How to Perform a Complete Ultrasonographic Evaluation of the Pastern

1. Introduction

Although ultrasound has been used in musculoskeletal imaging for the past two decades, ultrasound of the equine pastern remains a challenge for many equine practitioners. Even veterinarians experienced in metacarpal/metatarsal ultrasound find the pastern difficult to understand because of its more complex anatomy. Injuries to the palmar/plantar soft-tissue structures of the pastern occur with relative frequency; however, many pastern injuries go undiagnosed, because the practitioner may be reluctant to ultrasound this region. The recent advent of magnetic resonance imaging (MRI) of the equine distal limb has increased our knowledge of distal extremity injuries [1-3]; however, the availability and affordability of MRI may limit an owner's and/or veterinarian's ability to use this modality. It is also important to review pastern ultrasonography to be able to correlate MRI findings. Standard and new techniques of pastern ultrasonography will reveal the presence of many soft-tissue injuries visible with MRI and should always be performed, especially when MRI is not available. This paper is intended to make the pastern more understandable by reviewing pastern anatomy and ultrasound technique. A systematic evaluation of the pastern will be emphasized. This paper will also review the distribution of pastern tendon and ligament injuries during a 5.5-yr study period (1999 - 2004) from the Large Animal Ultrasound Service at the University of California at Davis Veterinary Medical Teaching Hospital (UCD-VMTH).

2. Materials and Methods

A complete ultrasound evaluation of the pastern includes the deep digital flexor tendon (DDFT), straight distal sesamoidean ligament (SDSL), oblique distal sesamoidean ligament (ODSL), including its origins at the base of the proximal sesamoid bones (PSBs), the medial and lateral branches of the superficial digital flexor tendon (SDFT), and the medial and lateral axial and abaxial palmar/plantar ligaments of the proximal interphalangeal joint. Evaluation of the digital sheath, bony surfaces of the first phalanx (P1) and second phalanx (P2), pastern joint capsule, and medial and lateral collateral ligaments of the pastern joint can also be performed. Traditional pastern ultrasound involves evaluation of the palmar/plantar structures of P1. P1 is divided equally into three regions:P1A, P1B, and P1C (Fig. 1) [4]. Reference transverse and longitudinal ultrasound images are shown in Fig. 2a-c. Evaluation of the palmar/plantar structures of the P2 can only be performed with a microconvex transducer placed between the heel bulbs.

Figure 1. Sagittal anatomic pastern specimen showing the three zones (A, B, C) of P1 and the ultrasonographically visible region of P2.

Figure 2. (A) Proximal pastern region (P1A):normal transverse (left image) and longitudinal (right image) views obtained from palmar midline with a 10-MHz linear transducer at a scanning depth of 4 cm. Note the dorsal and central hypoechoic region of the normal DDFT that can be rectified by "rocking" the transducer proximal to distal. Lateral is to the right and medial is to the left on all transverse images. Proximal is to the right and distal is to the left on all longitudinal images. (B) Mid-pastern region (P1B):normal transverse and longitudinal images obtained from palmar midline. Note the insertion of the ODSL onto P1 (arrow) on the longitudinal view. This defines the distal extent of Zone P1B. Orientation is as described for A. (C) Distal pastern region (P1C):normal transverse and longitudinal images obtained from palmar midline. Note central hypoechoic area on transverse view of SDSL and corresponding central irregular fiber pattern at the SDSL insertion onto P2. Orientation as described for A.

Indications for pastern ultrasound include swelling of the pastern region, septic or non-septic digital sheath effusion, radiographic evidence of basilar fractures of the PSBs, and lameness localized to the pastern region. This includes horses whose lameness improves with abaxial sesamoid nerve blocks or palmar/plantar digital nerve blocks. Even the distal most placement of a palmar digital nerve block has the potential to anesthetize injuries of the distal DDFT, insertional regions of the SDSL, and branches of the SDFT. A positive response to digital sheath intrathecal anesthesia should also prompt pastern ultrasonographic evaluation to rule out injuries to the flexor tendons within the tendon sheath.

Ideally, the hair of the pastern should be clipped with #40 clipper blades. It is important to clip the hair proximally to the fetlock region to view the ODSL origins. It is often helpful to shave the palmar/plantar midline ridge of hair that grows in the opposite direction. This serves two purposes:it decreases standoff drag when sliding the transducer distally, which increases standoff life, and it eliminates the vertical edge artifact that is often seen extending through the image from this ridge of hair. The skin should be washed with soap and water, and ultrasound coupling gel should be applied. The patient should be placed in stocks when available and lightly sedated. The most comfortable position for the examiner is seated on the floor/ground. All pastern imaging at the UCD-VMTH is performed while seated on the floor; however, it is recognized that this is not ideal for clinician safety. An alternative is sitting on a short stool, but this places the back, arms, and shoulders higher than is ideal and requires bending and leaning in an uncomfortable position.

The best pastern images are obtained with a 10 - 14 MHz transducer; however, diagnostic images can be obtained with a 7 - 10 MHz transducer. A tendon-format transducer is ideally suited for pastern ultrasound, because both transverse and longitudinal images can be obtained. Rectal transducers may also be used, but longitudinal images cannot be obtained because of the positioning of the cord. A microconvex transducer is not necessary to obtain traditional views of the pastern, but it is required to fully evaluate the palmar/plantar structures of P2. Depth should be set at 3 - 5 cm for most horses. Generally, transverse views of each structure are obtained first and longitudinal views are obtained second. The small ligaments of the pastern (collateral ligaments and axial and abaxial ligaments) are best evaluated on longitudinal views.

DDFT

The DDFT (as well as the SDSL) is technically the easiest to evaluate in the pastern because of its palmar/plantar midline location throughout the length of the pastern. The transducer is placed immediately distal to the ergot, and the DDFT is evaluated as far distally as the heel bulbs will allow. The DDFT has an oval to half-circle shape in P1A. It is nearly impossible to obtain a homogeneously echogenic appearance of the DDFT in P1A on a single transverse view without "rocking" the transducer in a proximal to distal direction. A central and dorsal hypoechoic area is commonly seen in this location (Fig. 2a), but this can be rectified by changing the beam angle. The DDFT begins to appear bilobed in P1B and is a peanut shape in P1C. The DDFT is generally homogeneously echogenic with a linear fiber pattern. Injuries of the DDFT can occur at any location throughout the pastern, including P2. There is not a typical appearance for DDFT injuries. Tears may involve the lateral or medial lobes or both and can appear as small anechoic to hypoechoic core lesions (Fig. 3), irregularly shaped hypoechoic areas, or diffuse tearing (Fig. 4). Fiber tearing is usually easily visible on longitudinal views, even in chronic cases.

Figure 3. Small core lesion within medial lobe of DDFT in a 14-yr-old Trakhener dressage horse with persistent lameness localized to the pastern region. Note the thickening of the medial lobe (arrows) in Zone P1B compared with the normal lateral lobe (arrowheads). Fiber tearing has remained visible on longitudinal views after several months of a diligently followed controlled exercise program. Orientation is as described for Figure 2a.

Figure 4. Diffuse tearing of the DDFT in Zone P1A in a 14-yr-old Paint ranch horse with acute onset lameness and digital sheath distention. Scattered hypoechoic areas (arrowheads) and irregular fiber pattern (arrows) are easily seen on transverse and longitudinal views, respectively. Tearing continued distally throughout the pastern region. A large core lesion was also seen in the distal metatarsus. Orientation is as described for Figure 2a.

SDSL

The SDSL extends from its origin at the base of the PSBs to its insertion onto proximal P2. The SDSL is located deep to the DDFT throughout the length of the pastern. The SDSL shows a trapezoidal shape in P1A, a rectangular to square shape in P1B, and a square shape in P1C. The SDSL often shows a central hypoechoic area with a mildly to moderately irregular fiber pattern at P1C near its insertion onto P2 (Fig. 2c). Small hypoechoic areas may also be seen near its origin on the PSBs. Comparison to the contralateral limb will assist in differentiating normal from abnormal. SDSL injuries are relatively uncommon but do occur [5]. Small focal lesions may be seen in addition to hypoechoic core lesions (Fig. 5).

Figure 5. Large core lesion within the SDSL (arrows) of a 17-yr-old Quarter Horse pleasure riding horse with acute onset lameness. The lesion was also visible in Zones P1A and P1C. Rechecked ultrasound exams revealed slow improvement in echogenicity and fiber pattern over a 9 - 12-mo period. Orientation is as described for Figure 2a.

SDFT Branches

The SDFT is a thin structure between the skin and the DDFT in the proximal pastern region. The medial and lateral edges of the SDFT begin to enlarge to form its medial and lateral branches in P1A. The branches are completely formed in P1B and insert onto distal P1 and proximal P2. Evaluation of each branch is performed independently of the other. Evaluation of the lateral SDFT branch is performed by placing the transducer along palmar/plantar midline immediately distal to the ergot (as in evaluation of the DDFT and SDSL). The transducer is slid slightly laterally until the lateral edge of the SDFT is located in the center of the screen. The branch is then followed distally to its insertion. Evaluation of the medial SDFT branch is performed similarly, but the transducer is slid medially until the medial edge of the SDFT is located in the center of the screen. While most SDFT branch injuries are visible on transverse views, longitudinal imaging should be attempted either by rotating the transducer 90° from a transverse image or by sliding abaxially from a longitudinal image of the DDFT until the thin SDFT is visible deep to the skin. Beginning ultrasonographers often find longitudinal views difficult to obtain. The SDFT branches demonstrate a small tear-drop shape in P1A (Fig. 6), a larger tear-drop shape in P1B, and a triangular shape in P1C. The branches are uniformly echogenic in P1A and P1B; however, there is often a thin hypoechoic line that extends from their axial border in P1B that should be not misinterpreted as a lesion. The branches are often mottled at their insertions in P1C. Injuries to the SDFT branches can be seen along their length and may appear as hypoechoic to anechoic core lesions in the acute phase with fiber tearing visible on long axis (Fig. 7). Chronic lesions often appear mottled with a scattered irregular fiber pattern.

Figure 6. Reference normal image of lateral branch of SDFT (arrows) obtained from Zone P1A. Note the tear-drop shape on the transverse image and its close association with the lateral border of the DDFT. The medial branch of the SDFT shows a similar appearance. Orientation is as described for Figure 2a.

Figure 7. Recent acute tear of medial branch of SDFT (arrows) in a 22-yr-old Thoroughbred gelding sustained while stall confined for an unrelated illness. Note the central anechoic area of tearing in Zone P1B with corresponding fiber disruption centrally on the longitudinal view. Orientation is as described for Figure 2a.

ODSL

The ODSL is a "Y"-shaped structure with separate origins from the bases of the medial and lateral PSBs. The "branches" extend distally from the sesamoid bones and then join to form the body of the ligament at distal P1A. The ligament has a diffuse insertion onto P1 in the midpastern region at P1B. The majority of the ODSL is evaluated using the palmar/plantar midline approach described above for the DDFT and SDSL. From midline, the ODSL branches appear as small triangular structures deep to the SDSL and adjacent to the bony surface of P1 in the region of P1A. The ODSL shows a thin rectangular shape between the SDSL and the bony surface of P1 in P1B. The medial and lateral origins of the ODSL cannot be evaluated from palmar/plantar midline and must be evaluated separately using the following technique. Transverse views are obtained by placing the transducer at the base of the respective proximal sesamoid bone (Fig. 8). The transducer is held at an ~60 - 70° angle to the floor with the suspensory ligament branch bisecting the transducer. The neurovascular bundle can also be used as a landmark for transducer placement. Longitudinal images are obtained by rotating the transducer 90° from the optimal transverse image. The ODSL origins show a rounded triangular shape with an evenly echogenic appearance (Fig. 9). Care should be taken not to overanalyze linear edge artifacts extending from the edges of the overlying palmar/plantar digital artery and vein. Injuries most commonly involve the ODSL origins, although injuries may extend into the body and insertional regions. Anechoic to hypoechoic core lesions are most common (Fig. 10), although diffuse tearing can also be seen.

Figure 8. Proper positioning of ultrasound transducer at the base of the lateral PSB to obtain transverse images of the lateral origin of the ODSL. Images of the medial origin of the ODSL are obtained in similar fashion by placing the transducer at the base of the medial PSB.

Figure 9. Normal transverse and longitudinal images of medial or lateral origin of ODSL. The palmar digital vein and artery are seen in the near field on transverse views. The vertical edge artifacts (arrows) created from the vessel walls should not be misinterpreted as injury to the ODSL origins. Note the linear fiber pattern at the bone ligament interface on longitudinal views. Orientation is as described for Figure 2a.

Figure 10. Moderate hypoechoic core lesion (arrowhead) at the origin of the lateral ODSL (arrows) in a 13-yr-old Morgan pleasure horse with acute onset lameness. This lesion has remained visible despite 7 mo of controlled exercise. However, the lameness has resolved, and the horse is currently in moderate exercise. Orientation is as described for Figure 2a.

Axial and Abaxial Palmar/Plantar Ligaments

The medial and lateral axial and abaxial palmar/plantar ligaments are small ligaments that extend from P1 to P2 (Fig. 11). They are located palmar/plantar to the dorsally located collateral ligaments. The abaxial ligaments can be confused with the collateral ligaments. Although technically difficult for beginners, experienced pastern ultrasonographers should be able to image these ligaments successfully. Each ligament is best evaluated on longitudinal views. The axial ligaments are imaged by first locating the respective SDFT branch on longitudinal axis at the level of P1B. The transducer is then twisted slightly in an abaxial or dorsal direction until a small ligament is visualized deep to the SDFT branch. A linear fiber pattern should be seen. The abaxial ligaments are found by locating the respective SDFT branch on longitudinal axis at the level of P1B. The transducer is then slid slightly dorsally until a thin ligamentous structure is seen deep to the skin surface originating from P1. Injuries may appear as thickening of the ligaments with an irregular fiber pattern. Discrete areas of tearing can also be seen. Enthesophytes are a relatively common finding at the origin of the abaxial palmar/plantar ligaments on P1.

Figure 11. Palmar oblique view of pastern showing anatomical relationship of axial (A) and abaxial (B) palmar/plantar ligaments of the pastern joint to the branch of the SDFT (SDF Br). The axial ligament is located deep to the SDFT branch in the distal pastern region. The abaxial ligament is located dorsal to the axial ligament and palmar to the collateral ligament (C) of the pastern joint. The DDFT and distal sesamoidean ligaments are not shown in this diagram.

Collateral Ligaments

The medial and lateral collateral ligaments of the pastern joint are relatively thin structures located dorsal to the medial and lateral abaxial palmar/plantar ligaments. They are also best evaluated on longitudinal views. Injuries of the collateral ligaments are relatively uncommon, but traumatic lacerations of the distal limb may involve the collateral ligaments.

Evaluation of Palmar/Plantar Structures of P2

The proximal palmar/plantar eminences of P2 are located at approximately the level of the heel bulbs. Evaluation with a standard tendon- or rectal-format transducer is generally not possible, although removal of the standoff pad may allow evaluation of the proximal extent of P2 while pushing the transducer into the heel bulbs. A microconvex transducer is necessary to best evaluate this region. If a variable frequency transducer is used, it should be set to its highest frequency. The transducer is placed between the heel bulbs and angled slightly distally toward the direction of the toe or apex of the frog (Fig. 12). Ideal placement will depend on the conformation of the horse. Transverse and longitudinal images should be performed. While other abnormalities may be seen in this region, the primary purpose of these views is to evaluate the DDFT. Transverse views of the DDFT will be off-normal incident angle ("off beam"); however, the borders of the DDFT will remain visible, and the medial and lateral lobes should appear equal in size (Fig. 13). Injuries may appear as enlargement of one or both lobes, areas of decreased echogenicity, and/or hyperechoic areas that may represent calcification (Fig. 14). Longitudinal images may reveal thickening (dorsal to palmar/plantar) as well as defects in fiber pattern of the DDFT.

Figure 12. Proper positioning of the ultrasound transducer to obtain sagittal/longitudinal images of the palmar/plantar structures of P2.

Figure 13. Normal transverse image of the DDFT at proximal P2 obtained by using the technique shown in Figure 11. The DDFT is "off-beam" and appears hypoechoic; however, the borders of the tendon are easily visible (arrows). Note the symmetrical appearance of the medial and lateral lobes of the DDFT. Orientation is as described for Figure 2a.

Figure 14. Tear of lateral lobe of DDFT in a 9-yr-old Warmblood show horse with persistent right forelimb lameness. The lesion was most apparent at this level (P2). Note the enlargement of the lateral lobe with dorsal bulging compared with the normal medial lobe. Pinpoint hyperechoic foci are seen on transverse and longitudinal views within the area of tearing. Dorsal to palmar thickening of the DDFT is also seen on longitudinal views. Orientation is as described for Figure 2a.

3. Results

Records were reviewed of all horses that presented to the Large Animal Ultrasound Service at the UCD-VMTH from July 1999 to December 31, 2004. Eight hundred twenty-four pastern ultrasonographic examinations were performed during the 5.5-yr study period. This number includes 194 recheck examinations and 165 pastern exams performed as part of a digital sheath evaluation. Five-hundred thirty-three (533) forelimb and 291 hindlimb examinations were performed. Several horses underwent pastern evaluation of more than one limb. The majority of horses underwent pastern ultrasound because of obliteration or reduction in lameness with regional nerve blocks, swelling of the pastern region, and/or distention of the digital sheath. Some exams were performed at the request of the owner and/or the referral veterinarian without any localizing clinical signs.

No evidence of tendon or ligament injuries was found in 254 of 630 (40.3%) primary pastern examinations (corrected for recheck examinations). Many horses showed evidence of multiple injuries in the same limb or in multiple limbs. The distribution of injuries was ODSL (203), DDFT (122), superficial digital flexor branches (95), SDSL (61), abaxial palmar/plantar ligaments (40), and axial palmar/plantar ligaments (11). The relatively low number of axial and abaxial palmar/plantar ligament injuries is likely secondary to the fact that evaluation of those structures was not performed during the initial 1 - 2 yr of our study period.

The majority of the ODSL injuries were primarily or solely located at either the medial or lateral origin at the base of proximal sesamoid bone. Distribution of medial (105) and lateral (98) ODSL injuries was similar. Injuries were classified as mild (140), moderate (51), or severe (12). Many of the mild ODSL injuries were seen in middle-aged, well-campaigned performance horses. DDFT injuries were seen along the length of the pastern from P1A to mid P2. Lesions were classified as mild (63), moderate (34), or severe (25). Many horses with DDFT injuries also had digital sheath effusion and synovitis. DDFT injuries extended into the distal metacarpus/metatarsus in 27 horses. SDFT branch injuries were classified as mild (60), moderate (23), or severe (12). Distribution of medial (46) versus lateral (49) SDFT branch injuries was similar. SDSL injuries were characterized as mild (43), moderate (17), or severe (1). Severity of injuries was based on assessment of the cross-sectional area of the injured structure, size and echogenicity of the lesion, and amount and severity of fiber disruption or irregularity.

Recheck ultrasound examinations typically revealed a stable or only slightly improved appearance in the majority of the pastern tendon/ligament injuries. The lack of sonographic improvement may be reflected by the chronic history of lameness in many horses presenting to our hospital. However, many horses with DDFT injuries presented soon after onset of lameness. These acute lesions were not more likely to show sonographic healing than similar chronic lesions. The SDSL and SDFT branch injuries did seem to show improvement in echogenicity and fiber pattern but did so over a prolonged period of time. The mild ODSL injuries seemed to remain unchanged over time, although many of these horses were able to return to work.

4. Discussion

Ultrasonographic evaluation of the pastern region remains a viable, cost-effective modality to diagnose soft-tissue injuries of the pastern region. Positive findings were found in 59.7% of primary pastern examinations during our 5.5-yr study period. This is significantly greater than that reported in two previous studies of pastern soft-tissue injuries by Denoix et al. [6] and Reime [7]. The distribution of pastern soft-tissue injuries in our study also differs from these reports. While both authors report the SDFT branches to be most commonly affected, SDFT injuries were the third most common injury in our study. Injuries to the DDFT (n = 122) were fairly common in our study population, but only a few DDFT lesions were reported by Denoix et al. [6] and Reimer [7]. It should be noted that the highest percentage of severe injuries involved the DDFT (25 of 122; 20.5%) in our study population. Similar to previous reports, the SDSL was the least commonly injured structure.

The most likely contributing factor to injury distribution and prevalence is differing patient populations. The age, breed, and discipline distribution of horses presenting to UCD-VMTH Large Animal Ultrasound Service is quite varied. While many English and Western performance horses are seen at the VMTH, the patient population includes nearly all breeds and uses, ranging from the backyard or ranch horse to specialty breeds and broodmares. The fact that Thoroughbred racehorses do not constitute a large percentage of our practice may also explain the low rate of SDFT branch injuries compared with that reported by Reimer [7]. In another report on SDFT branch injuries, 11 of 12 affected horses were either flat or steeplechase racehorses, further supporting the high prevalence of SDFT branch injuries seen in racing disciplines [8].

The preponderance of ODSL injuries involving their origins has been documented in Thoroughbred racehorses [7,9]. This was also seen in our study group that included some racehorses but also numerous other disciplines. This strongly supports the use of the technique shown in this report to evaluate ODSL origins. The practitioner is reminded that evaluation of the ODSL origins cannot be performed from palmar/plantar midline. Complete evaluation of the pastern should include evaluation of the ODSL origins from the base of the PSBs as described above, especially in a predominantly Thoroughbred racehorse practice.

Although many significant tendon and ligament injuries were diagnosed during the study period, not all lesions identified with pastern ultrasound were considered clinically significant findings. In general, most moderate and all severe tendon and ligament injuries were considered clinically important findings. The majority of DDFT lesions were considered important and seemed to correlate well with the blocking history. On the other hand, some horses were diagnosed with primary bony or soft-tissue injuries in other limbs or regions using radiography and/or ultrasound. Mild injuries visualized on pastern ultrasound in these cases were often felt to be incidental or not the cause of clinical signs. Finally, many mild injuries involving the ODSL origins in the older show horses described above were felt to be chronic in nature. Many showed no enlargement of the ligament in association with the hypoechoic areas. Lameness was often localized to other areas in these cases or significant injuries were located elsewhere in the limb.

An excellent knowledge of anatomy is critical to the understanding of the pastern region. Current texts on equine distal-limb anatomy [10] and equine ultrasonography [11-13] are available to assist veterinarians with evaluation of the pastern region. Practitioners who remain intimidated by pastern ultrasonography may gradually introduce themselves to this methodology by focusing on the palmar/plantar midline structures of the pastern (DDFT and SDSL). Evaluation of these structures involves similar technique to the metacarpus and metatarsus that is familiar to many veterinarians. As the comfort level increases with these structures, the branches of the SDFT can be added to the exam as well as the origins of the ODSL. Evaluation of the small axial and abaxial palmar/plantar ligaments of the pastern joint should be attempted after evaluation of the other, more commonly injured structures are mastered. Finally, if and when a microconvex transducer becomes available, the palmar/plantar structures of P2 should be evaluated in each pastern ultrasonographic examination. A complete ultrasound evaluation of the pastern should become as important as radiographic evaluation in all horses with pastern swelling, lameness localized to the pastern region, and digital sheath effusion.