Abstract
The purpose of this study was to investigate the differences in three-dimensional carpal kinematics between type 1 and 2 lunates. We studied 15 instances of wrist flexion to extension (nine type 1, six type 2), 13 of radial to ulnar deviation (seven type 1, six type 2), and 12 of dart-throwing motion (six each of type 1 and 2) in 25 normal participants based on imaging with computerized tomography. Mean proximal translation of the distal articular midpoint of the triquetrum relative to type 2 lunates during wrist radioulnar deviation was 2.9 mm (standard deviation (SD) 0.7), which was significantly greater than for type 1 lunates, 1.6 mm (SD 0.6). The hamate contacted the lunate in ulnar deviation and ulnar flexion of wrists with type 2 lunates but not with type 1. We conclude that the four-corner kinematics of the wrist joint are different between type 1 and 2 lunates.
Keywords
Introduction
Two lunate morphologies, types 1 and 2, have been described (Burgess, 1990; Viegas, 1990). Type 1 lunates do not articulate with the hamate, but type 2 lunates have a medial articular facet with the hamate (Burgess, 1990; Viegas, 1990; Viegas et al., 1990). Lunate–hamate impaction, which has been observed with type 2 lunates (Nakamura et al., 2000), can cause proximal hamate arthritis or chondromalacia. A few carpal kinematic studies have evaluated the influence of lunate morphology. Nakamura et al. (2000) described the in vivo carpal kinematics of the lunate and the hamate in cases where the hamate impinged on the lunate in wrist ulnar deviation, but using two-dimensional X-rays and magnetic resonance imaging. Bain et al. (2015) described differences in the motion of carpal bones in relation to lunate morphology. However, this was a cadaver study and only analysed wrist flexion–extension and radioulnar deviation movements.
There have been reports of three-dimensional (3D) or four-dimensional (3D + time) computerized tomography (CT) for demonstrating wrist 3D kinematics (Carelsen et al., 2005; Got et al., 2016; Pickering et al., 2016; Rainbow et al., 2016). However, there are no in vivo 3D studies of kinematics of four-corner bones (lunate, hamate, capitate and triquetrum) associated with different types of the lunate. The purpose of this study was to evaluate the motion of the four-corner carpal bones, especially the hamate and triquetrum, in relation to the lunate in type 1 and 2 lunates.
Methods
This study was approved by our institutional review board. The in vivo kinematics of the normal four-corner wrist joint between the lunate and hamate, the lunate and triquetrum, and the lunate and capitate were evaluated during wrist flexion to extension (FEM), radial to ulnar deviation (RUD), and radial extension to ulnar flexion (dart-throwing motion). Images from healthy volunteers involved in previous studies (Moritomo et al., 2006) and from patients who presented at our institution with pathologies such as carpal instability or bone deformity of the contralateral wrist were obtained from our database. All participants provided informed consent to use their data for both primary research and future studies and were permitted to withdraw their consent whenever they wished to do so. Patients were eligible for inclusion if they had normal wrist CT data with three consecutive positional CT scans for at least one movement: flexion, extension and neutral (FEM); radial deviation, ulnar deviation and neutral (RUD); and radial extension, ulnar flexion and neutral (dart-throwing motion). Patients with carpal instability, as assessed by a hand surgeon, with disease or history of trauma confirmed in a medical interview or inquiry, with insufficient CT data such as the lack of serial positioning CT images, FEM, RUD or dart-throwing motion, or who were unable to keep the wrist in a static position during CT scanning were excluded.
CT (LightSpeed Ultra; General Electric, Waukesha, WI, USA or Aquillion; Toshiba, Tochigi, Japan) scanning conditions were varied as follows: tube voltage was 120 kV, tube current was 10–100 mA, slice thickness was 0.5–0.625 mm and pixels were 0.175 × 0.175 mm – 0.473 × 0.473 mm). CT imaging was performed with the participant in a prone position with their arms elevated above the head, and the forearm was positioned in neutral rotation position with the fist unclenched. Each static position was set in the individual’s active maximal position without a jig in order to demonstrate natural active kinematics of the wrist. Three-dimensional surface models of the carpal bones, radius and ulna were created from digital imaging and communications in medicine data using commercially available computer software (Bone Viewer; Orthree Co., Ltd, Osaka, Japan). Lunate morphology was determined from the coronal view CT by measuring the shortest distance between the capitate and the triquetrum (C–T distance) in the neutral position (Galley et al., 2007; Nakamura et al., 1997). A C–T distance of ≤3.0 mm indicated a type 1 lunate; a distance of >3.0 mm indicated type 2.
Kinematic variables
Kinematic variables were calculated by registering each carpal bone, the radius and the ulna from the extension, radial deviation or radial extension position to their neutral and flexion, ulnar deviation or ulnar flexion positions using a surface-based registration technique (Besl and McKay, 1992) and commercially available software (Bone Simulator; Orthree Co., Ltd, Osaka, Japan). Registration was performed automatically by the software using an iterative closest point algorithm and the accuracy of the registration was reported to be 0.5 mm in this system (Oka et al., 2009). We set up an orthogonal coordinate system with the radius in the neutral position as a reference (Belsole et al., 1991; Wu et al., 2005). The Y-axis was the inertial axis of the radius and ran from the distal to the proximal end. The Z-axis was in a plane perpendicular to the Y-axis and parallel to the orthogonal projection of the line that ran from the base of the concavity of the sigmoid notch toward the top of the radial styloid process, and the X-axis was mutually perpendicular to both the Y- and the Z-axes (Figure 1(a)). The origin was the intersection point of these three axes. Therefore, rotation around the Y-axis indicated internal rotation (+) or external rotation (−), rotation around the Z-axis indicated flexion (+) or extension (−) and rotation around the X-axis indicate ulnar deviation (+) or radial deviation (−). According to the recommendations of Wu et al., (2005), the coordinate system for the lunate was set to be parallel with the coordinate system for the radius and the origin of the coordinates was set to its volumetric centre (Wu et al., 2005).
(a) Coordinate system for the radius in a neutral wrist position with the positive X-axis in the volar direction, positive Y-axis in the proximal direction and positive Z-axis to the radial side. (b) Red spheres on triquetrum indicated distal volar and dorsal apices of the lunate–triquetrum joint. The yellow sphere on the triquetrum was the midpoint of two red spheres and defined as the distal articular midpoint. The distance of movement of the midpoint was calculated as lunate–triquetrum articular congruity.
A video was made from three static positions using the software (Bone Simulator). The screw axis method (An et al., 1988), which describes spatial movement of the rigid body by rotation around a unique axis and translation along the axis was used to create interpolating positions. Interpolations were equally divided in 5–10 ways between both the extreme and neutral positions. The videos were representative cases of each lunate type rather than generated from average kinematic data.
Range of motion and direction of movement
The 3D rigid body movement was initially described by screw axis rotation (An et al., 1988; Panjabi et al., 1981). The screw displacement axis (SDA) data described range of motion around a unique axis of rotation. To analyse direction as well, the Euler angle method (An et al., 1988), which describes the orientation of the rigid body with respect to a fixed coordinate system, was used. Therefore, range of motion was finally calculated as rotation around each axis of the coordinate system using Euler angle methods (in Y-X-Z order) and considering each axis of the coordinate system and the location of the origin. Translation was defined as the 3D distance between the volumetric centres of the bone at the extreme wrist positions.
Global wrist motion was assessed by the motion of the capitate relative to the radius, both by SDA and the Euler angle methods, because there was little motion between the third metacarpal and the capitate (Patterson et al., 1998). The direction of global wrist motion was ascertained by the Euler angle.
Lunate–triquetrum distal articular congruity
To describe distal articular congruity between the lunate and triquetrum during wrist motion, we set distal volar and dorsal apices of the triquetrum/lunate articular surface and the midpoint of these two points was defined as the distal articular midpoint of the triquetrum/lunate (Figure 1(b)). The distance of movement of the midpoint of the triquetrum with respect to the midpoint of the lunate between extreme wrist positions (between flexion and extension, radial and ulnar deviations and radial extension and ulnar flexion) was calculated as lunate–triquetrum articular congruity.
Lunate–hamate distance
To assess lunate–hamate impaction, the distance between these two bones was measured by proximity mapping (Marai et al., 2004) with a threshold distance of 5.0 mm using Bone Simulator. The shortest distance between the lunate and the hamate was measured at each position only in wrists with type 2 lunates, because only type 2 lunates have articular facets with the hamate.
Four-corner motion with respect to lunate
Motion of the capitate, hamate and triquetrum relative to the lunate were assessed by SDA and the Euler angle methods based on the coordinate system of the lunate. Midcarpal motion was assessed by the motion of the capitate relative to the lunate.
Statistics
Continuous variables were described as means and standard deviation (SD). Multivariate analysis of variance with Pillai’s trace and unpaired t-tests with Bonferroni corrections as post-hoc tests were used to analyse the difference between type 1 and 2 lunates for global wrist motion Lunate-Hamate (L-H), Lunate-Triquetrum (L-T), Lunate-Capitate (L-C) in each type of movement. The unpaired t-test was used to analyse the motion of the distal articular midpoint of the triquetrum relative to the lunate. The paired t-test was used to analyse type 2 lunate–hamate distance between extreme wrist positions in each motion. Statistical significance was defined as p < 0.05.
Power analysis showed that we needed 15 patient pairs with a significant level of 0.05, power of 0.8 and effect size of 0.8 for the paired t-test. However, we did not meet this sample size criterion because this study was performed within institutional review board acceptance as a retrospective study. We assessed at least six patient pairs for each movement and described p values, effect size and power (1-ß).
Results
Global wrist motion in wrists with type 1 and type 2 lunates.
SDA was analysed with the unpaired t-test. Euler angle and translation were analysed by multivariate analysis of variance.
Positivenegative indicates ulnar deviation–radial deviation.
Positivenegative indicates internal rotation–external rotation.
Positivenegative indicates flexion–extension rotation.
FEM: flexion to extension; RUD: radial to ulnar deviation; SD: standard deviation; SDA: screw displacement axis.
Lunate–triquetrum distal articular congruity
Proximal translation of the distal articular midpoint of the triquetrum in type 2 lunate wrists from radial deviation to ulnar deviation was 2.9 mm (SD 0.7), which was significantly greater than that observed in type 1 lunate wrists (1.6 mm (SD 0.6), p = 0.007, effect size = 1.8, 1-ß = 0.92) (Figure 2, Videos 3 and 4). There were no noteworthy differences between type 1 and 2 lunates for other movements (Table 2).
A representative case of both type 1 (a) and 2 (b) lunate during wrist RUD viewed from the ulnar side. The lunate (white model), triquetrum, radius, ulna (translucent model) and the distal articular midpoint of the triquetrum are shown at both ulnar deviation (blue sphere) and radial deviation (red sphere). The midpoint of the triquetrum translates proximally during wrist ulnar deviation and the distance was greater in type 2 lunates (b) than in type 1 lunates (a). Coloured arrows indicate the coordinate system of the lunate. Lunate–triquetrum distal articular congruity. Motion of distal articular midpoint of the triquetrum with respect to lunate was compared between lunate type 1 and 2 in each wrist motion using the unpaired t-test. FEM: flexion to extension; RUD: radial to ulnar deviation; SD: standard deviation.
Lunate–hamate distance
The distance between the type 2 lunates and the hamate was smallest in ulnar deviation (0.4 mm (SD 0.3)), followed by ulnar flexion (0.6 mm (SD 0.3)). Both distances were significantly smaller than that observed at radial deviation (4.7 mm (SD 1.0), p < 0.001, effect size = 4.0, 1-ß > 0.999) and radial extension (2.9 mm (SD 0.9), p < 0.001, effect size = 3.4, 1-ß > 0.999). The area of closest contact in ulnar deviation and ulnar flexion was the centre of the proximal pole of the hamate (Supplemental data 2). The result indicated that the hamate contacted the type 2 lunate at extreme ulnar deviation and ulnar flexion of the wrist, but maintained the distance from the lunate in FEM (1.2 mm (SD 0.5) in extension, 1.8 mm (SD 0.9) in flexion).
Four-corner motion with respect to lunate type
Motion of the capitate, hamate and triquetrum relative to lunate observed during FEM, RUD and dart-throwing motion in wrists with type 1 and 2 lunates are shown in Videos 1–6 and Supplemental data 3–5. There were no noteworthy differences in L–H, L–T and L–C rotation between type 1 and 2 lunates. In both type 1 and 2 lunate wrists, the capitate, hamate and triquetrum always rotated in a similar direction relative to lunate in each motion. The three bones rotated mainly from extension towards flexion in FEM and from radial extension toward ulnar flexion in RUD. In dart-throwing motion, the direction of rotation was intermediate to that seen in FEM and RUD. The range of rotation of the triquetrum relative to lunate was about one-third that of the capitate in any wrist motion (Supplemental data 3–5). There was no noteworthy difference in the percentage contribution of midcarpal (capitate–lunate motion) to global wrist motion (capitate–radius motion) between type 1 (58%) and type 2 lunates (53%) in FEM, RUD (86% in type 1 and 93% in type 2) or dart-throwing motion (83% in type 1 and 85% in type 2).
Discussion
Our study revealed that a significant difference existed in motion of the distal articular midpoint of the triquetrum relative to the midpoint of the lunate between type 1 and 2 lunates in RUD (Figure 2). In type 2 wrists, the distal articular midpoint of the triquetrum translated in a proximal direction along the articular surface of the lunate during RUD, and the sliding distance was significantly greater than in type 1 wrists. In videos of RUD, the triquetrum in type 2 wrists translated proximally to avoid hamate–triquetrum impingement and to increase the range of midcarpal motion because of the offset of the distal articular surfaces of the lunate and triquetrum (Figure 3, Videos 3 and 4).
A representative case of 3D model of the four-corner carpal bones of type 1 ((a) and (b)) and type 2 ((c) and (d)) lunates in radial deviation ((a) and (c)) and ulnar deviation ((b) and (d)) viewed from the dorsal aspect. (d) As the hamate facet of the lunohamate joint is located proximally in type 2 lunates, the distal articular surface of the triquetrum translates proximally in type 2 lunates (arrow). Lunatohamate impaction occurs at ulnar deviation (arrow head), and the lunotriquetral interosseous ligament (coloured model in (c) and (d) is stretched by triquetrum motion.
Ulnar deviation includes slight pronation and flexion (Kobayashi et al., 1997) and the direction of capitate motion relative to the lunate gradually shifts from the FEM plane toward the dart-throwing motion plane as the wrist motion changes from FEM to RUD (Moritomo et al., 2006). Our results are consistent with these findings. As shown in Figure 3 and Video 4, lunate–triquetrum motion appeared to add a shearing force to the lunotriquetral interosseous ligament (LTIL), which may be described as a triquetrum ‘shearing’ motion. Because LTIL tears have been associated with type 2 lunates (Harley et al., 2004; Pfirrmann et al., 2002) and hamate arthritis (Burgess, 1990), triquetrum shearing may give rise to LTIL damage in type 2 lunate wrists.
Regarding the distance between the lunate and hamate, we found the hamate seemed to contact type 2 lunates at extreme ulnar deviation and/or at ulnar flexion of the wrist (Figure 3 and Supplemental data 2), whereas the hamate maintained the distance from the lunate during FEM. In addition, we found that the closest point and area in ulnar deviation and ulnar flexion was the centre of the proximal pole of the hamate (Supplemental data 2). We speculate repetitive movement could induce hamate arthritis in type 2 lunates more easily. Our result is consistent with those of previous reports on hamate impingement of the lunate in ulnar deviation (Nakamura et al., 2000) and increased occurrence of arthritis in type 2 lunates (Burgess, 1990; Viegas, 1990). Our observation of motion between the hamate and type 2 lunates is also consistent with a clinical report, which stated that type 2 lunate wrists are more vulnerable to LTIL tear and hamate arthrosis (Harley et al., 2004).
There are some limitations in this study. First, it included a small number of participants. However, we did find a significant difference in motion of articular midpoint of the triquetrum relative to the lunate between type 1 and 2 lunate during ulnar deviation of the wrist. Second, because this study was performed retrospectively, the number of images in each motion for each type of lunate was not standardized. Third, the models were derived from static 3D views in three positions for each wrist motion. Fourth, we did not confirm the presence or absence of a pre-existing LTIL tear by MRI. However, no participants complained of wrist pain or carpal instability. Therefore, we assumed that the participants did not have LTIL tears that greatly affected the kinematic data. Finally, the participants were CT scanned in their active position without jig, which causes the variability of the direction of global wrist motion. However, there were no significant differences in global wrist motion between type 1 and 2 lunates.
This study demonstrated 3D kinematics of four-corner wrist motion involving type 1 and 2 lunates. It is hoped that these findings will contribute to a better understanding of the involvement of wrist carpal kinematics in wrist disorders and assist orthopaedic surgeons in the diagnosis of ulnar wrist pain related to type 2 lunate. First, increased lunotriquetral shearing motion in type 2 lunates during RUD may cause LTIL tears in type 2 lunate wrists; lunotriquetral fusion in type 2 lunate wrists may influence postoperative restriction of range of motion. Second, hamate contact with the type 2 lunate at ulnar deviation and at ulnar flexion may cause proximal hamate arthrosis. We suggest that these kinematic mechanisms associated with LTIL tears, and the hamate arthrosis on type 2 lunate could be responsible for ulnar wrist pain.
Supplemental Material
Supplemental Material1 - Supplemental material for Three-dimensional kinematics of the lunate, hamate, capitate and triquetrum with type 1 or 2 lunate morphology
Supplemental material, Supplemental Material1 for Three-dimensional kinematics of the lunate, hamate, capitate and triquetrum with type 1 or 2 lunate morphology by Shingo Abe, Hisao Moritomo, Kunihiro Oka, Kazuomi Sugamoto, Kenji Kasubuchi, Tsuyoshi Murase and Hideki Yoshikawa in Journal of Hand Surgery (European Volume)
Supplemental Material
Supplemental Material2 - Supplemental material for Three-dimensional kinematics of the lunate, hamate, capitate and triquetrum with type 1 or 2 lunate morphology
Supplemental material, Supplemental Material2 for Three-dimensional kinematics of the lunate, hamate, capitate and triquetrum with type 1 or 2 lunate morphology by Shingo Abe, Hisao Moritomo, Kunihiro Oka, Kazuomi Sugamoto, Kenji Kasubuchi, Tsuyoshi Murase and Hideki Yoshikawa in Journal of Hand Surgery (European Volume)
Supplemental Material
Supplemental Material3 - Supplemental material for Three-dimensional kinematics of the lunate, hamate, capitate and triquetrum with type 1 or 2 lunate morphology
Supplemental material, Supplemental Material3 for Three-dimensional kinematics of the lunate, hamate, capitate and triquetrum with type 1 or 2 lunate morphology by Shingo Abe, Hisao Moritomo, Kunihiro Oka, Kazuomi Sugamoto, Kenji Kasubuchi, Tsuyoshi Murase and Hideki Yoshikawa in Journal of Hand Surgery (European Volume)
Footnotes
Acknowledgement
We thank for Ryouji Nakao, a computer programmer, who contributed to the study.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: this work was supported by AMED Grant Number 17hk0102023h0003 and by JSPS KAKENHI Grant Number 26350522.
Ethical approval
This study was approved by our institutional review board.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
