Simon R. Platt, BVM&S, MRCVS, DACVIM (Neurology), DECVN
The diagnosis and treatment of spinal injury is a very controversial topic in veterinary medicine. Direct comparisons of study results for diagnostic and treatment modalities are difficult because of the variability of neurologic dysfunction, duration of clinical signs and lack of long-term follow studies. The outcome for an individual patient can be dependent on a multitude of different factors, not all of which we can control for, and hence the prognostication for an individual can be difficult. An owner's decision to pursue therapy may be solely based upon such prognostic advice and so we desperately need continued investigation into this area and more in the way of long-term studies. Here we will review the assessment and current treatment recommendations for dogs with acute spinal injury.
Assessment of the Spinal Injury Patient
Complete routine blood work and urinalysis should be obtained if possible; otherwise a PCV, total protein level, BUN assessment, glucose and electrolyte levels should be ascertained. Cardiovascular stability should be investigated with the aid of an ECG and blood pressure measurements. The important systemic parameters to monitor, their suggested reference values after trauma and management protocols are similar to those following head trauma.
A thorough physical and orthopedic examination can follow the initial patient evaluation and stabilization. Consideration should also be given to obtaining a coagulation panel, a buccal mucosal bleeding time and a platelet count if there has been associated hemorrhage. A patient which has experienced blood loss or which is expected to during a surgery, should be blood-typed or cross-matched and appropriate blood products should then be obtained.
It is essential to complete a thorough examination of the nervous system. Errors in diagnosis commonly occur when the only the region of the neurologic deficit is examined and more subtle alterations in other parts of the nervous system are overlooked. A modified neurological examination is necessary when a patient with a suspected spinal cord injury is stabilized in lateral recumbency, to prevent further injury that may result from movement of unstable vertebrae. The extent and severity of a spinal cord injury usually may be assessed accurately during an initial neurologic examination. Five groups of clinical signs are seen to a varying degree in all animals that have a spinal cord injury:
1. Reduction or loss of voluntary movement
2. Alteration of spinal reflexes
3. Changes of muscle tone
4. Muscle atrophy (with chronic injuries)
5. Sensory dysfunction
Careful assessment of each of these groups facilitates lesion localization to one of four major regions of the spinal cord: C1-C5; C6-T2; T3-L3; and L4-Cd5. When multiple fractures occur, the clinical signs of a more caudal lesion may mask those of a second lesion located further cranially. Pain assessment is extremely important in spinal cord trauma patients. Analgesics should not be given until after deep pain has been assessed. Pain perception is assessed by applying a painful stimulus and observing for a brain-mediated (often behavioral) response. The stimulus applied to the digit may result in withdrawal of the limb by a spinal reflex mechanism, even though the spinal cord may have been severed cranial to the spinal cord segments mediating the reflex. When assessing the response to a painful stimulus in animals in which deep pain is apparently absent. It is essential to take the time to ensure the patient is as relaxed as is possible, so that heart rate, respiration rate and pupil size can be noted prior to the noxious stimulus.
Thoracic radiographs should be evaluated after a significant trauma, looking for pleural effusions, contusions, pneumo-mediastinum and--thorax as well as the possibility of pericardial effusions and diaphragmatic herniation. If a vertebral injury is suspected, it is recommended to take survey radiographs of the entire spine prior to additional manipulation of the animal. Sites particularly predisposed to fracture and luxation include the atlantoaxial junction, the thoracolumbar junction and the lumbar and lumbosacral spine. As some fractures can be subtle, good quality and well-positioned radiographs from 2 different planes are necessary. This may be accomplished with the animal awake and immobilised. Poor radiographic technique resulting in rotation of the spine (especially in the cervical area) can make assessment for unstable and malaligned vertebral segments difficult. Extreme care should be taken in positioning the animal for ventrodorsal views: horizontal beam radiographs can be taken if the equipment is available. Sedation may be necessary to achieve accurate positioning for radiographs in some animals. This should not be performed, however, if the examiner is unsure of the physical diagnosis, as sedation often influences the results of the neurological examination. Additionally, sedation or anesthesia results in the loss of voluntary paraspinal muscle contraction and unstable vertebral segments may be more likely to subluxate. It is important to remember that radiographs provide a static record of the location of the vertebrae at the time of the study; however, they do not allow for assessment of how extensive the displacement of the vertebrae was at the time of the injury and prior to radiology. As a result of the strong paraspinal musculature, vertebrae can be significantly displaced acutely at the time of injury but then subsequently pulled back into a more normal position. A scheme has been devised for predicting spinal instability in human patients based upon the degree of vertebral damage, which has been modified for use in animals. In this model the vertebrae are divided into three compartments. Damage to two or more components indicate the need for internal or external stabilization.
Myelography or other advanced imaging such as computed tomography (CT) or magnetic resonance (MR) imaging is needed to evaluate spinal compression and to ensure that additional lesions unidentifiable on survey radiographs are not present. CT is invaluable in identifying bone defects that may not be apparent with survey radiography. Three-dimensional reconstruction from CT images may provide additional anatomic information regarding bone contour for surgical planning. MR imaging has the distinct advantage of showing intramedullary spinal disease and soft tissue injuries as well as multiplanar images, which at least in the case of dogs without nociception can be prognostic.1,2 Multiple sites of disc compression can be found when the spinal injury is due to disc disease on account of the sensitivity of MRI and as such the determination of the most significant lesion can be difficult.3,4 MRI studies on disc disease in dogs have failed to identify an association of the degree of compression with severity of disease but they are able to differentiate between compressive and non-compressive acute disc disease, which can help with surgical treatment planning.5,6
Factors influencing the prognosis of the acutely paralyzed animal include the underlying disease, the severity of the neurological signs and the duration of those signs.
Severity of Neurological Signs
Deep pain perception or nociception in affected limbs is the single most important prognostic indicator of injury severity that can be obtained from the neurological examination.7 Nociception is tested for by applying heavy pressure to the bones of the digits with large hemostatic forceps or heavy pliers such that pressure is applied to the underlying periosteum. Nociception is present if the animal displays a conscious awareness of the stimulus rather than simply a reflex withdrawal of the limb. Typically both medial and lateral digits of each paralyzed limb are tested and the tail is tested. Lack of deep pain perception implies functional spinal cord or peripheral nerve transection at the time of testing and provides useful prognostic information. As a general rule, animals without nociception due to acute disc disease have approximately 60% potential to recover motor function, if the underlying disease process can be prevented from progressing.7 This potential seems to drastically reduce within hours of the injury to the spinal cord and doesn't seem to exist at all if the injury is due to a spinal fracture or luxation.7
The prognosis of paraplegic animals that lack deep pain perception in their pelvic limbs varies with the cause and generally worsens the longer the duration of signs. It is unusual to be presented with a tetraplegic animal that lacks deep pain perception as functional cervical spinal cord transection causes paralysis of the respiratory muscles. In addition, the sympathetic nervous system is interrupted as it runs down the cervical spinal cord, resulting in pronounced bradycardia. As a result, animals with extremely severe cervical spinal cord injuries die rapidly as a result of hypoventilation and cardiac arrest.
Medical treatment of acute spinal cord concussion and ischemia is aimed at limiting the final extent of secondary tissue damage. Minimizing secondary injury is generally achieved by ensuring adequate perfusion and oxygenation of the animal and administration of neuroprotective agents. However treatment must be initiated as soon as possible after injury as the majority of secondary tissue damage has occurred within 24 hours of the primary injury. The first consideration is systemic blood pressure and oxygenation, particularly in the trauma victim. In the normal spinal cord, perfusion is maintained in the face of changes in systemic blood pressure by an autoregulatory process. Autoregulation is lost in the injured spinal cord, and hypotension results in a further decrease in the already compromised perfusion of the injured segment. Hypoxaemia exacerbates the local energy failure. Hypotension should be treated by administration of appropriate fluids and oxygen supplementation provided by face mask, nasopharyngeal or transtracheal catheter if necessary.
Unfortunately there is little if any objective information available on the most effective treatment for acute spinal cord injury in dogs, and so treatment protocols for humans with acute spinal cord injuries have been adopted by veterinarians. It should be emphasized that the efficacy of these protocols has not been established in dogs with spontaneous spinal cord injuries. Although many different therapeutic agents, including opioid antagonists and agonists, calcium channel blockers, and glutamate receptor antagonists have been shown to be protective experimentally, the only drugs shown to be of benefit in randomized, prospective clinical trials are methylprednisolone sodium succinate (MPSS), and its derivative, tirilazad.
MPSS remains the standard of care in humans. MPSS has been shown to be effective because of its free radical scavenging properties rather than its anti-inflammatory effect. In order to obtain this effect, it must be used at high doses and treatment should be initiated within 8 hours of injury. Suggested protocols include initiation of treatment with MPSS within three hours of injury at a dose rate of 30mg/kg given intravenously and followed with either a constant rate infusion of 5.4mg/kg/hr for 24 hours, or second and third boluses of 15mg/kg at two and six hours after the first dose and then a constant rate infusion of 2.5mg/kg/hour for 24 hours. If treatment is initiated between 3 and 8 hours after injury, the recommendation is to continue treatment for 48 hours rather than 24 hours. Delaying initiation of treatment for more than 8 hours has a detrimental effect on outcome in people. As MPSS has both glucocorticoid and free radical scavenging effects, it is postulated that delaying treatment until after the majority of free radical induced damage has occurred is more likely to result in glucocorticoid side-effects. Indeed, although there continues to be wide spread use of glucocorticoids, such as dexamethasone, to treat acute spinal cord injuries in veterinary practice, there is no good evidence that such drugs are beneficial and the side effects have been well documented and at this time they cannot be recommended as efficacious treatments.
Recently, substances that fuse membranes ("fusogens") have been proposed as a treatment for acute spinal cord injury. Polyethylene glycol (PEG) is a hydrophilic polymer that targets damaged membranes following intravenous administration and seals membranes preventing intracellular leak of ions and subsequent axonal disruption, restoring axonal conduction. In an experimental model of spinal cord injury in guinea pigs, there was rapid improvement of electrophysiological parameters and of the cutaneous trunci reflex following treatment with PEG.8 No adverse effects were reported in a phase I trial of PEG completed in dogs with paralysis and loss of pain sensation due to acute disc herniations;9 60% of the dogs did recover function but this preliminary study was not blinded and included a historical control group. A large multicenter trial coordinated by NCSU is currently underway evaluating this drug for the treatment of dogs with severe spinal injury due to disc disease.
Dogs with compressive lesions and/or spinal instability should ideally undergo surgical decompression and/or stabilization as soon as possible. Surgery for vertebral fractures or luxations is the most effective means of accurately realigning and stabilizing affected vertebrae and of rapidly and effectively decompressing the spinal cord. With few exceptions, vertebral column fracture-luxations are a surgical emergency. Surgical stabilization of the vertebral column permits early patient mobilization and physical therapy because the risk of further injury from movement of an unstable vertebral column is minimized. This approach applies to all spinal cord injuries, even those that result in apparent loss of deep pain perception.
1. Ito D, et al. J Am Vet Med Assoc 2005;227(9):1454.
2. Platt SR, et al. Vet Radiol Ultrasound 2006;47(1):78-82.
3. Besalti O, et al. J Am Vet Med Assoc 2006;228(6):902.
4. Besalti O, et al. Can Vet J 2005;46(9):814.
5. Penning V, et al. J Small Anim Pract 2006;47(1):78.
6. Chang Y, et al. Vet Rec 2007;160(23):795.
7. Olby N, et al. J Am Vet Med Assoc 2003;222(6):762.
8. Shi R, et al. J Neurophysiol 1999;81:2406.
9. Laverty PH, et al. J Neurotrauma 2004;21(12):1767.