Disease of the podotrochlear apparatus, also known as “navicular disease,” is a common, potentially career-ending and life-threatening cause of thoracic limb lameness in horses. Pathology of the soft tissues, in addition to the distal sesamoid bone (DSB), may be involved. Since the DSB is a dynamic structure, it is plausible that soft tissue pathologies, such as adhesions, may alter its normal function through the course of the disease. The motion of the DSB in normal thoracic limbs is not well understood.
Kinematics is a branch of classical mechanics that studies the motion of objects in 3-dimensional (3-D) space, without regard to the forces applied to those objects. Such studies typically assess the translation, defined as uniplanar sliding motion, and rotation about predetermined axes, in order to analyze object motion in 3-D space. Kinematics has been applied to describe coxofemoral arthritis,1 prosthetic limbs,2 and diseased gaits such as Parkinson’s disease3 in people. Kinematics in nonequid veterinary species has been used to compare normal stifles to those with cranial cruciate ligament rupture in dogs4 and to assess the use of hydrotherapy postoperatively for rehabilitation following canine cranial cruciate ligament repair.5 In horses, the earliest kinematic studies6–9 used photography and film to define various gaits but have since modernized to incorporate high-speed video analysis of more specific anatomical regions like the spine and distal thoracic and pelvic limbs. Most recent biomechanical studies4,5,10,11 have focused on lameness but often use skin surface markers, which have a tendency to slide over the palpable bony landmarks and cannot assess deeper bony structures. An algorithm was devised to correct for skin displacement in equine kinematic gait analysis studies,10 although it has limited clinical application.
Other kinematic studies11 have been performed to estimate the concussive forces to the distal limb joints but cannot evaluate deep, bony structures of the foot, such as the DSB. Kinematic evaluation of deeper structures in the equine stifle has been performed using radioisometric analysis whereby radiopaque markers were embedded in the articular cartilage; however, studies12 such as these often require the removal of soft tissues, which may potentially alter results. One such study13 in ponies determined there was substantial movement of the DSB relative to the middle phalanx. This study13 used pony thoracic limbs transected at the humerus to assess joint angles and excluded linear translation of the DSB. Removal of proximal soft tissues in cadaveric kinematic studies is common, although the effect on distal limb bony structures in the horse has not been determined. The role of the superficial (SDFT) and deep digital flexor tendons (DDFT) and muscles in augmenting the kinematics of distal orthopedic structures such as the DSB is unknown.
The objectives of this study were (1) to quantify translation and angular rotation of the DSB using a combination of computed tomography (CT) and medical modeling software, and (2) to compare the kinematics of the DSB in full thoracic limbs to those with proximal flexor tendon attachments transected (partial limbs). We hypothesized that the motion of the DSB of thoracic limbs was quantifiably different from zero and that the movement of the DSB would be greater in partial limbs compared to full limbs.
Materials and Methods
Bilateral thoracic limbs from 15 skeletally mature horses euthanatized for reasons unrelated to thoracic limb lameness were collected (n = 30 limbs). Palmaroproximal-palmarodistal radiographs at 65° were obtained with a MinXRay HF8015 (60 kVp, 0.06 mAs) to ensure DSBs were free from any radiographically evident foot pathology (enthesiophytes, corticomedullary changes, etc). Limbs were categorized into 3 groups: 12 partial limbs (transected immediately proximal to the chestnut), 8 full limbs (collected from scapula to digit), and 10 full limbs, which were studied and subsequently transected and evaluated as partial limbs (matched pairs). The matched pairs underwent kinematic CT evaluations as full limbs, were then transected, and underwent additional kinematic CT evaluation again as partial limbs. The hooves were trimmed and cleaned and a 1.3-cm hole was drilled through the dorsodistal aspect. A 114-kg capacity cable tie was loosely placed proximal to the carpus (so as not to produce force on the DDFT at the metacarpal or pastern regions), and a ratchet device was used to achieve different positions. Bone volume CT images were acquired at 0.5 X 0.3-mm data volumes (Toshiba Aquillion CT Scanner, Cannon Medical Systems) in 3 positions: neutral (metacarpophalangeal [MCPJ] joint angle of 180° with the dorsal bone cortices aligned with a 2.54 X 10-cm wooden board), 200° MCPJ joint extension, and maximum attainable MCPJ flexion (110°) (Figure 1). MCPJ joint angles were measured with a goniometer centered over the epicondylar fossa of the lateral condyle. The CT images were imported into a medical segmenting software (Materialise Mimics v. 21) to create the 3-D models of the DSB, middle phalanx, and distal phalanx, which were used for spatial manipulation (Materialise 3-Matic v. 15). The middle phalanges from all 3 positions for each thoracic limb were aligned and superimposed over each other while maintaining the spatial relationship between them and their respective DSBs (Figure 2). Maximal mediolateral length, dorsopalmar width, and proximodistal height of each DSB were recorded as measured in 3-D space. A coordinate system was established based on the anatomic axes of the DSB (dorsopalmar, proximodistal, and mediolateral), and the translation distance each DSB traveled in each plane (eg, dorsally or palmarly, etc) was measured. This was designed to make measurements transposable to similarly positioned standing lateromedial radiographs in a horse with a neutral hoof-pastern axis (Figure 3). Images acquired from specimens in the neutral position were used as the reference location against which all other positions were compared. Measurements were acquired via creation of a transformation matrix, which details linear translation distances required to superimpose each DSB over the neutral DSB in each plane in 3-D space, followed by rotation about each axis to match the distal sesamoid in neutral position. Finally, translation in each direction was calculated as a percentage of the mean mediolateral length, dorsopalmar width, or proximodistal height of the DSBs.
Bone volume CT images were acquired with the limbs in 3 different positions: neutral, with the dorsal cortices aligned using a board to achieve a neutral hoof-pastern axis (A), 200° metacarpophalangeal (MCPJ) extension (B), and maximal attainable MCPJ flexion (110°) (C).
Citation: American Journal of Veterinary Research 83, 7; 10.2460/ajvr.21.07.0090
Spatial manipulation software images of the distal sesamoid bone (DSB) and middle and distal phalanges. A and B—Charcoal images were acquired with the limb in neutral position, red images were acquired with the limb in extension, and yellow images were acquired with the limb in flexion. A—The middle phalanges of each position are superimposed over one another, maintaining their spatial relationship to their respective DSBs. Note the charcoal (neutral) image alignment, which demonstrates a neutral hoof-pastern axis. B—The middle and distal phalanges are hidden, demonstrating the range of motion of the DSB.
Citation: American Journal of Veterinary Research 83, 7; 10.2460/ajvr.21.07.0090
Radiographic representation of range of motion of the DSB and inertial axes of the DSB in full limbs. Length of arrows corresponds to distance of travel, which demonstrates the highly variable position of the DSB in the hoof capsule.
Citation: American Journal of Veterinary Research 83, 7; 10.2460/ajvr.21.07.0090
Graphical representation of means ± 95% CI for translation and rotation of the DSB. A—Dorsopalmar translation was significantly different (*) between extension (Ext) and flexion (Flex) positions for full and partial limbs. B—Mediolateral translation was not different between full and partial limbs, or between positions. C—Proximodistal translation is significantly different between full and partial limbs (†) and between extension and flexion of full limbs (‡). D—Rotation about the dorsopalmar axis of the DSB is not different between full and partial limbs or between positions. E—Rotation about the mediolateral axis is significantly different between extension and flexion for full limbs only (§). F—Rotation about the proximodistal axis is not different between full and partial limbs, or between positions.
Citation: American Journal of Veterinary Research 83, 7; 10.2460/ajvr.21.07.0090
Statistics
Data were evaluated for normality using the Shapiro-Wilk test. A two-way ANOVA test was performed to determine differences among the groups with the limb as the subject and limb position (flexion, extension, or neutral) as one factor and the limb condition (full or partial limb) as the other factor. A post hoc Tukey’s test was performed to determine differences between the groups for pair-wise comparisons. Significance was set at P ≤ 0.05. Data were reported as means ± SE.
Results
Thoracic limbs from 15 horses were enrolled in the study comprising a total of 30 limbs: 12 partial limbs, 8 full limbs, and 10 matched full and partial limbs. Horse breeds included 7 American Quarter Horses, 4 Thoroughbreds, 2 Tennessee Walking Horses, 1 Irish Draft, and 1 Warmblood. Mean age of horses was 16.8 ± 1.98 years, and mean weight was 482 ± 25.85 kg. There were 9 mares, 5 geldings, and 1 stallion.
Distal sesamoid bone measurements
Mean mediolateral length, dorsopalmar width, and proximodistal height of the DSBs were 5.59 cm (± 0.06), 1.54 cm (± 0.02), and 2.21 cm (± 0.04), respectively.
Translation
Mean values of translation of the DSB in each limb category and position are summarized in Table 1. Mean dorsal and palmar translation accounts for approximately ([1.44 cm dorsal + 0.23 cm palmar]/1.54 cm mean width) 108% of the dorsopalmar width of the DSB. There was a significant difference (P < 0.001) in dorsal translation of the DSB between paired full limb extension and paired partial limb extension (Figure 4). In addition, within each limb condition (full or partial) there was a significant difference between the extension position and the other positions (flexion, P < 0.001). However, there was no significant difference among translation in the dorsopalmar plane for paired partial limbs or between flexion positions for paired full limbs (P = 1.0).
Mean translation from neutral position.
| +DorsoPalmar− | −ProximoDistal+ | −MedioLateral+ | +DorsoPalmar− | −ProximoDistal+ | −MedioLateral+ | |
|---|---|---|---|---|---|---|
| Full limbs | Partial limbs | |||||
| Extension | 1.35 ± 0.18 | 6.81 ± 0.74 | −0.07 ± 0.12 | −1.01 ± 0.3 | 0.08 ± 1.8 | −0.15 ± 0.2 |
| Flexion | 0.25 ± 0.1 | −2.01 ± 0.56 | 0.09 ± 0.2 | −0.1 ± 0.17 | 0.22 ± 1.03 | 0.05 ± 0.07 |
| Paired—full | Paired—partial | |||||
| Extension | 1.44 ± 0.4a,b | 5.35 ± 0.7d,f | 0.21 ± 0.16 | 1.3 ± 0.23a,c | 6.22 ± 0.6d,g | 0.08 ± 0.11 |
| Flexion | 0.23 ± 0.18b | −1.94 ± 0.37e,f | 0.01 ± 0.07 | 0.06 ± 0.13c | −1.12 ± 0.36e,g | 0.00 ± 0.14 |
Statistically significant differences.
Mean proximal and distal translation accounts for approximately ([1.94 cm proximal + 5.35 cm distal]/2.2 cm mean height) 331% of the proximodistal height of the DSB. Proximodistal translation was significantly different between full limb extension and partial limb extension, between full limb flexion and partial limb flexion, and between the extension and flexion positions for full and partial limbs (P < 0.001).
Mean medial and lateral translation accounts for approximately ([0.21 cm medial + 0.01 cm lateral]/5.59 cm mean length) 4% of the mediolateral length of the DSB. Translation was not significant in the mediolateral plane for any position, and no difference was observed between full and partial limbs (P = 0.64).
Rotation
Mean values of rotation of the DSB in each limb category and position are summarized in Table 2. No statistical significance in mean rotation about the dorsopalmar axis was detected between full and partial limbs or between positions for rotation about the dorsopalmar axis of the DSB (P = 0.73).
Mean rotation from neutral position.
| DorsoPalmar | ProximoDistal | MedioLateral | DorsoPalmar | ProximoDistal | MedioLateral | |
|---|---|---|---|---|---|---|
| Full limbs | Partial limbs | |||||
| Extension | 0.09 ± 0.39 | 0.39 ± 0.57 | 15.54 ± 1.8 | 0.01 ± 0.24 | 0.13 ± 0.23 | 1.92 ± 4.19 |
| Flexion | 0.77 ± 0.53 | 0.23 ± 0.21 | 4.16 ± 1.23 | 0.49 ± 0.22 | 0.08 ± 0.12 | 0.41 ± 2.46 |
| Paired—full | Paired—partial | |||||
| Extension | 0.39 ± 0.45 | 0.49 ± 0.42 | 11.18 ± 2.41a | 0.81 ± 0.47 | 0.6 ± 0.57 | 11.89 ± 1.02 |
| Flexion | 0.52 ± 0.42 | 0.08 ± 0.25 | 2.59 ± 1.25a | 0.49 ± 0.42 | 0.49 ± 0.37 | 1.3 ± 1.18 |
Statistically significant differences.
No statistical significance was detected in mean rotation about the proximodistal axis between full and partial limbs or between positions for rotation about the proximodistal axis of the DSB (P = 0.70).
Rotation about the mediolateral axis was statistically different (P < 0.001) between extension and flexion positions in full limbs (Figure 4). There was no significant difference in mediolateral axis rotation between full and partial limbs (P = 0.31).
Discussion
To the authors’ knowledge, this study represents the first quantification of translation and angular rotation of the DSB during different positions of the limb. The DSB lies deep to the hoof capsule, making surface marker assessment impractical. As imaging technology advances, the DSB is being reconceptualized as an enthesis organ designed to redistribute forces applied to the distal limb.14 In this report, a combination approach of CT and medical imaging software was used to analyze the movement of the DSB. Our first hypothesis that DSB motion deviates from zero (static) was supported, and therefore, the DSB should be considered a dynamically mobile structure. We found that as the foot is brought into extension, the DSB moves distally and rotates about its mediolateral axis.
Our second hypothesis that translation and rotation of the DSB would be greater in partial limbs than in full limbs was not supported. We found that translation and rotation of the DSB was greater in full limbs than in partially transected limbs. Consequently, dorsopalmar translation of the DSB decreases by a combined 19% of its width, proximal translation decreases by 16% of its height, and mediolateral translation decreases by 2% by severing the soft tissue attachments in converting them from full to partial limbs. Interestingly, the distal translation of the DSB increased by 37% of the DSB’s height in partial limbs compared to full limbs. This suggests that transecting the SDFT and DDFT through their proximal muscular origins (as occurs in collection of partial limbs) leads to a reduction in DSB motion; using full limbs with intact musculotendinous units permits the full range of motion of the DSB to occur. Initially, we believed that severing the SDFT and DDFT muscles would reduce forces on the DSB through decreased tension inside the hoof and thus lead to a greater freedom to move within the hoof capsule during motion. However, the evidence from this study suggests the converse, that the forces exerted by the SDFT and DDFT across the DSB during different positions (as can only be achieved by full limbs with intact proximal muscular units) may be important for DSB rotation and translation.
The difference in dorsopalmar translation between full and partial limbs and between extension and the other positions is likely a result of a vector sum of dorsal force by the DDFT on the DSB. The DDFT is composed of a muscular portion that originates proximally on the caudal aspects of the humerus, radius, and ulna, a tendinous portion distally that inserts on the semilunar notch of the distal phalanx, and is joined by an accessory ligament to the deep digital flexor tendon (ALDDFT) that originates on the caudal aspect of the third metacarpal bone.15,16 We believe that as the foot is brought into extension, the origin of the DDFT on the proximocaudal humerus and its insertion on the distal phalanx act as anchor points as the tendon tightens between the 2, causing the DSB to translate dorsally. Consequently, there was no significant translation in the dorsopalmar plane in the flexion position for either full or partial limbs. This is likely because the DDFT is no longer under tension, resulting in a lack of a dorsal force acting on the DSB.
Similarly, we believe the difference in proximodistal translation of the DSB between the extension and flexion positions is likely a result of the distal sesamoidean (impar) ligament applying tension on the DSB as the foot is brought into extension. The distal sesamoidean ligament originates proximally on the distal border of the distal sesamoidean bone and inserts distally on the proximopalmar aspect of the distal phalanx.15,16 As the foot is brought into extension, the distal phalanx likely exerts tension on the distal sesamoidean ligament, which in turn causes the DSB to translate distally; conversely, when the foot is brought into flexion, the tension is reduced, allowing the DSB to translate proximally.
The lack of differences in translation in the mediolateral plane in all positions and for both full and partial limbs is likely due to limitation by the proximal sesamoidean ligaments of the DSB, as well as a lack of experimental forces exerted in a mediolateral direction. As the foot is brought into extension, the forces exerted by the proximal sesamoidean ligaments should be symmetrical under normal circumstances. The dorsal aspect of the DSB is also mildly incongruent where it articulates dorsally with the distal aspect of the middle phalanx, which may result in a physical barrier to mediolateral movement in a lock-and-key fashion.
Angular rotation of the DSB was found about the mediolateral axis between extension and flexion positions and between full and partial limbs. The rotation around this axis is likely a result of the simultaneous actions of the distal sesamoidean ligament applying tension on the DSB distally and the dorsal force of the DDFT, which causes the DSB to move dorsally as described previously. Interestingly, rotation about the dorsopalmar axis was different between full and partial limbs. Rotation about this axis was not expected, and the authors believe that this may have resulted from torsional rotation of the foot that occurred in partial limbs (but not full limbs), when the ratchet device was used to bring the limb into the flexion position due to a lack of stabilizing proximal soft tissue structures. However, while there was a difference between full and partial limbs, there was no difference in the quantified rotation about the dorsopalmar axis between the positions, which is likely more relevant to the standing horse. Similarly, no appreciable difference between positions or between full and partial limbs was observed for rotation about the proximodistal axis. This was expected, as the DSB does not have any structures to exert forces that would cause it to rotate in this direction and would be physically hindered by the distal phalanx.
Our study suggests that as the foot is brought into extension, the DDFT and distal sesamoidean ligament exert forces on the DSB causing simultaneous dorsodistal translation and rotation about the mediolateral axis. The most likely cause for these differences in translation between full and partial limbs is the transection of the proximal soft tissue attachments of the DDFT. By transecting the proximal muscular attachments, it is likely that the DDFT was able to slide in the tendon sheath without applying the normal dorsally compressive force to the DSB, resulting in the reduced motion of the DSB seen in the partial limbs.
While the inclusion criteria eliminated known lameness issues, the distal limb morphology of the included specimens was not identical. Very little trimming was performed in an attempt to leave the feet “as-is” and therefore representative of a cross-section of horses that might be encountered in clinical practice. Naturally, a protocol for standardization for “normal” hoof morphology may be required in future studies involving standing, living horses, where abnormalities such as long toes may change the kinematics of the DSB. However, because the cadaveric specimens were not bearing weight or subjected to the same forces as living horses, the authors believe the data gathered here to be an accurate representation of the range of motion of the DSB.
The decreased movement of the DSB in the partial limbs observed in this study has implications for future kinematic studies. The data acquired in this study suggest that kinematic studies involving the equine distal limb require the use of full limbs, including the scapula and proximal soft tissue attachments of the flexor tendons, in order to more accurately mimic the forces these soft tissues apply to the bony structures of the distal limb. This will, in turn, more accurately represent the movement of the DSB in living horses, in which extension and compression due to the horse’s weight will likely be the primary forces on the DSB. The authors anticipate that there will be an increase in the flexion component of DSB translation relative to the results reported here due to the compression applied by the standing weight of the horse. The expectation is that the data acquired in this study may help to serve as a baseline for prognostication of horses with pathologic disease of the podotrochlear apparatus; however, further study is needed.
A limitation of this study includes mild carpal flexion that occurred when the cadaveric limbs were positioned in lateral recumbency due to the concave nature of the CT table. This could be corrected with a splint in future studies to more accurately represent a weight-bearing position. The proximal attachments of the muscular heads of the DDFT and the ligamentous attachments of the ALDDFT may be affected by mild carpal flexion, which may result in a subsequent decreased estimation of the full extent of DSB in extension.
In conclusion, the equine DSB translates dorsally and distally about a mediolateral axis of rotation as the foot is brought into extension. The proximal superficial and deep digital flexor tendon soft tissue attachments have a significant, previously unrecognized effect on the kinematics of the DSB in horses. Future ex vivo distal limb kinematic studies should use full limbs including the scapula in order to more accurately represent the forces exerted on distal limb structures by the proximal soft tissue attachments.
Acknowledgments
Funding was provided by an intramural resident College of Veterinary Medicine research grant at the University of Florida.
The authors declare that there were no conflicts of interest.
Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing and publication of the manuscript.
The abstract for this article was presented at the American College of Veterinary Surgeons Symposium in Las Vegas, NV on October 19, 2019.
References
- 1. ↑
Hodge WA, Fijan RS, Carlson KL, Burgess RG, Harris WH, Mann RW. Contact pressures in the human hip joint measured in vivo. Proc Natl Acad Sci. 1986;83:2879–2883.
- 2. ↑
Segal AD, Orendurff MS, Klute GK, et al. Kinematic and kinetic comparisons of transfemoral amputee gait using C-Leg and Mauch SNS prosthetic knees. J Rehab Res Dev. 2006;43:857–870.
- 3. ↑
Morris ME, McGinley J, Huxham F, Collier J, Iansek R. Constraints on the kinetic, kinematic and spatiotemporal parameters of gait in Parkinson’s disease. Hum Mvmt Sci. 1999;18:461–483.
- 4. ↑
Brown S, Stubbs NC, Kaiser LJ, Lavagnino M, Clayton HM. Swing phase kinematics of horses trotting over poles. Equine Vet J. 2015;47:107–112.
- 5. ↑
Harrison SM, Whitton RC, Kawcak CE, Stover, SM, Pandy MG. Evaluation of a subject-specific finite-element model of the equine metacarpophalangeal joint under physiological load. J Biomech. 2014;47:65–73.
- 7.
van Weeren R. Equine biomechanics: from an adjunct of art to a science in its own right. Equine Vet J. 2012;44:506–508.
- 8.
Faber M, Johnston C, Schamhardt HC, van Weeren PR, Roepstorff L, Barneveld A. Three-dimensional kinematics of the equine spine curing canter. Equine Vet J Suppl. 2001;33:145–149.
- 9. ↑
Egan S, Brama P, McGrath D. Research trends in equine movement analysis, future opportunities and potential barriers in the digital age: a scoping review from 1978-2018. Equine Vet J. 2019;51:813–824.
- 10. ↑
Roach JM, Pfau T, Bryars J, Unt V, Channon SB, Weller R. Sagittal distal limb kinematics inside the hoof capsule captured using high-speed fluoroscopy in walking and trotting horses. Vet J. 2014; 202:94–98.
- 11. ↑
Ratzlaff MH, Wilson PD, Hyde ML, Balch OK, Grant BD. Relationships between locomotor forces, hoof position and joint motion during the support phase of the stride of galloping horses. Acta Anatomica. 1993;146:200–204.
- 12. ↑
Halley SE, Bey MJ, Haladik JA, Lavagnino M, Arnoczky SP. Three dimensional, radioisometric analysis (RSA) of equine stifle kinematics and articular surface contact: a cadaveric study. Equine Vet J. 2014;46:364–369.
- 13. ↑
van Dixhoorn ID, Meershoek LS, Huiskes R, Schamhardt HC. A description of the motion of the navicular bone during in vitro vertical loading of the equine forelimb. Equine Vet J. 2002;34:594–597.
- 14. ↑
Osborn ML, Cornille JL, Blas-Machado U, Uhl EW. The equine navicular apparatus as a premier enthesis organ: functional implications. Vet Surg. 2021;50:713–728.
- 15. ↑
Dyce KM, Sack WO, Wensing CJG. The forelimb of the horse. In: Textbook of Veterinary Anatomy. 4th ed. Saunders Elsevier; 2010:586–623.
- 16. ↑
Budras KD, Sack WO, Röck S. Anatomy of the Horse: an Illustrated Text. 2nd ed. Mosby-Wolfe; 1994:10–11.
