Biomechanical Evaluation of the Porcine Carpus as a Potential Preclinical Animal Model for the Human Carpus

Abstract

Advancing successful treatments for carpal instabilities of the wrist are hindered due, in part, to limited preclinical animal models. The purpose of this study was to evaluate the forelimb of the Yucatan minipig (YP) as a preclinical animal model for the human wrist by quantifying carpal biomechanics in vitro in the intact and after two ligament transection conditions. Porcine wrist biomechanics (n = 12, 5M, 7F) were determined in 28 range of motion (ROM) directions, in pronation-supination, and in volar-dorsal translation using a six-axis robotic musculoskeletal simulator. Testing was implemented in three conditions – intact, and after sequential transection of radial intermediate ligament (RIL) and dorsal intercarpal ligament (DIC). Mixed models were employed to examine differences in direction and conditions among male and female specimens. The intact ROM envelope was elliptical in shape and oriented toward ulnar flexion with the largest ROM about 15° from the flexion-extension axis. Transection of RIL and DIC did not alter the ROM envelope orientation, however, subtle increases in ROM were observed in extension and radial deviation following transection of both RIL and DIC. Pronation in neutral was greater than supination in all three test conditions. Volar translation increased subtly in the RIL+DIC condition. This novel study investigated the multidirectional biomechanics of the YP forelimb. ROM in the general directions of extension, radial and ulnar deviation were less than in humans, while flexion was substantially larger. These specific ligament transections had minor effects on the biomechanics of the YP forelimb.

1. Introduction

Arthroplasties of the hip, knee, and shoulder produce high success and satisfaction rates due to well-developed protocols for standardized in vitro testing, preclinical animal models, and clinical trials (Swiontkowski et al., 2022; Wadhwa et al., 2019). In comparison, the wrist joint lags behind these larger joints in terms of satisfaction and clinical outcomes (Boeckstyns and Herzberg, 2024; Kopriva et al., 2023). The lack of success may be due, in part, to the absence of a characterized preclinical animal model for bony and ligamentous anatomy and joint biomechanics, as well as the complexity of human wrist anatomy and biomechanics.

The human wrist joint owes its complexity to the balanced arrangement of eight unique, small carpal bones divided into two rows, articulating proximally with the radius and ulna and distally with the base of the metacarpals (Figure 1). The human carpus lacks any notable muscle insertions, so the stability of these eight carpal bones is achieved solely through the shape of their articulations and a complex ligamentous system (Andrews and Youm, 1979; Kitay and Wolfe, 2012; Tang and Varacallo, 2024). Bony or ligamentous trauma has the potential to permanently disrupt systemic mechanical stability resulting in irregular postures and movements, altered joint loading, and irreversible degeneration (Kitay and Wolfe, 2012) placing the patient at an increased risk for osteoarthritis (OA) (Fleming et al., 2005; Gabay and Gabay, 2012; Kwok et al., 2011). The scapholunate interosseous ligament (SLIL) is the primary stabilizer of the scapholunate (SL) joint and the most frequently injured carpal ligament (Kitay and Wolfe, 2012). The dorsal intercarpal ligament (DIC) functions as a secondary stabilizer to the SL joint and is a critical stabilizer to the proximal carpal row (Kitay and Wolfe, 2012; Wolfe and Kakar, 2022). Studies have demonstrated the functional impact of SLIL and DIC ligament insufficiencies on human carpal stability (Badida et al., 2021; Padmore et al., 2019; Pérez et al., 2019; Short et al., 2002); however these methods have not been examined outside of humans.

Figure 1.

Bone surface models from CT images of the human and porcine wrist demonstrating two distinct rows of carpal bones proximally articulating with the radius/ulna and distally articulating with the metacarpals. The RCB and ICB are bound together tightly by the deep (colored red) and superficial components of the RIL. The superficial RIL supports the proximal row laterally from the RCB to the ulnar carpal bone (UCB) and the DIC diagonally extends from the RCB to the 4^th Carpal Bone.

The Yucatan Minipig (YP) has been employed as a preclinical model in orthopedics owing to the homologies of its musculoskeletal systems (Cone et al., 2017). Studies evaluating ligament injuries, bone fracture, and cartilage damage have used porcine models because of the homologies of the porcine knee in bony anatomy, bone mineral density, morphology, and healing to humans (Aerssens et al., 1998; An and Friedman, R.J., 1999; Beveridge et al., 2019; Thorwarth et al., 2005). For example, in the knee, the YP has been used for investigating anterior cruciate ligament (ACL) repair techniques (Karamchedu et al., 2021) and cartilage degeneration due to meniscus damage (Bansal et al., 2020). Previous upper extremity/forelimb animal studies have utilized canines, rabbits, and rats to mechanically evaluate carpal tunnel compliance for carpal tunnel syndrome (Tung et al., 2010). A recent study employed the accessory carpal bone in the YP carpus as a model for the biologic joint replacement for trapeziometacarpal osteoarthritis (Stoeckl et al., 2021). The only other identified study in the forelimb sought to characterize the YP as a preclinical model for fracture non-unions of the scaphoid (Behrends et al., 2015).

The skeletal anatomy of the YP carpus has numerous bony and ligamentous homologies with the human carpus (Figure 1). Of the ligaments identified in the YP carpus, two are postulated to be homologous with the critical stabilizers of the human carpus; the radial intermediate ligament (RIL) with the SLIL and the dorsal intercarpal ligament (DIC) with the identically named DIC. To date, the biomechanics of the porcine carpus has not yet been investigated. This study aims to evaluate the Yucatan minipig as a preclinical, large animal model of the human wrist through quantifying the biomechanics of the carpus in the intact and ligament transection conditions. We hypothesized that the biomechanics of the YP wrist will vary with the direction of motion and that there will be an increase in range of motion (ROM) and a decrease in stiffness (KROM) after both ligaments, the RIL and DIC, are transected.

2. Methods

2.1. Anatomy

Gross anatomical review of the YP carpus revealed that the carpus includes 8 individual carpal bones, each with a unique size and geometry. The carpal bones are assembled in two distinct rows bound together by a system of connective tissues with four metacarpals distally. Two ligaments identified in the YP carpus were postulated to be homologous to the critical ligaments of the human carpus: RIL and DIC. The RIL was found to be homologous to the human SLIL in placement and function, connecting the radial carpal bone (RCB) to the intermediate carpal bone (ICB), which are positionally homologous to the scaphoid and lunate, respectively (Figure 1). The RCB and ICB articulate proximally with the radius. Unlike humans, the radius and ulna are fused and do not supinate or pronate relative to each other. The RIL, like the SLIL, has three main components: volar, dorsal, and membranous (Wolfe and Kakar, 2022). The YP DIC connects the proximal and distal carpal row through attachment on the dorsal face of the RCB laterally to the fourth carpal bone (Figure 1). Additionally, three tendons homologous to the human wrist flexors and extensors, namely, the superficial dorsal extensor tendon (SDE); the superficial radial carpal flexor tendon (SRCF); and the superficial ulnar carpal flexor tendon (SUCF) were also identified.

2.2. Specimen Preparation

Twelve YP forelimb specimens (5M, 7F, average age: 18 months; average weight: 55.4 kg) were acquired from an unrelated study approved by the institution’s animal care and use committee (IACUC, Rhode Island Hospital, Providence, US). The skin was removed, phalanges disarticulated, and the base of the olecranon was removed from the proximal end of the radius, leaving about 10 cm of the distal radius. The SDE, SRCF, and SUCF tendons were isolated, and whip stitched for attachment of constant force springs. The distal metacarpals and proximal radius and ulna were potted in fast setting urethane resin (SmoothOn, Macungie, PA). Once cured, six non-coplanar fiducial marker holes (diameter: 2 mm) were drilled to random depths into each potting fixture for registration of specimen-specific anatomical coordinate systems to the coordinate system of the robotic musculoskeletal simulator.

2.3. CT Imaging and Post-Imaging Processing

Each specimen underwent computed tomography (CT) scanning (GE, Milwaukee, WI) and post-processing for defining specimen specific anatomical coordinate systems, as described previously in homologous testing of human wrists (Badida et al., 2020). After CT scanning, the radius, ulna, metacarpals, and fiducial holes were segmented from the CT images to create 3D surface models using MIMICS v19 (Materialise, Leuven, Belgium). The surface models were exported to Geomagic Wrap (3D Systems^®, SC, US) and the coordinates of the anatomical and fiducial points on the radius, metacarpals, and their respective potting were identified.

2.4. YP Biomechanics Testing Procedure

A. Mounting and Digitization

Using a pre-established robotic protocol developed to determine human wrist biomechanics (Badida et al., 2020), a 6-DOF robot (KUKA KR6 R700, Augsburg, DE) operated by a LabView based software simVITRO (Cleveland Clinic, OH) was used for biomechanical testing. Each specimen was thawed for 10–12 hours before testing. For testing, the radial potting was secured to the base of the robot and the metacarpal potting was secured to the robot end-effector (Figure 2). The isolated tendons (SDE, SRCF, and SUCF) were attached to constant force springs (4.45 N each, McMaster-Carr, Robbinsville, NJ). A 6-DOF digitizing probe (Optotrak, NDI, Ontario, CA) was used to locate the load cell and the fiducial markers to register the anatomical landmarks from CT image’s reference frame (CT_GCS) to the robot’s reference frame (ROB_GCS) and create specimen specific radial (RCS) and metacarpal (MCS) coordinate systems (Badida et al., 2020).

Figure 2.

Left porcine forelimb mounted to the KUKA robot. Radius and ulna are fixed to the base, and the metacarpals secured to the end-effector. The radius coordinate system (+Rx = Volar, +Ry = Proximal, and +Rz = Radial) and the metacarpal coordinate system (+Mx = Volar, +My = Proximal, +Mz = Radial) in robot’s frame of reference registered via the digitization of fiducial markers.

B. Joint Coordinate System (JCS) and Neutral Position

The JCS generated within the ROB_GCS described the position of the MCS relative to the RCS (Figure 2). The JCS was generated with the flexion-extension (FE) axis (Z) fixed in the radius, the pronation-supination (PS) axis (Y) fixed in the metacarpals, and a floating radioulnar deviation (RUD) axis (X). The neutral position was set as: (1) 0° flexion-extension, (2) 0° radial-ulnar deviation, (3) 0 Nm pronation/supination torque, and (4) 0 N joint forces. After setting these criteria, forces and torques less than ±1 N and ±0.05 Nm were achieved at the neutral position.

C. Testing Protocol

Three testing protocols were performed: 1) Multidirectional Range of Motion (MROM), 2) Carpal Pronation/Supination ROM (PSROM) and 3) Translational ROM (TROM). Each specimen was preconditioned for 3 cycles, and the entire protocol was completed within one day per specimen to minimize the re-thawing and re-digitization process. These test protocols were adopted from previous biomechanical testing studies of human wrists (Badida et al., 2021, 2020).

For the MROM tests, the specimen was rotated in 28 directions relative to its neutral position: four anatomical directions of flexion, extension, radial and ulnar deviation, and twenty-four directions that were combinations of these four anatomical directions (e.g., ulnar flexion, ulnar extension, radial flexion, and radial extension). For example, the testing direction midway between ulnar deviation and flexion was defined as ‘UF45’. Pronation and supination were set to 0° throughout the MROM tests. The metacarpals were rotated at 1°/sec until a resultant torque of 1 Nm, defined as the vector sum of torques in all 3 DOF was reached. From a total of 1008 MROM tests (no. of specimens × test conditions × directions = 12 × 3 × 28), 104 tests were manually stopped before 1 Nm was reached because the robot’s linkages were at risk of collision.

The carpal PSROM protocol was performed at neutral wrist position and at 45° of wrist rotation in the direction of UF75 (15° from the FE axis in the ulnar direction), which was determined to be the direction of the maximum ROM of the intact specimens. Specimens were then pronated and supinated at the two wrist positions at 1°/sec until a resultant torque of 1 Nm.

Lastly, the TROM tests were performed by translating the metacarpals in pure volar and dorsal directions with the wrist in neutral position at 1mm/sec until 30 N of force in the direction of translation was reached.

2.5. Ligament Sectioning

Each specimen was tested (1) intact, and after sequential sectioning of (2) RIL (referred to as RIL group), and (3) DIC (referred to as RIL+DIC group). The carpal capsule was opened to obtain a clear view of the dorsal plane, and all components of the RIL were carefully sectioned between the RCB and ICB ensuring the DIC was not unintentionally sectioned. An elevator tool was placed in the joint space between the RCB and ICB to confirm the thick, membranous portion of the RIL was fully sectioned through to the volar side. This same procedure was used to section the DIC that lies distal to the RIL. The joint capsule was sutured closed using a simple interrupted stitch between each condition.

2.6. Data Processing and Statistical Analysis

For the MROM protocol, wrist motion was described using the ZXY cardan angle sequence and ROM was calculated as $\sqrt{F{E}^2+RU{D}^2}$, the resultant of flexion-extension (FE) and radial-ulnar deviation (RUD) angles of the metacarpals with respect to the radius and ulna, relative to the neutral position. Torque-rotation curves (Figure 3) were fitted with 5^th order polynomial (MATLAB, Mathworks, Natick, MA) and analyzed for ROM and stiffness in all 28 directions. ROM was defined as the wrist rotation at torque values of 1 Nm and are presented as polar plots defining the ROM envelope of the YP carpus. Stiffness was calculated as the slope of the fitted polynomial between 0.75 Nm and 1 Nm. Of the 104 tests manually stopped, ROM values from 28 tests were extrapolated from the fitted polynomial as they exceeded the predefined threshold of 0.5 Nm. The remaining 76 tests were not extrapolated as they did not exceed 0.5 Nm and thus were designated as ‘missing’ values.

Figure 3.

Typical torque-rotation curves in the four primary anatomical directions during multidirectional range of motion (MROM) testing.

For the carpal PSROM protocol, ROM was calculated as the rotation about the PS axis at 1 Nm. For the TROM protocol, translational ROM was calculated as the displacement at 30 N of force in each respective direction. For both protocols, torque-rotation curves and load-displacement curves were fitted with 5^th order polynomials. Stiffness was calculated as the slope of the fitted polynomial between 0.75 Nm to 1 Nm for PSROM, and between 22.5 N to 30 N for TROM.

Mixed models were utilized to evaluate the differences in kinematics as a function of testing direction, ligament condition, and sex. The difference in ROM between male and female specimens were consistently less than 1°, accordingly reported results were not stratified by sex. Mixed models were chosen for the analytic method as they allow appropriate handling of data structures associated with repeated measurements and maximize the information from specimens with missing observations. The outcome distribution and model residuals were examined to determine the distribution and link function to be applied to each model. To satisfy the assumption of normality, the outcomes were logarithmically transformed where necessary. Classical sandwich estimation was used to further protect against possible model misspecification. Pairwise comparisons between study conditions (i.e., Intact, RIL, RIL+DIC) were conducted within the model via orthogonal contrasts. Repeatability testing on a single specimen for 9 cycles resulted in average standard deviations in ROM of 0.9° in intact, 2.9° in RIL cut and 0.7° in RIL+DIC cut across all directions. The Holm test was used to correct for multiple comparisons, maintaining a 2-tailed familywise alpha at 0.05 across the hypotheses tested. The data is reported using adjusted p-values and model-based means and 95% confidence intervals (CIs). An adjusted p-value<0.05 was used to determine statistical significance. All analyses were conducted in SAS version 9.4 (SAS Institute Inc., Cary, NC).

3. Results

3.1. Intact Biomechanics

The intact ROM envelope had a general elliptical shape with the mean principal axis oriented in an ulnar flexion to radial extension direction, 15.9° from flexion-extension (Figure 4A). Across all 28 directions, the largest ROM was 97.6° (CI: 85.0, 112.1) in the direction of ulnar flexion (UF75). In the four primary anatomical directions, the ROM was 36.8° (CI: 23.1, 58.6) for flexion, 9.4° (CI: 7.5, 11.9) for extension, 3.1° (CI: 2.2, 4.4) for radial deviation, and 6.0° (CI: 5.2, 7) for ulnar deviation. The intact stiffness envelope had an orientation orthogonal to the ROM envelope with its major principal axis oriented from ulnar extension to radial flexion, 100.4° from the flexion-extension axis. Specimens demonstrated the greatest stiffness in radial deviation 0.52 Nm/° (CI: 0.46, 0.57) and least stiffness in UF75 0.01 Nm/° (CI: −0.04, 0.06).

Figure 4.

ROM envelopes across the 28 motion directions during multidirectional range of motion (MROM) testing of the YP carpus for (A) Intact, (B) RIL transection, and (C) RIL+DIC transection conditions. Average (solid line), + 1 SD (dashed line), and − 1 SD (dotted line) (deg).

For carpal PSROM protocol, pronation ROM at the neutral position and at UF75 were 18.3° (CI: 15.1, 21.5) and 13.7° (CI: 9.7, 17.8), respectively (Figure 5). Supination ROM at the neutral position was 9.3° (CI: 7.8, 10.9) and less than pronation (p<0.05). Pronation ROM and supination ROM at UF75 were 13.7° (CI: 9.7, 17.8) and 19.9° (CI: 17.9, 22), respectively. Supination stiffness at the neutral position was 0.21 Nm/° (CI: 0.19, 0.24) and greater than pronation stiffness of 0.17 Nm/° (CI: 0.14, 0.2; p<0.05). Conversely, supination and pronation stiffness along UF75 were 0.15 Nm/° (CI: 0.13, 0.18) and 0.16 Nm/° (CI: 0.13, 0.19). For the TROM protocol, volar translation of 7.5 mm (CI: 6.4, 8.6) was marginally greater than dorsal translation of 7.2 mm (CI: 6.2, 8.2). Volar and dorsal translation stiffness were 7.9 N/mm (CI: 7.2, 8.6) and 8.2 N/mm (CI: 7.2, 9.4), respectively.

Figure 5.

Pronation and supination of all three conditions in the defined neutral position and the direction of max ROM which varied between pure flexion and the greater degrees of ulnar flexion. Significant differences (p-values<0.05) are shown with *.

3.2. Ligament-sectioned Biomechanics

Minimal kinematic changes were observed after ligament sectioning in all three test protocols; hence we only report the changes that demonstrated statistical differences. ROM extension in RIL and RIL+DIC increased (p<0.05) by approximately 2° to 11.1° (CI: 9.8, 12.6) and 12.1° (CI: 10.2, 14.2), respectively. Accordingly, stiffness in extension decreased after RIL+DIC to 0.23 Nm/° (CI: 0.19, 0.27) (p<0.05).

For carpal PSROM protocol at the neutral position, supination ROM in RIL+DIC increased (p<0.05) to 10.5° (CI: 8.9, 12.1) compared to 9.7° (8.2, 11.2; p<0.05) in RIL and 9.3° (CI: 7.8, 10.9; p<0.05) in intact. Pronation ROM remained greater than supination ROM across all conditions (p<0.05). Supination stiffness remained greater (p<0.05) than pronation in all conditions, whereas pronation stiffness of 0.15 Nm/° (CI: 0.13, 0.18) in RIL+DIC was lower than 0.17 Nm/° (CI: 0.14, 0.2) in intact (p<0.05).

For carpal PSROM in the UF75 position, pronation increased (p<0.05) by approximately 3° to 16.2° (CI: 12.4, 19.9) in RIL and 15.6° (CI: 11.1, 20) in RIL+DIC. Supination increased (p<0.05) by approximately 2° to 21.9° (CI: 18.9, 24.8) in RIL+DIC. Mean supination stiffness in RIL+DIC (0.12 Nm/°, CI: 0.09, 0.15) was less than the stiffnesses for both intact (0.15 Nm/°, CI: 0.13, 0.18; p<0.05) and RIL (0.15 Nm/°, CI: 0.12, 0.18; p<0.05).

For TROM protocol, volar translation increased (p<0.05) in RIL+DIC by approximately 1 mm to 8.4 mm (CI: 7.3, 9.4). Accordingly, dorsal stiffness decreased (p<0.05) in RIL+DIC to 8 N/mm (CI: 7.1, 8.4).

4. Discussion

The motivation for this study was to assess the Yucatan minipig as a potential preclinical animal model for the human wrist by quantifying carpal biomechanics in vitro during intact and ligament transection conditions. The YP was chosen because of similarities in anatomy, size, and cartilage physiology to the human carpus (Behrends et al., 2015; Stoeckl et al., 2021). The ligament sectioning sequence was modeled after the methodologies used in previous instability studies that quantified human cadaveric wrist kinematics by simulating wrist rotations under healthy and ligament injury conditions (Badida et al., 2021; Short et al., 2007). The RIL and DIC both insert on the dorsal surface of the RCB and were sectioned to mimic scaphoid and lunate carpal instability from an SLIL tear. We note, however, the YP DIC inserts on the carpal bones in a reversed diagonal pattern compared to humans; the YP DIC functionally connects the proximal row to the distal carpal row, but does not insert on the 1^st and 2^nd carpal bone (trapezium and trapezoid homologs) as it does in the human wrist (Wolfe and Kakar, 2022).

To the best of our knowledge, the biomechanics of the YP carpus has not been previously reported precluding any direct comparison of our findings. However, as the main objective was to evaluate the YP as a potential biomechanical model for the human wrist, we compared our findings to human wrist biomechanics. Our finding of 97.6° being the largest mean intact ROM in ulnar flexion is consistent with 98° in the same direction of human wrist as reported by Badida et.al (Badida et al., 2021) and 111.5° as reported by Crisco et al (Crisco et al., 2011). Our findings of mean intact ROM in principal directions of flexion, extension, radial and ulnar deviation were 36.8°, 9.4°, 3.1° and 6° respectively. These values are substantially less than the 65° and 75° in flexion, 64° and 82° in extension, 21° and 25° in radial deviation, and 42° for ulnar deviation in both studies, as reported for human wrist specimens (Badida et al., 2021; Crisco et al., 2011). However, the orientation of the ROM envelope was similar to that previously reported for human specimens, which has been reported to align with the dart-thrower’s motion (Moritomo et al., 2007). Even though the motion envelopes have the same elliptical shape and orientation, the human wrist inherently possesses a greater total ROM. The pig is a fore bearing animal that places approximately 60% of its weight on the forelimbs when standing (Kaiser et al., 2020); thus, passively limiting extension likely reduces the cost of metabolism during gait and increases stability. We also measured pronation and supination of the YP carpus. However, because the YP radius and ulna are fused, the pronation and supination data reflect the rotation of the carpus with respect to the radius, not forelimb rotation. We also found that the intact volar translation was greater than dorsal translation but could not locate any similar testing in the human wrist for comparison.

We hypothesized, based on previous biomechanical studies of the human wrist (Badida et al., 2021; Padmore et al., 2019; Pérez et al., 2019), an increase in ROM and a decrease in stiffness of the YP carpus after transection of two stabilizing ligaments, the RIL+DIC, but not after transection of the RIL alone. In humans, the dorsal component of the SLIL is the strongest and the primary stabilizer with the ability to withstand forces up to 260 N, while the membranous portion does not provide any notable stabilization (Andersson, 2017). While we were able to detect increases in ROM, and accordingly decreases in stiffness, likely due to our repeated measures study design and precision testing with the robotic musculoskeletal simulator, these differences were quite subtle in magnitude, in contrast to carpal instability that is exacerbated after injury to the extrinsic carpal ligaments (Bui et al., 2021). In comparison to human wrist that showed mean ROM increases of 3.3° in radial deviation, 8.8° in radial extension and 13° in ulnar flexion after DIC cut (Badida et al., 2021), the YP carpus exhibited a marginal increase of ~3° in extension only, which was greater than the variability in our repeatability testing. Thus, the structures that are responsible for maintaining the overall stability of the porcine carpus remains to be determined.

In this study, we chose 1 Nm torque and 30 N force as the end condition based on pilot studies that demonstrated that these values were sufficient to characterize the non-linear behavior of the YP carpus without damaging the specimen. A 4.45 N constant force spring was applied to each major tendon to provide joint stability, consistent with our previous studies of human wrists (Badida et al., 2020; Crisco et al., 2011). In comparison to previous YP knee models that observed gait patterns, forelimb gait has not been documented, therefore specific loading of the wrist is unknown and could be assessed in the future (Boakye et al., 2020). Lastly, future in vitro or in vivo testing could implement motion tracking of individual carpal bones (Akhbari et al., 2019; Stoesser et al., 2017), during robotic testing, CT imaging, or observation of porcine gait to understand if there are effects of intact and ligament transection conditions on carpal bone kinematics.

A limitation of this study is that individual carpal bone motion was not measured. Secondly, we only tested a single ligament transection sequence of RIL and then DIC. Further exploration into the additional stabilizing ligaments in the carpus may yield greater ROM differences. Lastly, due to the high ROM in ulnar flexion directions of specimens, from 1008 tests, ROM values for 28 tests were extrapolated from torque-rotation curves and a total of 76 tests were considered ‘missing’ as they could not be extrapolated. Our findings are limited to the specific robotic control protocol we employed, and alternate control protocols, such as a displacement or other hybrid control as previously used in ACL and spine testing (Bell et al., 2015, 2013), could yield different results.

In summary, we assessed the Yucatan minipig by quantifying the biomechanics of the intact and ligament transected YP carpus for its use as a potential animal model for the human carpus. The YP carpus motion envelope being similarly shaped and oriented as the human carpus, along with the described anatomical and physiological similarities, provide evidence that the YP has the potential to serve as a preclinical animal model. Its feasibility as a preclinical model will, however, depend on the specific research question being asked. Documenting the similarities and differences between the porcine and human carpus could inform its use as a preclinical model for testing of new carpal bone and wrist arthroplasties, in vitro or in vivo, by providing a functional, physiological system for evaluating implant features and new technologies.

Figure 6.

Subsequential translation measurements in all three conditions of ligament sectioning. Significant differences (p-values<0.05) are shown with *.