Chapter 44The Tarsus

Anatomical Considerations

The tarsus consists of the tarsocrural, talocalcaneal, talocalcaneal-centroquartal (proximal intertarsal), centrodistal (distal intertarsal), and tarsometatarsal joints. The bones include the talus and calcaneus and the central, first and second fused, third, and fourth tarsal bones.¹ The tarsocrural joint is a ginglymus joint based on the shape of deep grooves on the cochlear articular surface of the distal end of the tibia with the extensive surface of the trochlea of the talus. The articulation of these joints is at an angle of 12 to 15 degrees dorsolateral to the sagittal plane of the limb. The talocalcaneal-centroquartal (proximal intertarsal), centrodistal, and tarsometatarsal joints are plane joints and are capable of only small amounts of gliding (shear) movement.

The tibia follows an almost circular path along the talar ridges, and most movement occurs in the tarsocrural joint; however, there are translations and rotations at other sites.² Hock flexion is associated with abduction of the distal part of the limb caused by an oblique axis of motion.³ There is outward rotation of the third metatarsal bone (MtIII) during the swing phase, followed by a small inward rotation during stance, when the hock is compressed. Thus at the gallop the hindlimbs swing outside the forelimbs as the tarsus flexes and then move back toward the midline as the tarsus extends before landing. Excessive rotation leads to visible wobbling of the tarsus. There is a locking mechanism among the talus, the central and third tarsal bones, and the MtIII.⁴ A process on the distal plantar aspect of the talus fits into an indentation on the proximal aspect of the central tarsal bone; a ridge on the distal aspect of the central tarsal bone engages with a fossa on the proximal aspect of the third tarsal bone. A triangular plantar process of the third tarsal bone prevents the third tarsal bone from sliding across the proximal aspect of the MtIII. The oblique dorsal ligament extends distally from a small origin on the talus, spreading across the central and third tarsal bones, limiting motion.

Biomechanical studies have shown greatest compression on the distal medial aspect of the tibia and the proximal lateral aspect of the MtIII, suggesting that compressive load is transferred from medial to lateral through the tarsus. Radiopharmaceutical uptake (RU) in normal horses is greatest dorsally and laterally in the distal aspect of the tarsus, suggesting increased adaptive bone modeling in response to high load.⁵ There are differences in subchondral bone and cartilage thickness in the central and third tarsal bones from medially to laterally, varying dependent on intensity and type of exercise history, reflecting loading.^6,7 The distal tarsal joints function to absorb energy in the early stance phase and contribute to propulsion in the late stance phase. Conformation influences tarsal function.⁸ Horses with large tarsal angles (>165.5 degrees) have been shown to have less flexion and less energy absorption in the impact phase at the trot compared with horses with intermediate (155.5 to 165.5 degrees) or small (<155.5 degrees) angles.⁸ Horses with large hock angles generated less vertical impulse than horses with small angles, and net extensor force was lower.⁸ Thus large tarsal angles may result in less propulsion and less absorption of concussion, which may influence both performance and soundness. However, horses with small hock angles had greater flexion during the stance phase, which may compress the dorsal aspects of the bones. Abnormal tarsal conformation can lead to lameness. Anecdotally, horses with small hock angles (sickle hocks) often develop osteoarthritis (OA) of the distal hock joints, curb, and soft tissue injury in the distal, plantar aspect of the hock and may be prone to the development of fracture of the central and third tarsal bones. Horses with large hock angles (straight-hock conformation), surprisingly, may be less prone to the development of tarsal pain, but they appear predisposed to the development of suspensory desmitis and stifle region pain.

Flexion of the stifle and hock is synchronous because of the reciprocal apparatus; there is rapid flexion of the tarsus at the beginning of the stance phase of the stride and maximal extension at the end of the stance phase. There is a peak of flexion in the middle of the swing phase. There is also coupling of movement of the tarsus and fetlock in flexion and extension because of passive action of the superficial digital flexor tendon and the long digital extensor tendon, respectively.

The fibrous part of the combined tarsal joint capsule surrounds all synovial compartments and attaches proximally to the tibia and distally to the distal tarsal bones, blending with the collateral ligaments (CLs), and to the metatarsal bones.² The dorsomedial aspect of the joint capsule is thin and uncovered by any tendons or ligaments and forms a fluctuant swelling over the medial trochlea of the talus. The proximal aspect of the plantar fibrous joint capsule is also thin and extends proximally about 5 cm caudal to the distal aspect of the tibia.² The plantar distal aspect of the joint capsule consists of the plantar and tarsometatarsal ligaments, which are thick and tightly adherent to the distal tarsal bones. The tarsus is composed of four synovial compartments. The tarsocrural compartment lubricates the tarsocrural joint and the dorsal aspect of the proximal intertarsal joint. The tarsocrural compartment is the largest compartment and is composed of four pouches: the dorsomedial, dorsolateral, plantaromedial, and plantarolateral. This provides several sites at which arthrocentesis can be performed. The proximal intertarsal synovial sac lines the talus and calcaneus proximally and the plantar aspect of the central and third tarsal bones distally and communicates dorsally with the tarsocrural joint. The large fenestration between the tarsocrural and the proximal intertarsal joints allows loose osteochondral fragments to move freely between the conjoined joints. Disease of one joint obligatorily involves both compartments, because they function as one. In horses <1 year of age this opening appears small or slitlike or is not readily apparent when evaluated arthroscopically. The centrodistal joint lubricates the articulation between the central and third tarsal bones and the bones on each side, and the tarsometatarsal joint lubricates the third and fourth tarsal bones with the second metatarsal bone (MtII), the MtIII, and the fourth metatarsal bone (MtIV). The superficial location of the joint capsules makes them susceptible to penetration and the introduction of infection in association with trauma to the tarsus.

There is communication in 8% to 39% of the centrodistal and tarsometatarsal joints, but not necessarily in both hocks of the same horse,^9-11 although substances may spread between them by diffusion even in the absence of frank communication. ^12,13 Communication between the distal tarsal joints and the tarsocrural joint was demonstrated in 3% of horses after injection of radiodense contrast medium into the tarsometatarsal joints in live horses¹⁰; however, in a cadaver study mepivacaine injected into the centrodistal or tarsometatarsal joints spread to the tarsocrural joint within 15 minutes of injection in 88% and 92% of limbs, respectively.¹²

Anatomical aspects of the gastrocnemius and calcaneal bursae and the tarsal sheath are discussed elsewhere (Chapters 76, 79, and 80). The distal tendon of the long digital extensor muscle is enclosed within a synovial sheath as it passes over the dorsolateral aspect of the tarsus. This sheath is compressed by transverse retinacular bands, resulting in a loculated appearance if the sheath is distended. Such longitudinal swelling is a quite common incidental finding in both sound and lame horses either unilaterally or bilaterally and is only rarely associated with pain and lameness. Focal lesions of the long digital extensor tendon have been identified ultrasonographically, but usually unassociated with lameness.

Numerous ligaments surround the hock. Both the lateral and medial CLs have one long and three short components.¹⁴ The long lateral CL is superficial; originates at the caudal aspect of the lateral malleolus of the distal tibia; inserts on the distal lateral aspect of the calcaneus with additional fibers to the fourth tarsal bone, the MtIII, and the MtIV; and forms a canal for the tendon of the lateral digital extensor muscle. The short lateral CLs lie deep to the long lateral CL and originate from the cranial aspect of the lateral malleolus, pass plantad, and attach to the lateral surface of the calcaneus and the proximolateral plantar aspect of the talus. The long medial CL originates from the caudal aspect of the medial malleolus, attaches distally to the MtII and the MtIII, and also attaches to the medial aspect of the distal tarsal bones. The short medial CLs lie deep to the long medial CL and extend from the medial malleolus to the medial aspect of the calcaneus and sustentaculum tali. The long plantar ligament is a strong, flat band that originates at the proximal plantar surface of the calcaneus, extends distally, and attaches to the fourth tarsal bone and the MtIV. The dorsal tarsal ligament spreads out distally from the distal tuberosity of the talus and attaches to the central and third tarsal bones and proximal aspect of the MtIII and the MtIV. Numerous short intertarsal ligaments connect adjacent bones of the tarsus and have connections between tarsal and metatarsal bones.

The suspensory ligament (SL) also has an accessory ligament that extends proximally to originate from the plantar aspect of the fourth tarsal bone and the calcaneus.¹⁵ This anatomical relationship between the SL and the tarsus may explain why distal hock joint pain and suspensory injury may coexist, why some horses with primary SL pain show a positive response to tarsal flexion, and why there may be confusion with diagnostic analgesic techniques used to differentiate distal hock joint pain from proximal suspensory desmitis.

Diagnosis

Clinical Signs

Swelling is a variable feature of hock-related lameness. There are numerous common swellings of the tarsus. but the presence of swelling is not pathognomonic for tarsal region pain (see Chapter 6, page 58, and Figure 6-28 Figure 6-29 Figure 6-30 Figure 6-31 Figure 6-32). Swellings include capped hock, which must be differentiated from lateral dislocation or luxation of the superficial digital flexor tendon and swelling of the calcaneal bursae; tarsal sheath–associated or, less commonly, non–sheath associated thoroughpin; bog spavin (effusion of the tarsocrural joint); bone spavin (bony prominence of the dorsal and medial aspects of the distal tarsal region); curb (see Figure 78-1), a collection of soft tissue injuries involving the distal plantar aspect of the hock; and diffuse soft tissue swelling in the distal aspect of the crus and dorsal aspect of the tarsus associated with injury of the fibularis (peroneus) tertius. Rapid development of diffuse periarticular swelling may follow trauma of the tarsus or may herald periarticular cellulitis. Distention of the tarsocrural joint can be an incidental finding but may reflect primary joint pathology. Distention of the distal hock joint capsules is rarely detectable clinically, but if it occurs in association with OA, there may be firm enlargement on the medial aspect of the limb reflecting periarticular new bone and overlying fibrous tissue. The presence of this swelling, bone spavin, is neither required for horses to be lame as a result of distal hock joint pain nor pathognomonic for distal hock joint pain in horses in which it is found. Swelling restricted to the medial or lateral aspect of the tarsus may reflect CL injury and malleolar fractures of the distal aspect of the tibia.

Some horses with tarsal region pain resent passive flexion, but exaggerated lifting of the limb to avoid maximal flexion is more likely to reflect stifle pain. Although there are many advocates of the Churchill test for the detection of distal hock joint pain (see Chapter 6), we do not find this test particularly useful because there are many false-positive and some false-negative responses. This may reflect the degree of pressure applied, because one author (SJD) is rarely able to induce pain, whereas the other (MWR) frequently does elicit a pain response. Some horses with distal hock joint pain do manifest a positive reaction to palpation of the medial soft tissue structures and the proximal medial metatarsal region, likely reflecting the presence of periarticular soft tissue pain associated with OA. Horses often manifest a positive response to compression of the distal medial aspect of the tarsus and proximal medial aspect of the metatarsal region statically, and lameness can often be exacerbated with compression of this region followed by trotting, a dynamic test; yet pain in these horses is localized with diagnostic analgesia to the distal portion of the hindlimb. Proximal suspensory desmitis and lameness associated with the metatarsophalangeal joint are common diagnoses in these horses. It is possible that horses with distal hindlimb pain move abnormally and have coexistent pain in the region compressed using the Churchill test (see Figure 6-33 Figure 6-34 Figure 6-35). The absence of a positive Churchill test result should not lead the examiner to conclude the horse does not have tarsal region pain, a false-negative response. Unless clinical signs are diagnostic or a horse is suspected to have a fracture, diagnostic analgesia should always be used to confirm the presence of tarsal region pain.

Shoe wear can be excessive in horses with chronic tarsal region pain. Shoes are often worn on the toe or, most commonly, on the dorsal and lateral aspects (see Figure 6-38). Often, the fullering and toe grab, if present, are worn completely through, a finding that is most prominent in horses shod in aluminum shoes. Horses with pain causing lameness from other sources in the distal aspect of the hindlimb can manifest similar shoe wear, so this finding is not pathognomonic of distal hock joint pain. Horses with chronic tarsal region pain often have coexistent signs of pain on palpation of the gluteal and thoracolumbar area. Concurrent back and gluteal pain is likely secondary to chronic abnormal limb carriage causing muscle pain.

Lameness can range from mild to severe. Horses often warm out of lameness and are often able to perform with low-level distal hock joint pain. In racehorses, particular the Standardbred (STB), trainers often comment that the “horse throws away the lameness at speed.” There are no pathognomonic clinical signs or gait deficits associated with tarsal region pain. Gait abnormalities associated with hock-related lameness are very variable both in degree and in character, depending on the underlying cause of lameness. A reduction in the cranial phase of the stride at the trot is a consistent finding but is a common finding in horses with any source of pain causing hindlimb lameness. When observed from behind, a trotting horse with tarsal region pain during protraction swings the affected hindlimb medially toward the midline and then stabs laterally while landing, often called a “stabby” hindlimb gait. This is particularly noticeable when horses have bilateral hindlimb lameness. Based on the clinical observation of a stabby hindlimb gait and positive response to upper limb flexion (see later) a diagnosis of distal hock joint pain is made, particularly in Western performance horses or gaited breeds, and empirical treatment is undertaken and is successful. However, this stabby hindlimb gait is not pathognomonic for tarsal region pain, and horses with lameness as a result of pain originating anywhere in the distal aspect of the hindlimbs can manifest this type of gait abnormality. Diagnostic analgesia should be used if horses do not respond to initial treatment. Some horses with severe tarsal region pain and swelling carry the limb wide, avoid flexion, and swing the limb laterally during protraction. These horses are often reluctant to flex the hock while standing during the palpation and manipulation portions of a lameness examination.

Kinematic gait measurements were recorded after endotoxin-induced lameness of the distal tarsal joints.¹⁶ Both fetlock and tarsal joint extension during stance phase decreased, fetlock joint flexion and hoof height during swing phase increased, limb protraction decreased, and vertical excursion of the tuber coxae became more asymmetrical. These observations are not entirely consistent with the observations made in natural disease.

More recently, three-dimensional kinematic gait analysis was performed before and after experimental induction of synovitis of the centrodistal and tarsometatarsal joints resulting in mild lameness.¹⁷ There were significant decreases in tarsal joint flexion and in dorsal translation (sliding) of the metatarsal region relative to the tibia during the stance phase.¹⁷ Measurement of ground reaction forces with subtle lameness indicated reduced weight bearing on both the lame hindlimb and the contralateral forelimb but no change in the opposite hindlimb. Thus the gait may appear to have less bounce, rather than overt lameness being detectable. With mild increase in lameness there was reduced tarsal flexion in the stance phase and decreased forward sliding of the distal joints. This results in reduced propulsion, influencing quality of the gait, and less absorption of concussion. Movements that require maximum tarsal flexion are likely to be most painful, and resistance to perform such movements may be apparent before lameness is observed.

Horses with tarsal region pain often respond positively to an upper limb flexion test (see Chapter 8, page 84 and Figure 8-8). This test has erroneously been called the spavin test and a positive response should not be interpreted as pathognomonic for tarsal region pain. When it is suspected that a horse has tarsal region pain, this test often gives false-positive results. The most marked response to an upper limb flexion test may indeed be observed in horses with tarsal region pain, but a positive response can be seen with pain originating from the stifle, the hip region, and even the distal aspects of the limb. The upper limb flexion test is not specific, nor is the “hock extension test” (see Figure 8-10), a test used to exacerbate lameness in horses with tarsal region pain. There is no substitute for accurate diagnostic analgesia.

Diagnostic Analgesia

Analgesia of the hock region can be accomplished by perineural analgesia of the fibular (peroneal) and tibial nerves or by intraarticular analgesia. Although the latter is theoretically more specific, there are a number of important limitations. False negative results can occur in the presence of subchondral bone pain and/or extensive cartilage pathology. Horses with incomplete fractures of the talus or central tarsal bone often show little response to intraarticular diagnostic analgesia but do show a marked improvement after perineural analgesia of the fibular and tibial nerves. Lameness may be improved rather than abolished, and although this may be easy to appreciate in a moderately lame horse, it is less easy in a horse with a subtle lameness. Intraarticular analgesia has limited ability to influence periarticular soft tissue structures that may be contributing to pain causing lameness. The capacity of the centrodistal and tarsometatarsal joints is relatively small, and injection of too large a volume of local anesthetic solution results in leakage; the close proximity of the plantar outpouchings of the tarsometatarsal joint to the SL means that intraarticular analgesia of the tarsometatarsal joint may remove pain from the SL. With advanced joint space loss or periarticular new bone formation, intraarticular analgesia may not be physically possible.

Perineural analgesia of the fibular and tibial nerves is well tolerated by most horses (see Chapter 10). One of us (SJD) deposits a subcutaneous bleb of local anesthetic solution at both injection sites using a 25-G needle before performing the blocks. Although a 3.7-cm needle is adequate for most horses, to reach the deep branch of the fibular nerve requires a 5-cm needle in large (700 kg) horses. These nerves are large, and it takes longer for analgesia to develop than with more distal limb blocks. Do not be in a hurry to move proximally in the limb and perform intraarticular analgesia of the stifle or hip joints, because these blocks may take a full hour to work. We usually first evaluate the horse 20 minutes after injection and for up to an hour. It is important to recognize that successful blocks may increase or create a toe drag; improved stride length, rhythm, and symmetry of the hindquarters must therefore be used to evaluate improvement in lameness. Occasionally a horse will stumble slightly after fibular and tibial nerve blocks, but we commonly see horses ridden before and after and do not believe there are undue risks.

When performing intraarticular analgesia of the distal tarsal joints, we routinely block the tarsometatarsal joint first, because this is relatively easy and safe to perform, and commonly find that horses with centrodistal joint pathology are improved. If there is partial or no response, then the centrodistal joint is blocked. In a study using 66 cadaver limbs, mepivacaine (5 mL) was injected into either the tarsometatarsal or centrodistal joints; synovial fluid samples were collected from the tarsocrural, centrodistal, and tarsometatarsal joints 15 minutes later.¹² Concentrations of mepivacaine that were potentially analgesic were found in the centrodistal and tarsocrural joints after injection of the tarsometatarsal joint in 64% and 4% of limbs, respectively After injection of the centrodistal joint, analgesic concentrations of mepivacaine were found in the tarsometatarsal and tarsocrural joints in 60% and 24% of limbs, respectively. Despite these results we have never clinically recognized improvement in tarsocrural joint pain after intraarticular analgesia of the centrodistal and/or tarsometatarsal joints. We use a maximum volume of 3 to 4 mL of mepivacaine and evaluate the response 5 to 10 minutes after injection. Larger volumes will leak out of the injection site, increasing the likelihood of inadvertent analgesia of the proximal SL and other metatarsal structures. False-negative results may occur even in the absence of detectable radiological abnormalities, and the response to intraarticular medication may be substantially better. However for the much larger tarsocrural joint we use at least 20 mL of mepivacaine and will wait up to 30 minutes before declaring that the response is negative. One of us (MWR) has found that volumes of local anesthetic solution of up to 40 to 50 mL may be required to abolish pain associated with severe OA of the tarsocrural joint, and often the examiner needs to wait 45 to 60 minutes for the full effect of the intraarticular block to work.

Diagnostic Imaging

Four radiographic images of the tarsus are required: lateromedial, dorsolateral-plantaromedial oblique, dorsomedial-plantarolateral oblique (DM-PILO), and dorsoplantar.¹⁸ Lesions may be detectable only in a single view; thus in our opinion all four views should be obtained routinely. To cut through the centrodistal and tarsometatarsal joint spaces, it is important that the horse be standing with the metatarsal region vertical and bearing weight evenly on each hindlimb. Because the centrodistal and tarsometatarsal joints slope distally from laterally to medially, to reliably cut through the joint spaces in a lateromedial image, the x-ray beam should be angled 10 degrees distally. If a horizontal x-ray beam is used, it can be difficult to cut through the entire centrodistal joint space in a dorsoplantar projection, and it may appear that one side of the joint is narrowed. An additional dorsal 5° proximal-plantarodistal oblique image helps to determine whether there is genuine joint space narrowing.

Some lesions may be missed in dorsoplantar images, such as an axial osteochondrosis lesion of the medial malleolus of the tibia or a parasagittal fracture of the talus. Additional views obtained by angling the x-ray beam slightly obliquely (dorsal 10- to 20° lateral-plantaromedial oblique image) may be required in selected horses. A flexed lateromedial image and a flexed skyline image can give important additional information in some horses, particularly those with lesions involving the proximal and plantar aspects of the talus, those with osteochondral fragments involving the plantar pouch of the tarsocrural joint, and those with radiolucent defects on the calcaneus. A flexed dorsoplantar image can be useful to evaluate the proximal aspect of the talus in horses with incomplete fractures of this bone but is difficult to obtain.

Nuclear scintigraphy is most useful when pain causing lameness has been localized to the tarsus but no radiological or ultrasonographic abnormalities are detected to explain the lameness.¹⁹ The identification of increased radiopharmaceutical uptake (IRU) may prompt acquisition of additional radiographic images that yield a diagnosis. Scintigraphy may also be useful in horses that are difficult to nerve block or those that are examined for poor performance rather than overt lameness. It is important to recognize that RU in the distal tarsal bones can be influenced by the type of work history. For example, there is greater RU in the dorsal aspect of the central and third tarsal bones in elite show jumpers compared with horses from other disciplines.⁵ In a plantar scintigraphic image of the tarsus it is common and considered normal to have greater RU in the subchondral bone of the lateral aspect of the tarsometatarsal joint and the proximal aspect of the MtIV compared with medially.²⁰ This finding should not be interpreted as supportive of a diagnosis of distal hock joint pain. In young racehorses with early OA of the centrodistal and tarsometatarsal joints, focal moderate-to-intense IRU is found most often in the dorsal and lateral aspects of these joints, corresponding to radiological abnormalities detectable on the dorsomedial-plantarolateral radiographic image.²¹ The most common location for fractures of the central and third tarsal bones is on the dorsolateral aspect, and fractures can most reliably be seen in the dorsomedial-plantarolateral oblique radiographic image (Figure 44-1). These findings differ somewhat from most anecdotal reports, suggesting that it is the medial aspect of these joints that show early radiological changes and may be a reflection of a difference between racehorses and sports horses. In older racehorses with advanced OA and sports horses, scintigraphic and radiological evidence of OA is commonly found on both the medial and lateral aspects of the distal hock joints.

Fig. 44-1 A, Lateral (left two images) and plantar scintigraphic images of a 2-year-old Standardbred pacing filly with right hindlimb lameness as a result of a central tarsal bone fracture. Focal, moderate increased radiopharmaceutical uptake can be seen involving the dorsolateral aspect of this bone (arrows), a common location to find fractures or early osteoarthritis in racehorses. B, Dorsomedial-plantarolateral (DM-PlLO) digital radiographic image showing an incomplete fracture of the central tarsal bone (arrows). Fractures of the central and third tarsal bones and early signs of osteoarthritis can be seen most commonly in a DM-PlLO radiographic view, indicating disease in young racehorses is found in predominantly the dorsal and lateral aspects of the distal hock joints.

Ultrasonography is invaluable for the assessment of both periarticular soft tissues and the tarsocrural joint. An excellent review of normal anatomy and examples of abnormality are published elsewhere.²² Magnetic resonance imaging (MRI) and computed tomography (CT) have the potential to yield valuable additional information when a diagnosis cannot be reached by other means. Normal MRI and CT anatomy references are available.^23-25

Articular Diseases of the Tarsus

Osteoarthritis of the Distal Hock Joints and Distal Hock Joint Pain

Distal hock joint pain is common in horses from all disciplines and is often associated with OA. Distal hock joint pain is known colloquially as bone or jack spavin or occult or blind spavin in the absence of radiological abnormalities. The term juvenile spavin has been used to describe early OA that had a prevalence of 20% in a group of horses younger than 2 years of age.²⁶ Although it is usually seen in mature horses used for sport or pleasure, distal hock joint pain can occur in young Thoroughbred (TB) and STB racehorses and Western performance horses (see Chapter 120). Distal hock joint pain may be a sequela to incomplete ossification of the central and third tarsal bones (see page 517). Certain conformational abnormalities (such as sickle-hock, in-at-the-hock, or cow-hock conformation) or excessive straightness of the hindlimbs may predispose to distal hock joint pain, although this condition frequently occurs in normally conformed horses. Traditionally it has been proposed that OA of the distal hock joints is caused by excessive compression and rotation of the distal tarsal joints as the horse jumps or stops, which results in abnormal tension on the intertarsal ligaments. However, this theory is not consistent with the common recognition of distal hock joint pain in pleasure horses or its high incidence in Icelandic horses, a breed in which OA is thought to be a heritable condition.²⁷ Distal hock joint pain is classically thought to begin on the dorsomedial aspect of the joints and to progress dorsally. However, it is our experience that in OA, scintigraphic or radiological abnormalities are frequently first identified only on the dorsolateral aspect of the joints, a region of high compressive strain. Previous exercise history and thus loading of the joints may be influential.²⁸ Nuclear scintigraphic studies of clinically normal mature horses in active work has shown mild greater uptake of radiopharmaceutical on the lateral aspect, which is consistent with increased modeling, presumably the result of a relative increased loading laterally compared with medially.⁷

The centrodistal and tarsometatarsal joints are most commonly affected, either individually or together, but OA of the proximal intertarsal joint does occur, usually in association with OA of the more distal joints. The condition may be unilateral but is often bilateral. Occasionally OA of the talocalcaneal joint occurs either in isolation^29,30 or together with OA of the centrodistal and tarsometatarsal joints³¹ (see page 516).