Chapter 38The Carpus

Anatomy

The carpus consists of the antebrachiocarpal, middle carpal, and carpometacarpal joints. The antebrachiocarpal and middle carpal joints are considered ginglymi, but they are not typical of hinge joints; the carpometacarpal joint is arthrodial.¹ Arthrodial joints also exist between carpal bones in each respective row. Effective movement of the carpus originates from the antebrachiocarpal and middle carpal joints. The carpometacarpal joint does not open, but it is subject to shear stress. The antebrachiocarpal joint lies between the distal aspect of the radius and the proximal row of carpal bones. The distal, dorsal aspect of the radius has deep grooves in which run the tendons of the extensor carpi radialis and common digital extensor muscles. In flexion the tendons compress the dorsal aspect of the antebrachiocarpal joint, limiting visibility when arthroscopic examination is performed. The proximal row of carpal bones includes the accessory carpal bone, which articulates with the distal aspect of the radius and the ulnar carpal bone. The accessory carpal bone forms the lateral border of the carpal canal. From lateral to medial, the ulnar carpal, the intermediate carpal, and the radial carpal bones complete the proximal row.

The middle carpal joint lies between the proximal and distal rows of carpal bones. The number of bones in the distal row varies but always includes, from medial to lateral, the second, third, and fourth carpal bones. A first carpal bone is present unilaterally or bilaterally in approximately 50% of horses¹ and should not be mistaken on radiographs for an osteochondral fragment. The first carpal bone articulates with the second metacarpal bone (McII) and the second carpal bone, and its presence is often associated with radiolucent areas in the McII. A fifth carpal bone is rare, but if present is small, articulates with the fourth carpal bone and the proximal aspect of the fourth metacarpal bone (McIV), and can be confused with an osteochondral fragment. The second, third, and fourth carpal bones articulate with the McII, the third metacarpal bone (McIII), and the McIV, respectively. The articulation of the second carpal bone and the McII is broader than is that of the fourth carpal bone and the McIV, and hence the McII receives greater load, an important fact to consider with fractures of the McII and the McIV. The third carpal bone, the largest bone in the distal row, has two fossae separated by a distinct ridge, the intermediate (lateral) and radial (medial). The radial fossa is largest, receives greater load, and is more commonly injured. The third carpal bone is L shaped and has a large, dense palmar portion that is rarely injured.

The medial and lateral palmar intercarpal ligaments resist displacement and dissipate axial forces by allowing abaxial translation of carpal bones.^6,7 The long and short medial and lateral collateral ligaments originate on the radius and attach to the proximal aspects of the McII and the McIV, and the abaxial surface of the carpal bones, respectively. The collateral ligaments provided the major resistance to dorsal displacement of the proximal row of carpal bones during experimental loading, but the small but important palmar intercarpal ligaments contributed 23% resistance.² The lateral palmar intercarpal ligament mostly attaches proximally on the ulnar carpal bone and distally on the third carpal bone and may be divided,³ findings different from those previously reported—that the distal attachment was mostly on the fourth carpal bone.⁸ The medial palmar intercarpal ligament has four bundles that vary in size, and it courses between the radial carpal bone proximally and the palmaromedial surface of the third carpal bone and palmarolateral surface of the second carpal bone distally.³ Tearing of the medial palmar intercarpal ligament and to a lesser extent the lateral palmar intercarpal ligament was observed in horses with carpal disease and was recently proposed to be associated with cartilage and subchondral bone damage (see the following discussion).^8,9

The carpus has a dense joint capsule dorsally that blends with the overlying fascia and retinaculum. Synovium in young horses is often thickened or folded dorsally in the middle carpal joint and can interfere with visibility during arthroscopic surgery. This fold appears to smooth as horses age or as osteoarthritis develops. The antebrachial fascia blends with the retinaculum that functions to restrain extensor tendons. Retinaculum thickens and forms the medial and palmar borders of the carpal canal. The palmar retinaculum is sometimes severed in horses with carpal tenosynovitis and tendonitis (see Chapter 75). Anatomical considerations and flexor and extensor tendon injuries are discussed elsewhere (see Chapters 69 and 77). The sheathed extensor carpi radialis and common digital extensor tendons, located dorsally and dorsolaterally, respectively, limit carpal palpation and restrict access. Cul-de-sacs of distended antebrachiocarpal and middle carpal joint capsules can be palpated medial to the extensor carpi radialis tendon or between the extensor carpi radialis and common digital extensor tendons in a standing horse. Arthrocentesis and arthroscopic examination require careful placement of needles and instruments in these portals to avoid injury to tendons and sheaths. These portals can be easily felt as distinct depressions when the carpus is flexed. The sheathed lateral digital extensor tendon, located on the lateral aspect, should be avoided during arthrocentesis of the palmarolateral pouches. The sheathed extensor carpi obliquus tendon is small and passes obliquely over the antebrachiocarpal joint from lateral to medial to attach to the McII. This tendon can readily be seen medially during arthroscopic examination of the antebrachiocarpal joint. Extensor tenosynovitis must be differentiated from middle carpal and antebrachiocarpal joint effusion and hygroma. The antebrachiocarpal and middle carpal joints each have a palmarolateral and a palmaromedial outpouching through which arthrocentesis and arthroscopic evaluation can be performed. Unless greatly distended, the palmarolateral outpouchings are larger than the corresponding palmaromedial outpouchings. The palmarolateral outpouching of the antebrachiocarpal joint is in close proximity to the carpal sheath, and inadvertent penetration of the carpal sheath can occur during arthrocentesis or arthroscopic examination even when the palmarolateral outpouching is distended.

Knowledge of the communications and boundaries of the carpal joints is important in understanding the extent of disease processes and the results of diagnostic analgesia (see Chapter 10). The antebrachiocarpal joint is considered solitary, although in a single specimen in a cadaver study the joint communicated with the middle carpal and carpometacarpal joints.¹⁰ In some horses a communication appears between the antebrachiocarpal joint and the carpal sheath. The middle carpal and carpometacarpal joints always communicate (see Figure 10-8 Figure 10-9 Figure 10-10). Communication between the middle carpal and carpometacarpal joints and the carpal sheath is rare. The carpometacarpal joint has distinct distopalmar outpouchings located axial to the McII and the McIV that have secondary pouches interdigitating within the proximal aspect of the suspensory ligament (SL). These outpouchings explain inadvertent analgesia of the carpometacarpal and middle carpal joint while performing high palmar analgesia and possibly why lameness abates during middle carpal analgesia in horses with avulsion fractures of the proximopalmar aspect of the McIII or proximal suspensory desmitis.¹¹

Conformation

Racehorses, especially Thoroughbreds (TBs), with offset-knee (bench-knee) and back-at-the-knee (calf-knee) conformation are predisposed to develop carpal lameness. Mild in-at-the-knee (carpus valgus) deformity is common and of little concern, but if the deformity is severe, it can predispose to carpal lameness similar to that in horses with out-at-the-knee (carpus varus) conformation (see Chapter 4).

Clinical Characteristics and Diagnosis of Carpal Lameness

Carpal lameness is a common finding in many sports horses, but it is most common in racehorses. Former racehorses used in other disciplines may have chronic osteoarthritis (OA) or recurrence of osteochondral fragmentation. Primary carpal lameness in nonracehorses occurs from trauma such as from falls, kick wounds, and hitting fences; hyperextension injury resulting in fractures of the accessory carpal bone; and occasionally primary OA. Old horses, in particular Arabian horses, appear prone to develop inexplicable chronic, often severe OA of the carpometacarpal joint. Palmar carpal injury occurs in horses that fall or during recovery from general anesthesia and can result in moderate to severe lameness and subsequent OA. Palmar carpal injury most commonly involves the antebrachiocarpal joint, and clinical signs may not develop until hours or a few days after recovery from general anesthesia. Injury may result from trauma when the affected carpus is hyperextended or flexed and usually occurs in horses during rough recoveries but can occur, inexplicably, in horses with seemingly uneventful recoveries.

Few historical facts are pathognomonic for carpal lameness unless severe swelling and lameness develop acutely or a trauma is observed. TB racehorses may lug (bear) in, lug out, or fail to change leads. A Standardbred (STB) racehorse may be on a line. Horses may be racing poorly, particularly those with bilaterally symmetrical lameness. In racehorses with right forelimb carpal lameness, signs may be worse on the turns. Although most STBs are on the ipsilateral line, rarely a horse with right forelimb carpal lameness will be on the left line, presumably because the horse is bearing away from medially located pain or has a shortened stride in the right forelimb. Nonracehorses may have poor performance, fail to change leads, and hit or refuse fences. They may be uncomfortable when studs are placed in or removed from the shoes. Ponies with antebrachiocarpal joint pain may start to stumble.

Degree of lameness varies with the type and severity of carpal injury. Horses with early or mild chronic OA have mild lameness, whereas those with acute osteochondral fragments, slab fractures, or other more serious injuries have more severe lameness. Horses with infectious arthritis and comminuted carpal or other severe fractures may not bear weight at the walk. Dynamic angular deformity, carpus valgus or varus, may be seen in horses attempting to bear weight with comminuted fractures and loss of joint integrity. Lameness may be intermittent in horses with early and incomplete osteochondral fragments and may be apparent only after training or racing. Racehorses with bilaterally symmetrical lameness may not show overt lameness but have a wide, short gait bilaterally. Advancing and placing the affected limb wide while walking or trotting (pacing) is typical of carpal lameness (see Chapter 7). Horses with severe OA and natural carpal ankylosis or surgical arthrodesis swing (abduct) the limb because the carpus cannot flex. Advancement and placement of the limb in a lateral (abducted) position is not pathognomonic for carpal lameness, and horses with proximal palmar metacarpal pain or those with pain originating laterally in the digit may manifest similar signs. However, carpal tenosynovitis does not result in this typical carpal gait. Horses with carpal lameness have a shortened cranial phase of the stride. Lameness can be worse with the limb on the inside or outside of the circle depending on whether the location of pain is medial or lateral, but, in general, lameness in most horses with carpal lameness is worse with the limb on the outside of the circle.

Increased temperature (heat) over the dorsal surface of the carpus is a reliable indicator of carpal disease, but false positive and negative findings occur. If horses have been clipped for painting or blistering, or effects of topical counterirritation are still present, the area can be warm and sensitive without carpal lameness. The dorsal surface of each carpus should be evaluated and compared, but differences are difficult to detect if lameness is bilateral. Effusion is usually a reliable indicator for carpal synovitis, but it is not pathognomonic for carpal lameness. Horses with subchondral bone injury without overlying cartilage damage or incomplete osteochondral fragments often have carpal lameness without effusion. Effusion is suppressed in horses that have recently had corticosteroid injections. Effusion occurs commonly in young horses with early carpitis and in horses with advanced cartilage damage, osteochondral fragments, and infectious arthritis. Swelling associated with distention of the antebrachiocarpal or middle carpal joints is orientated horizontally on the dorsal aspect of the carpus, whereas effusion in a tendon sheath results in a longitudinally orientated swelling, compressed at intervals by horizontally orientated retinacular bands. With the horse’s limb in a weight-bearing position, the clinician’s fingers are used to ballot fluid within a distended carpal joint capsule. Capsular swellings are located between extensor tendons and medial to the tendon of the extensor carpi radialis. Older horses with chronic carpal changes often have mild or moderate effusion but can perform satisfactorily. In some horses there may be focal dorsal herniation of the antebrachiocarpal or middle carpal joint capsules.

Horses with acute carpal lameness, especially those with synovitis, often show a marked response to static flexion of the carpus. Horses with carpal tenosynovitis or other palmar carpal lameness also respond. Horses with pain from the proximal aspect of the limb, such as myositis in the proximal aspect of the antebrachium or elbow region pain, respond to carpal flexion (false-positive response). Static and dynamic flexion can be negative in horses with carpal lameness, especially in those with subchondral bone pain. Degree of flexion is usually decreased in horses with chronic OA because of joint capsule fibrosis. Careful palpation of all bony and soft tissue structures of the carpal region should be performed with the limb in standing and flexed positions, and responses should be compared with the contralateral limb. Swellings are best felt with the horse’s limb in a standing position. Swellings of nearby sheaths should be differentiated from authentic carpal effusion by location and ballottement. Horses with chronic OA often have firm, fibrous thickening at joint capsule attachments. Dorsomedial bony swelling of the radial and third carpal bones is observed and palpated in horses with chronic severe OA. The proximal aspect of the McII and the McIV should be palpated for bony and soft tissue swelling associated with fracture or exostoses. The proximal palmar metacarpal region should be palpated with the horse in the standing and flexed positions to differentiate pain in this region from carpal pain. The superficial digital flexor tendon should be carefully palpated to detect enlargement and pain, because horses with proximal superficial digital flexor tendonitis often manifest a positive response to carpal flexion. However, swelling of the superficial digital flexor tendon is easily overlooked because of compression by the extensive palmar retinaculum.

In my experience, the carpal flexion test is the most specific of any flexion test used, and if the test result is positive, carpal region pain is highly probable. A positive test result does not always incriminate the carpal joints, and the surrounding soft tissue structures must be kept in mind. A negative test result does not rule out carpal pain. Horses with subchondral bone pain, usually young racehorses with sclerosis of the third carpal bone, often have a negative or equivocal response to flexion. Horses with palmar metacarpal or elbow pain can respond positively.

Diagnostic Analgesia

In many horses, clinical signs and characteristic gait may allow a tentative diagnosis of carpal lameness, but in most horses diagnostic analgesia should be performed. Analgesia of the middle carpal and antebrachiocarpal joints should be performed independently and sequentially. Dorsal intraarticular techniques are most common, but in horses with scurf from previous counterirritant application or dorsal wounds, the palmarolateral pouches are used. Careful selective perineural and intraarticular analgesic techniques should be performed to differentiate between proximal palmar metacarpal pain and authentic carpal pain (see Chapter 10). Intraarticular analgesic techniques are highly specific, but false-negative results may occur.¹² False-positive results may also result because of abolition of carpal sheath pain or proximal palmar metacarpal region pain. Subchondral bone pain may not always be eliminated by intrasynovial deposition of local anesthetic solution, because nerve fibers may be located in bone or travel to the site by another extrasynovial route. The median and ulnar nerve block, although lacking specificity, is useful in horses with suspected carpal region pain that is not abolished by intraarticular techniques.

Laboratory analysis of synovial fluid is reserved for horses in which acute inflammation or infectious arthritis is suspected, but color and viscosity should be evaluated, and abnormalities may help to convince an owner or trainer of a carpal problem. In a normal flexed carpus, fluid does not readily drain from a small-gauge needle, and compression of the joint capsule at a distant site is usually necessary. Horses with effusion have thin synovial fluid that drips spontaneously without compression of the nearby capsule. Horses with true serosanguineous fluid (as opposed to contamination by penetration of capsular or synovial vessels) likely have cartilage damage with exposed subchondral bone or an osteochondral fragment. Hemarthrosis can be caused by trauma or bleeding from a torn intercarpal ligament.

Imaging

Radiography

A minimum of six well-exposed and positioned radiographic images are necessary for comprehensive examination of the carpus, including the dorsopalmar (DPa), lateromedial (LM), dorsal 45° lateral-palmaromedial oblique (DL-PaMO), dorsal 45° medial-palmarolateral oblique (DM-PaLO), and flexed LM images and the dorsoproximal-dorsodistal (tangential, skyline) image of the distal row of carpal bones. The skyline image is most important for assessing subtle radiological changes of the third carpal bone, but well-positioned images are often difficult to obtain. Evaluation of the radial fossa requires flexion of the limb in the sagittal plane with the metacarpal region beneath the antebrachium (Figure 38-1). Lateral positioning of the distal part of the limb results in overlap of the radial fossa of the third carpal bone and the radial carpal bone. The skyline image underestimates the amount of increased radiopacity of the third carpal bone and magnifies normal anatomy and lesions approximately twofold.¹³ The skyline image is not a true proximal-to-distal view of the second, third, and fourth carpal bones, and therefore lesions located palmar to the dorsal edge of the radial carpal bone cannot be seen. The skyline image cannot be used to evaluate fracture lines located more than 8 to 10 mm from the dorsal edge of the third carpal bone or to differentiate large osteochondral fragments from frontal slab fractures of the third carpal bone. Additional images, such as the tangential image of the proximal row of carpal bones (used to identify osteochondral fragments and unusually located frontal or sagittal slab fractures), flexed oblique images (e.g., DL-PaMO view with the limb held in flexion for evaluation of the articular surfaces of the third and radial carpal bones), and weight-bearing, oblique images of different obliquity (e.g., off DPa views, used to identify sagittal slab fractures of the third carpal bone) are sometimes useful. Considerable confusion arises in description of oblique images, with the use of terms lateral and medial oblique instead of naming the images according to the direction of the x-ray beam. To most clinicians, the lateral oblique is equivalent to a DL-PaMO image, but to others it is just the opposite. Follow-up radiographic examination is recommended in 10 to 14 days if fracture is suspected but initial radiological findings are negative or equivocal.

Fig. 38-1 A, A well-positioned dorsoproximal-dorsodistal (skyline) radiographic image of the distal row of carpal bones requires that the third metacarpal bone be aligned directly under the radius. This makes the radial fossa visible. B, Resulting digital radiographic image shows radial (R) and intermediate (I) fossae of the third carpal bone (C3) (medial is to the right and palmar is uppermost). There is dense increased radiopacity of the radial fossa. Sometimes overlap of the normal articulation between the third and fourth (C4) carpal bones appears as a linear radiolucent defect (white arrow) in the lateral aspect of C3, a finding confused with sagittal fracture of that bone. Often when digital radiographic techniques are used, fracture lines not visible on conventional radiographs can be seen, such as in this 3-year-old Thoroughbred colt with a small osteochondral fragment of the radial carpal bone (black arrow). The fracture fragment is superimposed on C3 but can readily be seen. C, Intraoperative arthroscopic photograph showing the chip fracture (arrow) and surrounding cartilage damage on the radial carpal bone (RC) in this horse. D, Intraoperative arthroscopic photograph showing thin articular cartilage and partial-thickness cartilaginous defects on C3 (arrows) that accompany sclerotic subchondral bone.

Normal radiological anatomy of the carpus is difficult because carpal bones overlap considerably, bones shift during flexion, and normal radiolucent defects and aberrant carpal bones can be difficult to interpret. In the skyline image of the distal row, the normal articulation between the third and fourth carpal bones can be superimposed on the lateral aspect of the third carpal bone and confused with a sagittal fracture (see Figure 38-1, B). On a DM-PaLO image the normal articulation between the second and third carpal bones should not be confused with a sagittal slab fracture, but this image is essential to diagnose sagittal fracture of the third carpal bone correctly, which runs parallel to this articulation. Radiolucent defects or osseous cystlike lesions are often seen in the ulnar carpal bone and are considered incidental findings, but when they appear in other bones, they can cause lameness regardless of whether communication with a joint exists. In LM and oblique images, the first and fifth carpal bones can be confused with osteochondral fragments. Radiolucent defects in the McII and the McIV often occur in the presence of the first (see Figure 37-7) and fifth (Figure 38-2) carpal bones but are normal. In a flexed LM image the radial carpal bone moves distally relative to the intermediate carpal bone. This normal finding is quite useful in determining the exact positioning of osteochondral fragments or other lesions on the proximal or distal surfaces of the radial and intermediate carpal bones. The flexed LM image is also highly useful for evaluation of the distal dorsal articular surface and subchondral bone of the radial carpal bone. Xeroradiography has largely been discontinued, but computed radiography and digital radiography are available at most institutions and many private practices and yield images superior to those obtained by conventional radiography, but positioning and exposure must still be optimized. Subtle radiological changes can be readily seen in most digital radiographic images, as can fragments, radiolucent defects, and other changes not previously visible on conventional images (see Figure 38-1). Care must be taken not to confuse normal articulations for fractures.

Fig. 38-2 Dorsal 45° lateral-palmaromedial oblique radiographic image of a left carpus in a 2-year-old Standardbred colt. A fifth carpal bone is present (arrow). Note the lucent area in the proximal aspect of the fourth metacarpal bone. Fifth and first carpal bones should not be confused with osteochondral fragments.

Computed tomography (CT) is available at some institutions and private referral hospitals and can be of value to determine fracture configuration and the presence of comminution, such as that demonstrated in a recent report of an Arabian filly with carpal instability, in which comminution that was not seen when digital radiography was used was confirmed with CT.¹⁴ Magnetic resonance imaging (MRI) has potential to be useful in the diagnosis of carpal region soft tissue and bony injuries but may have limited value because it is currently difficult to position a horse within some magnets to comprehensively evaluate the entire carpal region. Through use of ex vivo MRI, minor cartilage lesions and sclerotic subchondral bone have readily been seen in intact cadaveric specimens.¹⁵ Magnetic resonance (MR) contrast arthrography has been compared with arthroscopy and gross necropsy examination to evaluate and define the lateral palmar outpouching of the middle carpal joint.¹⁶ All structures of the palmarolateral outpouching including portions of the lateral collateral and lateral palmar intercarpal ligaments were visible in MR images, and information obtained compared favorably with arthroscopic examination performed after synovectomy using a motorized intraarticular blade.¹⁶ Ultrasonographic examination of the carpus can be useful to determine the extent of soft tissue damage, to determine if wounds or fistulous tracts communicate with carpal joints, and to diagnose extensor and digital flexor tendon injury, carpal tenosynovitis, and desmitis. Using ultrasonographic examination, the body and division into medial and lateral branches of the medial palmar intercarpal ligament could be seen from a dorsal approach, and the technique may be useful to image horses suspected of having injury of the ligament.¹⁷

Scintigraphy

Scintigraphy is especially useful to diagnose early stress-related subchondral bone injury and differentiate carpal lesions from those of the proximal metacarpal region (see Figure 19-19). A common finding in young racehorses is carpal lameness localized by clinical signs and diagnostic analgesia with negative or equivocal radiological abnormalities. Focal areas of increased radiopharmaceutical uptake (IRU) are often found unilaterally or bilaterally, most commonly in the third carpal bone (see Figure 19-16). Scintigraphy can be used to verify or refute the importance of sclerosis in the third carpal bone. Scintigraphy is useful in diagnosing unusual fractures of the palmar aspect of the third carpal bone, corner fractures or table surface collapse of the third carpal bone or other carpal bones, and lesions that are not apparent or are located in obscure areas not depicted radiologically. Focal areas of IRU occur with many carpal injuries, and although sensitivity is high, the specificity of scintigraphic images is low and differentiation of specific types of injuries is difficult. Scintigraphy is most useful in localizing the site of injury, based on which additional radiographic images are obtained, or rest is recommended, followed by repeated radiographic examination.

In general, scintigraphic examination is used when lameness is localized but a specific diagnosis cannot be made. In racehorses referred for evaluation of poor performance and obscure high-speed lameness, comprehensive scintigraphic examination of all limbs often reveals focal areas of IRU in the carpus, even when evidence of overt lameness is lacking. Areas of IRU are in locations typical for horses to develop osteochondral fragments and signs of OA. Whether these areas of IRU represent sources of pain causing high-speed or subtle lameness is unknown, but often radiological evidence of bone modeling (increased radiopacity, marginal osteophytes) or incomplete osteochondral fragments exists. Treadmill exercise of previously untrained horses has resulted in a significant IRU,¹⁸ a finding indicating scintigraphy can be useful to monitor skeletal response to exercise in normal horses. Mild IRU is seen as an adaptive response to bone modeling associated with exercise. In contrast, earlier work in normal horses showed exercise significantly increased subchondral bone density and uptake of radiopharmaceutical in the McIII condyles, but neither parameter increased significantly in carpal bones.¹⁹ More recently, in horses undergoing exercise after experimentally induced OA of the middle carpal joint, scintigraphic changes (IRU) were significantly correlated with lameness and increased beyond the normal adaptive response seen in association with exercise.²⁰