheath. 3) A shallow ECU groove might lead to ECU tendon instability and subluxation, and deepening the ECU groove 2 3 mm with a power bur has been recommended to improve the congruency of the tendon and groove. 3-5) However, there have been few studies regarding the ECU groove morphology, 6,7) and the ideal morphology of the ECU groove remains unknown. Furthermore, the shape of the ECU groove has not been discussed previously. The purpose of this study was to evaluate the morphology of the ECU groove, including

2 282 Sho Kohyama, et al. CT-based Evaluation of the ECU Groove its shape, and to develop a brief classification that could provide further understanding of ECU-related injuries. METHODS Computed tomography (CT) is routinely performed in our facility to assess fracture patterns in patients with a distal radius fracture. Following institutional review board approval, we reviewed 230 reconstructed axial plane CT scans (Brilliance CT 64, PHILIPS, Amsterdam, The Netherlands) of the wrists of patients who had visited our facility between August 2009 and December 2014 and were diagnosed with a distal radius fracture. Patients with a fracture in the ECU groove were excluded, resulting in 200 scans for evaluation. There were 63 men and 137 women with an average age of 59.4 (range, 17 86) years. The author (examiner 1) assessed the 200 CT scans twice, and a senior author (examiner 2) independently reexamined the CT scans once. We classified the ECU groove shape into three types based on the position of the deepest point of the groove: Type U, deepest point was deviated to the ulnar side; Type M, deepest point was in the exact center location of the groove, showing a symmetrical groove shape; and Type R, deepest point was deviated to the radial side (Fig. 1). The type of the ECU groove shape, depth, width, carrying angle, and radius of the curvature were assessed from a reconstructed axial slice with the largest ulnar head using the Picture Archiving and Communication System (Shade Quest/Serv ST, Yokogawa Medical Solutions Corporation, Japan). The techniques for measuring ECU groove depth and width are shown in Fig. 2. Groove depth was measured at the deepest portion of the groove as the distance of the line perpendicular to the line tangent to the most superficial borders of the groove. The distance of the 2 points tangent to the most superficial borders of the groove was recorded as groove width. The techniques for measuring the carrying angle of the ECU groove and radius of curvature are shown in Fig. 3. The carrying angle of the groove was measured as the angle formed from subtending the 2 end points of the groove width and the point at which groove depth was measured. The radius of curvature was calculated as the area of best fit of a superimposed circle. Fig. 1. Three major types of extensor carpi ulnaris groove shape based on the position of the deepest point of the groove. Type U: deepest point was deviated to the ulnar side; Type M: deepest point was in the middle of the groove; Type R: deepest point was deviated to the radial side. w d r ca d: depth w: width Fig. 2. Techniques for measuring extensor carpi ulnaris groove depth and width. Depth: distance of the line perpendicular to the line tangent to the most superficial borders of the groove to the deepest point of the groove; Width: distance of the 2 points tangent to the most superficial borders of the groove. ca: carrying angle r: radius of curvature Fig. 3. Techniques for measuring the carrying angle of the groove and radius of curvature of the extensor carpi ulnaris. Carrying angle: angle formed from subtending the 2 end points of the groove width and the point at which groove depth was measured; Radius of curvature: area of best fit of a superimposed circle.

3 283 The Journal of Hand Surgery (Asian-Pacific Volume) Vol. 22, No. 3, Statistical analysis The distributions of ECU depth, width, carrying angle, and radius of curvature were assessed using the Kolmogorov-Smirnov test. Intra-and inter-observer reliabilities of the type of ECU groove shape, depth, width and carrying angle were assessed using inter-rater correlation coefficients. The statistical correlations for depth and width, depth and carrying angle, and width and carrying angle were analyzed using Pearson correlation coefficients (r) for normally distributed values and Spearman s rank correlation coefficients (rs) for values that were not normally distributed. RESULTS The results of the measurements for the 200 patients are shown in Tables 1 and 2. The first measurements taken by the author are labeled as examiner 1-1, the second measurements taken by the author are labeled examiner Table 1. Number of each extensor carpi ulnaris groove type in 200 patients with a distal radius fracture, by examiner Number of patients (percentage) Examiner (44%) 74 (37%) 38 (19%) Examiner (42%) 79 (39.5%) 37 (18.5%) Examiner 2 80 (40%) 87 (43.5%) 33 (16.5%) Type U: deepest point was deviated to the ulnar side; Type M: deepest point was in the middle of the groove; Type R: deepest point was deviated to the radial side. 1-2, and the measurements taken by the senior author are labeled examiner 2. All of the depth measurements followed a normal distribution (examiner 1-1, p = 0.26; examiner 1-2, p = 0.47; examiner 2, p = 0.20), as tested using the Kolmogorov-Smirnov test. All of the width measurements also followed a normal distribution (examiner 1-1, p = 0.13; examiner 1-2, p = 0.15; examiner 2, p = 0.27), and all of the carrying angle measurements, except for those conducted by examiner 1-2, followed a normal distribution (examiner 1-1, p = 0.49; examiner 1-2, p = 0.018; examiner 2, p = 0.32). None of the measured radii of curvature followed a normal distribution (examiner 1-1, p = 0.009; examiner 1-2, p = 0.005; examiner 2, p = 0.008). The type of ECU groove shape, depth, and carrying angle had excellent intra- and inter-observer reliabilities (Table 3). However, both intra- and inter-observer reliabilities for the width were poor. To develop the classification, we considered the groove width unsuitable for correlation analysis as well as subgrouping based on the poor reliabilities of the measurements. The statistical analysis of the correlations for depth and carrying angle showed significant correlation (examiner 1-1, r = -0.81; examiner 1-2, r s = -0.79; examiner 2, r = -0.80), indicating a smaller carrying angle with a deeper ECU groove. Based on these results, to keep the classification as simple as possible, we decided to define subgroups based on depth, because of the good correlation between depth and carrying angle. In addition, we took into consideration the lack of a normal distribution of the carrying angle measurements conducted by examiner 1-2. Because the groove depth followed a normal Table 2. Measurements of the extensor carpi ulnaris groove in 200 patients with a distal radius fracture, by examiner Depth (mm) Width (mm) Carrying angle ( ) Radius of curvature (mm) Examiner ± 0.6* ( ) 9.2 ± 1.3* ( ) ± 13.9* ( ) 7.0 ± 2.8 ( ) Examiner ± 0.6* ( ) 9.0 ± 1.1* ( ) ± 13.4* ( ) 7.0 ± 3.0 ( ) Examiner ± 0.6* ( ) 9.0 ± 1.5* ( ) ± 13.7* ( ) 7.0 ± 2.9 ( ) Values are Mean ± Standard Deviation (range). *normal distribution. Table 3. Reliabilities of the measurements of the extensor carpi ulnaris groove in 200 patients with a distal radius fracture Type Depth Width Carrying angle Intra-observer (Examiner 1-1 & 1-2) Inter-observer (Examiner 1-1 & 2) Inter-observer (Examiner 1-2 & 2) The reliability was assessed using inter-rater correlation coefficients.

4 284 Sho Kohyama, et al. CT-based Evaluation of the ECU Groove d1 < 1.0 mm d mm d3 > 3.4 mm 0.5% 0.8% 1.5% 40.2% 1.3% distribution, we used ±2 standard deviations (SD) as the cutoffs to determine a shallow or deep groove: d1 (shallow, -2 SD), 1.0 mm; d2 (normal, ±2SD), mm; and d3 (deep, +2 SD), 3.4 mm. Considering the ECU groove shape and depth, the 200 wrists were classified as follows: 0.5% Ud1, 40.2% Ud2, 1.3% Ud3, 0.8% Md1, 38.3% Md2, 0.8% Md3, 1.5% Rd1, 16.4% Rd2, and 0.2% Rd3 (Fig. 4). DISCUSSION 38.3% 0.8% 16.4% 0.2% Fig. 4. New classification for the extensor carpi ulnaris (ECU) groove. The new classification includes the three major types (M, U, and R) of the ECU groove morphology and subtypes of the groove based on depth (d1, d2, and d3). We hypothesized that the morphology of the ECU groove, especially the shape, is related to ECU injuries and developed a brief classification to enable deeper understanding of ECU pathologies. To our knowledge, only 2 reports have been published regarding ECU groove morphology. 6,7) In 2014, a review of magnetic resonance images (MRI) of 60 patients by Iorio et al. 6) provided evidence of the normal ECU and tendon morphology, ECU groove depth, and comparative baseline values. The average measurements were 1.4 mm for depth, 9.0 mm for width, and 143 for carrying angle. Comparatively, deeper depths and smaller carrying angles were detected in the present study. These differences may be related with the radiographic tool used for measurements; Iorio et al used MRI, while we used CT. Intra-and inter-observer reliabilities for our width measurements were poor, potentially because of two reasons. First, when the groove is very shallow, it is difficult to determine the most superficial borders of the groove, especially when the groove is nearly flat; this makes it challenging for examiners to measure the groove width. Second, the accuracy of the PACS system that we used might have affected the reliabilities. It can be difficult to accurately point the target on the display using a mouse. For these two reasons, the reproducibility of the width measurement was poor; therefore, we considered the width unsuitable for correlation analysis as well as subclassification. Measurement inaccuracy might have affected the distribution of the carrying angle measurements conducted by examiner 1-2. Because the carrying angles were measured as the angle formed from subtending the 2 end points of the groove width and the point at which groove depth was measured, inaccurate width measurements can result in varying carrying angle measurements. Anatomical variations of the ECU grooves were reported in 1993, with the ECU groove for 240 cadaver upper limbs classified into 4 grades according to depth. 7) However, because the groove depth was classified only as deep, shallow, or flat, the classification was unclear; furthermore, a classification study similar to the present study has not been reported previously. Many studies in the literature have discussed ECUrelated injuries that commonly occur in athletes participating in tennis, baseball, hockey, or golf. 2,4,5,8-10) However, most previous studies have focused on the soft tissue structures surrounding the ECU tendon, 2,10) which is maintained within a bony groove of the ulna by the retinaculum and subsheath. If the subsheath is torn, stripped, or attenuated, ECU subluxation or dislocation can occur. 3) The ECU tendon has a direct course during pronation; however, it is subjected to maximal traction and has to adopt an angle of approximately 30 degrees to reach the base of the fifth metacarpal. 11) This anatomical feature is thought to play an important role in ECUrelated injuries. Deepening of the ECU groove has been reported recently to treat tendon instability or subluxation, based on the theory that a shallow or flat ECU groove leads to recurrent ECU tendon dislocations. 3-5) However, the optimal depth remains unknown, although mm has been suggested. 6) On the other hand, Graham et al. 9) cautioned against overaggressive deepening of the groove, noting that the tendon may have a greater tendency to subluxate if there is a loss of the medial buttress. As mentioned, Allende et al. 3) reported that volar subluxation of the tendon during supination is prevented by the transverse fibers of the medial wall of the subsheath. Therefore, the proper ECU groove has moderate depth with a medial buttress. Based on the present study, the ECU groove shapes that provide a bony medial but-

5 285 The Journal of Hand Surgery (Asian-Pacific Volume) Vol. 22, No. 3, tress are Types U and M, and the normal range for the groove depth is mm, the range of subgroup d2. In other words, Type R ECU grooves might be more prone to ECU-related injuries. Assuming that this theory is correct, the ECU groove should be deepened in the direction of the medial or ulnar side, no more than 3 mm in depth. However, this is just a hypothesis. The relationship between the classification and type of ECU-related injury should be clarified in future studies. Then, this classification could become very useful when selecting treatment options, including surgical procedures. There are several other limitations in this study. First, we only studied the Japanese population, and it is unknown whether this classification can be applied to other races. Second, the CT scans were reconstructed by several different radiologists, and any misregistration of the axial axis can affect the measurement outcome. Third, since this is a CT-based classification, the relationship between bony structures and surrounding soft tissues, including the ECU tendon, remains unclear. We developed a brief classification that describes the three major types (M, U, and R) of ECU groove morphology in detail and the subtypes of the groove based on depth (d1, d2, and d3) (Fig. 4). This new classification might facilitate treatment decisions for ECU tendonrelated injuries; however, further studies are necessary to understand the relationship between this classification and ECU-related injuries REFERENCES 1. Spinner M, Kaplan EB. Extensor carpi ulnaris. Its relationship to the stability of the distal radio-ulnar joint. Clin Orthop Relat Res. 1970;68: Taleisnik J, Gelberman RH, Miller BW, Szabo RM. The extensor retinaculum of the wrist. J Hand Surg Am. 1984;9: Allende C, Le Viet D. Extensor carpi ulnaris problems at the wrist classification, surgical treatment and results. J Hand Surg Br. 2005;30(3): MacLennan AJ, Nemechek NM, Waitayawinyu T, Trumble TE. Diagnosis and anatomic reconstruction of extensor carpi ulnaris subluxation. J Hand Surg Am. 2008;33(1): Oka Y, Handa A. Recurrent dislocation of the ECU tendon in a golf player: release of the extensor retinaculum and partial resection of the ulno-dorsal ridge of the ulnar head. Hand Surg. 2001;6(2): Iorio ML, Bayomy AF, Huang JI. Morphology of the extensor carpi ulnaris groove and tendon. J Hand Surg Am. 2014;39(12): Nakashima T, Hojo T, Furukawa H. Deep and shallow forms of the sulcus for extensor carpi ulnaris. J Anat. 1993;183(Pt 3): Campbell D, Campbell R, O Connor P, Hawkes R. Sportsrelated extensor carpi ulnaris pathology: a review of functional anatomy, sports injury and management. Br J Sports Med. 2013;47(17): Graham TJ. Pathologies of the extensor carpi ulnaris (ECU) tendon and its investments in the athlete. Hand Clin. 2012;28(3): Palmer AK, Skahen JR, Werner FW, Glisson RR. The extensor retinaculum of the wrist: an anatomical and biomechanical study. J Hand Surg Br. 1985;10(1): Montalvan B, Parier J, Brasseur JL, Le Viet D, Drape JL. Extensor carpi ulnaris injuries in tennis players: a study of 28 cases. Br J Sports Med. 2006;40(5):424-9.