Atlantoaxial Subluxation in the Dog

Atlantoaxial subluxation is typically a congenital or developmental disease affecting immature toy breed dogs [1,2]. Atlantoaxial subluxation has also been reported in other mammals including cats, horses, and cattle. In people, both the ligamentous laxity associated with Down's syndrome and the bony destruction of erosive rheumatoid arthritis can cause atlantoaxial subluxation [2-5]. In dogs with atlantoaxial subluxation, instability of the atlantoaxial joint results from a loss of ligamentous support of the axis, often with concurrent aplasia, hypoplasia or dysplasia of the dens. Acute or chronic spinal cord compression causing clinical signs consistent with an upper cervical myelopathy can result from this instability [2].

Anatomy

The occipitus, atlas, and axis are the three bones that comprise the craniovertebral junction. Through their specialized shapes and interrelationships of movement, they allow a wide range of motion of the head [6]. The atlanto-occipital joint is the cranial part of the overall occipito-atlanto-axial complex. It allows for dorsoventral motion of the head on the neck. The more caudal atlantoaxial joint allows for rotational movement of the head about the longitudinal axis of the vertebral column. At the center of this rotation is the dens, a bony protuberance that extends cranially from the axis and lies along the floor of the vertebral foramen of the atlas. The dens is the major point of attachment for ligaments that stabilize the atlantoaxial joint. The synovial joint cavities of the occipito-atlanto-axial complex are interconnected to form a single composite joint cavity [7].

Interosseous ligaments provide stability to the occipito-atlas-axis complex (Fig. 45-1) [8]. The apical ligament attaches the cranial aspect of the dens to the ventral aspect of the foramen magnum. The paired alar ligaments attach the dens to each of the occipital condyles. The transverse ligament is a tough fibrous band that acts as a restraint, keeping the dens against the body of the atlas. The transverse ligament allows rotational movement, but prevents the dens from tipping up into the vertebral canal. The dorsal atlantoaxial ligament and membrane extending between the vertebral arches of the atlas and axis also contribute to atlantoaxial stability [8].

Figure 45.1. The ligaments of the occiput, atlas, and axis in a normal dog. (Reprinted with permission from: Evans HE, Christensen GC. Millers Anatomy of the Dog, 3rd ed. Philadelphia: WB Saunders, 1993, p. 227.  - Available from amazon.com -  

Pathophysiology

Dogs with atlantoaxial subluxation usually have congenital abnormalities that predisposed them to injury. Although various congenital malformations of the entire occipito-atlanto-axial complex have been described in association with atlantoaxial subluxation, most commonly only aplasia or hypoplasia of the dens is present in dogs [2,9]. Dorsal deviation and so-called "nonunion" or separation of the dens have also been reported in the dog [2,10-12].

The cause of aplasia, hypoplasia, or dysplasia of the dens is unknown. At birth, the axis is composed of four separate ossification centers: the right and left neural arches, centrum 1 and centrum 2. Three additional ossification centers develop in the axis in postnatal life: intercentrum 2, the epiphysis, and the centrum of the proatlas (Fig. 45-2). An early hypothesis was that an assumed single ossification center for the dens fails to develop prenatally [11]. Subsequent investigation of the postnatal ossification centers of the atlas and axis in Miniature Schnauzers and other breeds revealed that the dens originates from two ossification centers [6,13]. The cranial one quarter originates from the centrum of the proatlas, and the caudal three quarters of the dens as well as the cranial part of the axis body originate from centrum 1 of the axis. The authors of this study thus contended that dens hypoplasia or aplasia is unlikely owing to congenital absence of an ossification center. They proposed that a vascular-related ischemia may lead to postnatal resorption of at least the middle part of the dens, resulting in dens dysplasia [13].

Figure 45.2. Ventral aspect of the axis from a 10-week-old Miniature Schnauzer, revealing 7 ossification centers: Cpa = centrum of the proatlas; C1, C2 = centrum 1 and centrum 2, respectively; ic2 = intercentrum 2; na2 = right neural arch center of the axis; e = epiphysis; x 5. (Photo courtesy of Dr. Alastair Watson. Reprinted from: Watson AG, Stewart JS. Postnatal ossification centers of the atlas and axis in Miniature Schnauzers. Am J Vet Res 51:264, 1990.  - PubMed -  

As the dens is a major attachment point for ligaments that stabilize the atlantoaxial joint, dogs with absent or hypoplastic dens are susceptible to trauma causing subluxation of the joint. Tearing of the dorsal atlantoaxial ligament and membrane often occurs with subluxation of the joint. With flexion of the head on the neck, ventral compression of the spinal cord occurs. Trauma to the spinal cord can come in the form of an apparently single acute episode or from more chronic progressive damage.

Atlantoaxial subluxation can occur in dogs with normal dens conformation as well. These cases usually result from traumatic rupture of the transverse ligament or fracture of the dens. A case of a dog with normal dens conformation with instability secondary to absence of the transverse ligament has been reported [14]. Such soft tissue abnormalities may be present in other cases of atlantoaxial subluxation in dogs with normal dens conformation. Dogs with atlantoaxial subluxation and a normal dens conformation may be at greater risk of severe spinal cord compression as the intact dens tilts dorsally into the vertebral canal. In one study, however, the conformation of the dens, whether it was normal, hypoplastic, or aplastic, did not affect the odds of a successful surgical outcome [9].

Clinical Findings

Atlantoaxial subluxation is a disease that most often affects immature, toy breed dogs of both sexes. Breeds most commonly affected include the Yorkshire terrier, toy and miniature poodle, Pomeranian, Pekingese, Chihuahua, and Maltese. However, atlantoaxial subluxation may occur in any size and breed of dog and at any age if significant trauma to the neck occurs.

Toy breed dogs with atlantoaxial subluxation often do present with a history of trauma, although this is often a mild incident such as jumping off a chair or being dropped by a child. Clinical signs of atlantoaxial subluxation may be mild to severe depending on the degree of compression and the chronicity of the disease. Most, but not all, dogs with atlantoaxial subluxation have neck pain, exhibited by rigid guarding of the neck, reluctance to move, or pain on cervical palpation. Mild cases of atlantoaxial subluxation may have ataxia or a spastic gait. Front and hind limb reflexes may be normal or increased, and postural reaction deficits may be present. More severe cases exhibit tetraparesis or tetraplegia. True paralysis from atlantoaxial subluxation is also associated with respiratory arrest and death owing to injury to the respiratory centers in the cervical spinal cord.

Occasionally, dogs with atlantoaxial subluxation have neurologic signs that may be localized to the caudal brainstem [9,15-18]. These may be seen more often in more neurologically compromised patients or following surgical manipulation. The cause of these signs often goes undiagnosed. They may be a result of concurrent congenital brain abnormalities, iatrogenic cervical spinal cord and brainstem trauma, or hypothetically, ischemia in the brainstem owing to compression of the basilar artery [18].

Diagnosis

The diagnosis of atlantoaxial subluxation is usually confirmed with radiography. Lateral radiographic views will show dorsal displacement of the axis in relation to the atlas, with an abnormally increased distance between the dorsal arch of the atlas and the dorsal spine of the axis, and dorsal displacement of the body of the axis (Fig. 45-3). On occasion, there may not appear to be enough distance between the dorsal aspects of the atlas and axis to be able to confirm atlantoaxial subluxation. General anesthesia may aid in getting diagnostic radiographs. Stressed views may be made by slightly flexing the neck; however, this maneuver can be very dangerous especially in the anesthetized unguarding animal. Fluoroscopy can aid in visualization of instability during flexion and extension. Myelography may also demonstrate spinal cord compression. Ventrodorsal radiographs of the atlantoaxial joint that demonstrate an absent or hypoplastic dens support a diagnosis of atlantoaxial subluxation (Fig. 45-4). An open-mouth radiographic view demonstrates the conformation of the dens; however, because flexion of the neck is required for this view, it is usually not recommended. Computed tomography and magnetic resonance imaging modalities are also useful in diagnosing atlantoaxial subluxation.

Figure 45.3. Lateral radiograph of caudal head and neck of an immature dog with atlantoaxial subluxation. Note the increased distance between arch of atlas and spine of axis.

Figure 45.4. Ventrodorsal radiograph of caudal head and the neck of an immature dog with atlantoaxial subluxation. Note the absence of the dens on the axis.

Differential diagnosis for cranial cervical spinal cord lesions in toy breed dogs includes focal myelitis associated with canine distemper, toxoplasmosis, or granulomatous myelitis and meningitis (most often immune-mediated in the young dog). Cervical disc disease is common in the middle-aged to older small breed dog. Neoplasia is also a rare possibility.

Treatment

Treatment of atlantoaxial subluxation consists of relieving spinal cord compression by reduction and permanent stabilization of the atlantoaxial joint. This has been achieved through both surgical and conservative methods.

Surgical Management

Surgical stabilization of the atlantoaxial joint is technically demanding. The vertebrae in toy breeds are extremely small and little margin for error exists when placing implants. The bone in immature dogs is soft. Implants can break the bone, migrate out of position, or cause impingement on the spinal cord. Manipulation of the atlantoaxial joint during surgery can also cause significant spinal cord trauma with subsequent death. However, most authors have recommended surgical stabilization for atlantoaxial subluxation. In dogs, both dorsal and ventral approaches to the atlantoaxial joint have been successfully used, with the goal of permanent fixation.

Dorsal Techniques

The first surgical stabilizations of the atlantoaxial joint in dogs were performed dorsally [1,15,16,19]. They consisted of orthopedic wire or suture material (even the nuchal ligament [19]) placed through the epidural space, then over the arch of the atlas, and finally, attached to the dorsal spine of the axis. A dorsal technique using a metal retractor [20] and a dorsal technique using a combination of pins and polymethylmethacrylate [21] have subsequently been described. Surgical implants applied on the dorsal surfaces of the atlas and axis have a theoretical biomechanical advantage as they are placed on the tension side of the joint subluxation; therefore, less implant strength is required to hold the joint reduced. Another advantage of the dorsal approach is the ease of surgical approach and the good exposure of the dorsal aspects of the atlas and axis. Complications of the dorsal techniques include resubluxation owing to suture or wire breakage or to these materials tearing through the soft immature bone. Also, life-threatening spinal cord damage may occur while attempting to pass material through the dorsal epidural space to the atlas. Dorsal surgical techniques do not allow access to the ventral articular surfaces of the atlantoaxial joint to attempt an arthrodesis. Long-term stability after dorsal procedures presumably is attributable to fibrosis across the dorsal aspects of the atlas and axis.

Ventral Techniques

Ventral stabilization techniques are advocated by most authors. The first ventral stabilizations involved placement of two transarticular Kirschner wires from the ventral aspect of the axis bilaterally through the joint to just medial to the alar notches on the wings of the axis [10,16]. This technique allowed for curettage of the articular surfaces of the atlantoaxial joint and the placement of bone graft to promote a permanent bony fusion. The ventral technique also allowed access to the dens for odontoidectomy procedures if necessary. More recently described ventral atlantoaxial stabilization techniques are modifications of the two-Kirschner-wire technique that often use additional Kirschner wires or bone screws in the atlas and axis [17,22,23]. The exposed ends of the Kirschner wires or bone screws are usually incorporated within a mass of polymethylmethacrylate bone cement. Other surgical methods that have been described to stabilize the atlantoaxial joint from a ventral approach include bone plating [16] and lag screw techniques [15,16].

Postoperatively, dogs should be managed in a cervical splint (as described in the following section) to protect the surgical fixation. Cage rest should be enforced for six to eight weeks. Radiographs may be made 8 weeks after surgery to confirm atlantoaxial alignment and progression of arthrodesis.

Conservative Management

Although most authors have recommended surgical stabilization, conservative management of atlantoaxial subluxation by use of a cervical splint has been reported [24], and more recently advocated [25]. If a cervical splint is used, the neck must be positioned in extension to allow for healing with the atlantoaxial joint reduced. Cervical splints must be light weight so that they are well tolerated by the pet. They may be circumferential or ventral only and should incorporate fiberglass, x-ray film, formable plastic splint material, or aluminum bars for stiffness. Cervical splints should be maintained for six to eight weeks. Radiographic evaluation should be performed prior to splint removal to confirm atlantoaxial alignment.

Restriction of activity is important to maintain atlantoaxial joint reduction and to keep the cervical splint from causing sores or loosening prematurely. Controlled, limited activity may have therapeutic value in the rehabilitation of dogs with neurologic compromise.

The success of conservative management relies on scar tissue formation across the atlantoaxial joint. As these dogs may be at risk of resubluxation later in life, long-term exercise restriction (e.g., preventing them from jumping off furniture) should be considered.

Corticosteroids such as oral prednisone have often been used in conjunction with both conservative management and surgical management. No scientific information is currently available showing that they improve the prognosis of these patients.

Prognosis

Nonsurgical management of atlantoaxial subluxation with a cervical splint was evaluated in a retrospective study [25]. Of 16 dogs, 10 (62%) were considered to have a good outcome 12 months after removal of the cervical splint. The authors of this study suggested that dogs with an acute onset of clinical signs and no prior history of neurologic disease, young dogs with immature bone in which surgical fixation may not provide adequate stability, or dogs for which there are financial constraints, should be considered for nonsurgical management. The authors of this study did not find that severity of deficits at admission was a reason to opt for a surgical approach.

Most authors consider surgical management the preferred treatment for atlantoaxial subluxation, especially in dogs with neurologic compromise. A 1995 review of the literature on surgical management of atlantoaxial subluxation revealed a 61% (32/52 dogs) success rate for dorsal procedures and a 62% (21/34 dogs) success rate for ventral procedures. Success was defined as improvement in clinical signs with no reported recurrence or necessity of reoperation [2]. These modest success rates have caused surgeons to strive to develop more reliable techniques to provide permanent rigid immobilization of the atlantoaxial joint. Three more recent studies reported modified surgical techniques with a combined 87.5% (35/40) of dogs having a good or excellent outcome.

A retrospective review of 46 cases of atlantoaxial subluxation in dogs identified risk factors affecting the outcome of surgical treatment [9]. Age of onset of clinical abnormalities of less than 24 months was significantly associated with greater odds of a successful first surgery and outcome. Duration of clinical abnormalities of less than 10 months was significantly associated with greater odds of a successful first surgery. Preoperative neurologic status was only marginally sensitive and specific in predicting a successful first surgery. Potential risk factors that did not affect the odds of a successful outcome include the type of surgery performed (the dorsal wire or suture technique vs. the ventral two-Kirschner-wire technique), grade of atlantoaxial joint reduction, conformation of the dens, and the need for a second surgical procedure. The long-term prognosis for dogs with atlantoaxial subluxation managed with surgery appears to be good provided the animal survives the perioperative period [9].