Successful medical management and rehabilitation of exercise-induced dorsal scapular luxation in an ultramarathon endurance sled dog with magnetic resonance imaging diagnosis of grade II serratus ventralis strain

Abstract

Dorsal displacement of the scapula in dogs is rare and often traumatic. This report describes dorsal luxation of the scapula in a sled dog. This case is unique given the injury was sport-related. Magnetic resonance imaging helped direct therapy and monitor healing; and medical management with rehabilitation resulted in full recovery and return to sport. One year after injury, the dog completed both a 482 km and a 1600 km endurance race, placing among the leading teams in the 1600-km race.

Dorsal luxation, or avulsion, of the scapula in dogs and cats is rare and has been described as resulting from trauma (such as vehicular trauma) that disrupts the supporting thoracoscapular muscular attachments, especially the serratus ventralis muscle (SV). This allows for a relative dorsal displacement of the scapula compared to the thoracic wall during weight-bearing (1–3). Surgical management has been reported to have a better prognosis than conservative management for these cases (2–4). Recently, successful surgical management of dorsal scapular luxation was described in 3 dogs: 2 of these dogs had a history of trauma and the other had an unknown history (3). A sport-related muscle strain has yet to be described in dogs and captured by imaging, whereas human muscle strains are often associated with intense eccentric contractions during exercise (5).

This report describes the presentation, diagnosis [including repeat magnetic resonance imaging (MRI)], and nonsurgical management of an acute exercise-related grade II serratus ventralis (SV) strain resulting in scapular luxation in an ultramarathon sled dog. This case is unique in that the injury was sport-related during eccentric exercise; MRI corroborated the clinical diagnosis and elaborated on the extent and severity of muscle damage beyond that implied by the clinical examination. Conservative management with physical therapy was supported by imaging and resulted in excellent recovery. This patient fully returned to sport and became a member of a top placing team at the following 1600 km Yukon Quest International Sled Dog Race. Based on this report it could be considered that dogs with suspected scapular luxations might benefit from MRI and rehabilitation in lieu of surgery for grade II SV strains.

Case description

A 5-year-old spayed female Alaskan husky dog participating in a 482 km sled dog race arrived at the 4th checkpoint (~402 km into the race) being carried in the team’s sled with the following trail history. The dog had been running in the lead position for most of the race but had slowed during the last run and had therefore been transferred into swing position (directly behind the lead dog) 64 km from the 4th checkpoint. The dog had become acutely lame and non-weight-bearing in the left forelimb after passing another team while descending a hill. There had been no fall or trauma observed. The team’s speed had increased just prior to the lameness, from ~16 to 19 to 21 km/h during the pass based on GPS data (all teams carry GPS on their sleds). Once lame, the dog had been immediately removed from the line and transported safely inside the sled to the 4th checkpoint where this history was provided.

Race veterinarians examined the dog at the 4th checkpoint. The patient vocalized during palpation of the left shoulder and had a flocculent swelling caudal to the left scapula. The dog initially refused to stand and was moderately dehydrated; however, once standing, dorsal luxation of the left scapula was noted. No diagnostics beyond physical examination were available at the remote checkpoint. The veterinary team recommended removing the dog from the race. The patient was then administered a 1-L IV bolus of 0.9% sodium chloride and given 50 mg tramadol (Sun Pharma, Princeton, New Jersey, USA) orally. The head veterinarian recommended that the dog be transported over 160 km to the nearest veterinary clinic for further diagnostics and stabilization.

The following day, on presentation to the clinic, the dog was quiet, alert, and responsive with a body condition score of 2.5/5 and was 6% dehydrated. The orthopedic examination remained unchanged with the exception of a subcutaneous hematoma visualized in the caudal left axilla. Mediolateral and caudocranial radiographs of the left scapula and humerus showed marked soft tissue swelling of the antebrachium and shoulder but no bony abnormalities. The left thoracic limb was placed in a Velpeau sling with instructions to maintain the bandage during the recommended 4 to 6 wk of crate rest. The patient was continued on oral tramadol with the addition of twice daily oral carprofen (Zoetis, Parsippany, New Jersey, USA) for 2 wk.

Eight days after injury, MRI (Toshiba Excelart Vantage 1.5T; Toshiba American Medical Systems, Tustin, California, USA) of the shoulder girdle was performed under general anesthesia at the University of Alaska Fairbanks to further assess and classify the degree of soft tissue damage. T2-weighted fast spin echo (T2-FSE) with fat saturation sequences were acquired in 3 planes. Proton density (PD) weighted sequences were performed in sagittal and transverse planes. The MRI showed extensive signal changes, in the form of increased signal intensity in T2w sequences, without change in volume in the muscles of the left trunk (Figure 1). These T2w hyperintensities consistent with edema were accentuated along fascial planes. Continuity of muscle fibers was best demonstrated in sagittal plane. Muscular changes involved the entire volume of the left m. serratus ventralis (cervical and thoracic portions), m. serratus dorsalis cranialis, m. rhomboideus cervicis, m. trapezius (cervical portion), and m. scalenus dorsalis, and extended from the tendinous raphe of the neck dorsally, medially to the (intact) m. longissimus capitis and m. iliocostalis, laterally to the medial surface of the scapula, ventrally to the level of the costochondral junction, and caudally to the level of the 7th rib. The space between the ribs was similarly affected, suggesting involvement of the m. intercostalis. Adjacent to the dorsal third of the left 3rd to 5th ribs, at a level corresponding to the site of attachment of the m. serratus dorsalis cranialis, there was a small flattened area of fluid signal intensity compatible with seroma or hemorrhage. The MRI changes were consistent with severe myotendinous strain without visible rupture.

Figure 1.

Dorsal (left) and transverse (right) MR images at the level of the 3rd rib pair 8 d after injury. The patient’s left is on the right side of the image. T2w FSE with fat saturation shows fluid signal along the fascial planes lateral to the m. serratus ventralis cervicis and m. scalenus (thick white arrow). A small pocket of fluid is present near the attachment site of the left m. serratus dorsalis cranialis (thin white arrow), representing either hematoma or edema. Bilaterally, feathery increased signal, consistent with benign muscle strain, is detected in the triceps muscle (black arrowhead, left); this is another muscle group presumably put under tension in the descent.

Seven weeks after injury, a follow-up MRI was performed at the same facility to assess healing. This MRI showed regressing left serratus ventralis and scalenus muscular damage with muscle atrophy and reduction in range and volume of the T2w hyperintensities. There was moderate residual T2w hyperintensity along the fascial planes of the m. serratus ventralis, extending lateral to the dorsal half of the rib cage and cranially to the transverse processes of the caudal most cervical vertebrae (Figure 2). The fluid pocket at the level of the 3rd to 5th ribs had resorbed.

Figure 2.

Dorsal (left) and transverse (right) MR images at the level of the 4th rib pair 7 wk after injury. The patient’s left is on the right side of the image. T2w FSE with fat saturation demonstrates regression of the changes observed in Figure 1, with mild residual hyperintensity medial to the mm. rhomboideus thoracis and serratus ventralis (white arrow). Tricipital changes have resolved.

At the time of the second MRI, the patient was ambulating relatively well at a walk. The musher had discontinued use of the Velpeau sling 5 wk before this evaluation. The dog had remained exercise restricted with the exception of recommended physical therapy. The dog’s owners had initially consulted with a canine sports medicine specialist (corresponding author) remotely by telephone and subsequently rehabilitated the patient according to recommendations with relatively good compliance. Crate rest for 3 wk was imperative. During that time, passive range of motion of the shoulder, elbow, and carpus was conducted while minimizing scapular motion. After 3 wk, isometric and low impact weight-bearing exercises were conducted, including weight shifting while standing and short leash walks. As the patient progressed the intensity of therapeutic exercise followed suit (such as weight-bearing with the uninjured forelimb elevated, mild to moderate grade hill-walking, and Cavaletti-like obstacle walking at or below elbow height). At 8 wk post-injury, eccentric weight-bearing exercises (such as steeper downhill-walking, jogging, and crawling) and a slow introduction to off-leash activity were implemented. By June (4 mo after injury), the dog was practicing short jogs (2.4 km) on glacier snow with short rests between sessions. The intensity of the protocol was gradually increased over the spring and summer in preparation for reconditioning back to sport in the fall. Such conditioning for endurance competition typically involves up to 3000 to 5000 km of mushing (6), of which the dog in this report completed about 5000 km (musher communications). One year after injury, the dog completed both a 482 km and a 1600 km endurance race, placing among the leading teams in the 1600-km race.

Discussion

In humans, muscle-tendon strains/ruptures are common and tend to occur from a combination of extrinsic and/or intrinsic factors (7–10). Extrinsic factors may include eccentric overload or chronic repetitive microtrauma. Intrinsic factors may include inhibited repair mechanisms, vascular supply, neuropathy, imbalance in extracellular matrix turnover, and/or altered biomechanics. Muscle becomes damaged at the same length of stretch; therefore, fatigued muscle is more likely to lengthen against external forces (7). Muscle contraction during lengthening is termed eccentric contraction and, although physiologically normal, is more likely to lead to injury than concentric or isometric contraction (7). A similar pathophysiology is suspected, but not described, in companion animals.

The sled dog in this report first experienced lameness before the 4th checkpoint while picking up speed and descending a hill. The described race situation accounts for extrinsic factors including fatigue (having raced over 340 km), overexertion (increasing speed with the center of gravity shifting further over the thoracic limbs), and intensified eccentric muscle contraction of the SV (downhill breaking during weight-bearing). Although no known intrinsic factors existed in the patient, they cannot be completely ruled out.

In dogs the forelimbs have a strut-like action, primarily engaging in weight-bearing and breaking functions during the gait cycle as ground reaction force vectors pass near the fulcrum (11–14). The bulky and wide-spanning SV muscle is the major weight-bearing muscle at the attachment of the scapula to the thorax (11,15). Increases in velocity or movement down an incline (such as occurred before acute lameness in our patient) result in further electromyographic activity of the SV as a result of eccentric contraction against external forces (11,12). The SV became injured when it could no longer withstand the forces placed upon it. Because the scapulothoracic articulation is a false musculotendinous joint (synsarcosis) in which the SV, trapezius, rhomboideus, and omotransversarius directly stabilize and facilitate movement (11,15), disruption or trauma to the SV may logically affect any or all of these other muscles. The MRI findings in this case described pathology of the trapezius and rhomboideus muscles in addition to the SV findings. Other muscles associated with the left thorax were also negatively affected but this may have been a result of indirect strain from compensatory gait or from disrupted fascial connections associated with SV injury.

The MRI grades of myotendinous strain injuries are reported in humans to parallel clinical grades (5). Bunching and retraction of the muscle with interposed hematoma formation is typical of 3rd degree strains (complete rupture) and was not observed in this dog. In 2nd degree strains (partial ruptures), MRI will demonstrate variable fluid intensities along fascial planes and possibly irregular thinning and mild laxity of tendon fibers. Hematoma formation at the myotendinous junction is diagnostic of a 2nd grade strain (16). The fluid pocket at the level of the left 3rd to 5th ribs may have represented such a myotendinous hematoma, probably representing compartmentalization of fluid medial and dorsal to the SV muscle attachment or, less likely, at the insertion site of the serratus dorsalis cranialis muscle.

There were some technical limitations to this MRI study that prevented a more refined assessment of these criteria, including those affecting spatial resolution (large voxel size), sequence selection preventing differentiation of seroma versus blood, and plane orientation restricted to standard coordinates, as opposed to conforming to the direction of the affected musculature. The serratus muscle is composed of thin, flat slips in which fibers run obliquely, forming a multipennate muscle architecture. Because of this specific anatomy, we suspect that a better evaluation of fiber integrity would have been achieved in a sagittal oblique plane oriented with the left rib cage. Nonetheless, differentiation between partial and complete tears cannot always be achieved with either MRI or ultrasonography (16). Persistent signal changes in a strained muscle may indicate an ongoing vulnerable period and higher risk of re-injury despite restoration of mobility (5). The dog’s clinical presentation mimicked a complete or near complete SV tear leading to dorsal displacement of the scapula in regard to the thorax during weight-bearing. In this case, such a displacement of the scapula may be more a reflection of myofiber dysfunction secondary to trauma compared to gross anatomical disruption of the muscle bellies.

Overall, little is reported on muscle strain injuries and treatment in dogs (17–21). The few studies which have been reported show a relatively high incidence of soft tissue injuries amongst canine athletes (22–24), in which many responded to conservative management (19). Due to the remote location of the dog kennel from the consulting rehabilitation specialist, the musher had to perform all recommended exercises without direct case follow-up. Although communication with the musher indicated relatively good rehabilitation compliance, it remains unknown to what extent the sled dog progressed through all the recommended rehabilitation. The therapeutic exercises for the dog were chosen to address clinical impairments and dysfunction through the healing process. Isometric activities to promote weight-bearing progressed toward slow incorporation of low impact eccentric activities to strengthen and lengthen affected muscle as the dog improved. Such physical therapy for muscle strains has been well-described in humans with hamstring injury (25,26). Unfortunately, successful or optimal rehabilitation protocols for specific canine diseases have yet to be validated or correlated to MRI findings. For this reason, evidence of canine rehabilitation success, especially with corresponding MRI findings, should be documented as a foundation upon which to develop prospective studies. Regardless, the balance between rest and therapeutic exercise during recovery showed tissue healing on repeat MRI and resulted in full return to ultra-marathon racing without surgery or re-injury in this dog.

This case remains unique in that the pathophysiology of the injury mimics those reported in acute exercise-related muscle-tendon strains in humans. Furthermore, diagnostic and follow-up MRI were obtained. Magnetic resonance imaging has been described as an important diagnostic and prognostic modality for myotendinous injury in human athletes (25–29). In contrast to a recently published case series by Jones et al (3), the gross examination in our patient was insufficient to determine the full degree and extent of injury to the supporting musculature of the thoracoscapular articulation. The MRI elucidated the site and degree of soft tissue injury beyond that of clinical diagnosis, demonstrating multiple muscle involvement within the synsarcosis and defining a grade II versus a complete SV rupture as described by Jones et al (3). The MRI findings helped direct a catered medical management approach including rest and rehabilitation to facilitate a full recovery in contrast to surgical exploration for diagnosis and/or treatment. Currently, MRI use in dogs for musculoskeletal disease is often precluded by availability, requirement of general anesthesia, and cost; however, further research may help broaden and better define its added value as a standard in diagnostics, prognostics, and helping to direct surgical versus non-surgical rehabilitative management.