The present paper documents that atlantoaxial subluxation due to absence of the dens can cause an acute non-ambulatory tetraparesis in two different breeds of large dogs. The cases described herein showed no sign of neck pain, hence, an absence of such pain is compatible with the diagnosis. Previous reports describe neck pain, ataxia and ambulatory tetraparesis as symptoms with the congenital or developmental form of atlantoaxial subluxation in large-breed dogs [10, 20, 22, 24, 25]. Thus, clinical signs associated with this condition in large dogs can vary considerably.
In this report, we describe one German shepherd (Dog 1) and three poodles that presented with an absent dens and acute non-ambulatory tetraparesis during the first year of life. However, signs of neck pain or gait disturbances were not detected in four older poodle dogs with an absence or hypoplasia of the dens. This indicates that large breed dogs without a dens run the greatest risk of severe atlantoaxial subluxation at an early age. In adult age, hypertrophy of atlantoaxial muscles, tendons, ligaments and the joint capsule may stabilize the joint to the extent that clinical signs of subluxation do not appear. A similar theory is presented by Patton et al., who reported a rottweiler dog with a nine year history of mild neurological deficits, an absent dens, but no associated displacement between the atlas and axis. It is speculated that in large breed dogs, the large neck muscle mass together with a thickening of the atlantoaxial joint capsule and the dorsal atlantoaxial ligament, can prevent atlantoaxial instability .
In dogs 1 and 2, atlantoaxial instability and pathological changes in the cervical cords were found at the postmortem examinations. Although neck pain is a frequent finding in dogs with atlantoaxial subluxation, it was not registered at the clinical examinations. Immediate repositioning of the atlantoaxial subluxation, cage rest and treatment with analgesics probably neutralized pain arising from articular joint capsules in both of these dogs. Another possible explanation for the apparent absence of neck pain is that the pathological changes were limited to spinal cord tissue, and insignificantly affected pain sensitive structures such as meninges, nerve roots, intervertebral discs or periosteum. In toy breed dogs with atlantoaxial subluxation, neck pain is reported to occur in 30–60% [1, 3].
The lateral and ventrodorsal radiographs of dogs 1 and 2 showed no signs of atlantoaxial subluxation, but increased atlantoaxial mobility was found in both dogs at the postmortem examinations. Most likely, subluxation would have been demonstrated radiographically in both dogs if flexed lateral view radiographs had been taken. However, this 'stressed' position may cause further spinal cord injury and, therefore, was not included as part of the diagnostic procedures. However, flexed lateral view radiographs could have been taken after euthanasia, as part of the post-mortem examinations.
The calcified body that was seen on the oblique view in the mother of dog 2 was interpreted as dystrophic calcification, and is most likely the result of a stress-induced injury to the dorsal atlantoaxial ligament. In dogs with subluxation of the atlantoaxial joint a total rupture of the dorsal atlantoaxial ligament, caused by hyperflexion of the head, is commonly registered [4, 30]. In the mother of dog 2 we presume that the dorsal atlantoaxial ligament has been partially ruptured by constant atlantoaxial instability. The severe degree of dens hypoplasia and the questionable alignment of C1 and C2 are also in accordance with presence of instability. This case shows that a discrepancy between radiological findings and clinical signs may exist in dogs with atlantoaxial instability.
In dogs, the development of the axis includes fusion of seven bony elements . One of these elements (centrum of proatlas) forms the apex of the dens, whereas another (centrum 1) forms the caudal part of the dens, the cranial part of the body and a smaller cranial part of the left and right cranial articular surfaces. In dog 2, an irregularly depressed area was noted in the normal position of the dens and both cranial articular surfaces were found to be smaller than normal. These observations indicate that all bone tissue originating from both centrum of proatlas and centrum 1 was missing in this dog, not only the bone tissue that should have formed the dens.
A slightly elevated body temperature was observed following cervical spinal cord injury in dogs 1 and 2. In dog 1, a distended bladder was also palpated indicating a neurogenic failure of bladder contraction. Expressed urine contained blood, but was not cultured so a urinary tract infection cannot be ruled out. In humans with traumatic spinal cord injury, episodes of fever are common and are frequently due to urinary tract infections or pneumonia . Similar findings could be expected in dogs. On the other hand, humans with cervical spinal cord injuries are at risk of autonomic dysfunction including thermo-dysregulation and pyrexia (quad fever), which may be fatal . Although the authors of the present paper are not aware of such descriptions in dogs, quad fever could be considered in dogs with cervical spinal cord injuries with unexplained pyrexia.
Presence or absence of deep pain sensation and findings on magnetic resonance (MR) images are significant information in prognostic evaluation of dogs with an acute-onset of tetraplegia/tetraparesis secondary to atlantoaxial subluxation [16, 34, 35]. Nevertheless, it may be difficult to make a certain prognosis and thereby hard to recommend euthanasia versus surgical treatment and rehabilitation of such dogs. In a study of nonsurgical treatment of atlantoaxial subluxation in 19 dogs, Havig et al. found that a good long-term outcome was associated with an acute duration of clinical signs, but not with degree of spinal cord dysfunction at admission, radiographic appearance of the dens, age, or a history of trauma. In a prospective study including MR images of 100 human patients with a traumatic cervical spinal cord injury, Miyanji et al. found that hemorrhage, swelling and maximum compression of the spinal cord were associated with a poor prognosis for neurologic recovery. In a one-year-old toy poodle dog presented for acute-onset tetraplegia and given a guarded prognosis for full neurological recovery, Kent et al. chose conservative treatment with external coaptation for eight days before re-examining the dog and evaluating the prognosis. In the present study, both dogs showed irreversible lesions in the atlantoaxial region of the spinal cord consisting of necrosis, loss of neural parenchyma and infiltration of gitter cells. Dog 1 was euthanased three days and eighteen hours after the inciting incident. The spinal cord lesion of this dog was extensive, affecting a large proportion of both white and grey matter, making recovery very unlikely. Although the lesions in dog 2 affected a smaller area of the spinal cord, a considerable area of both grey and white matter was affected also in this dog. It is uncertain, but considered unlikely, that dog 2 could have recovered based on the histological findings.
In the present study, a normal dens was found in both parents and in all littermates of dog 1 while an absence or hypoplasia of the dens was found in six out of the eighteen examined relatives of dog 2. The total number of seven poodle dogs with an abnormal dens represented both sexes (3 males and 4 females) and three generations of dogs. The coefficient of relationship between dog 2 and her relatives with an abnormal dens varied from 0.5 (mother, full sibling) to 0.125 (half niece/nephew). Hence, there is good evidence to suggest that absence or hypoplasia of the dens in Standard poodle dogs is inherited in an autosomal way, but the underlying mechanisms producing the changes are unknown.