Spinal Cord Injury in the Dog: Features of the Neurological Examination Affecting Prognosis
World Small Animal Veterinary Association World Congress Proceedings, 2008
Jacques Penderis, BVSc, MVM, PhD, CVR, DECVN, MRCVS
Institute of Comparative Medicine, Faculty of Veterinary Medicine, University of Glasgow
Glasgow, UK


The main aim of the neurological examination is to localise lesions affecting the nervous system. However, in lesions affecting the spinal cord, there are a number of other factors we can determine as part of the neurological examination, namely:

 What is the severity? This can be established most simply as the 5-point scale, originally developed by Ian Griffiths but modified by Wheeler and Sharp. The more advanced 14-point scale, subsequently developed by Natasha Olby, allows finer discrimination of recovery, but is less suited to rapid clinical assessment.

 Is there progression? Serial examinations can identify this.

 Does the lesion interrupt the local spinal reflexes to the limbs (specifically the cervical or lumbar intumescences)?

 How extensive is the lesion?

 Are there unfavourable underlying causes? For example, the identification of gradual progression of the lesion localised to the cranial-thoracic region would be more consistent with neoplasia than intervertebral disc disease.

Establishment of the Prognosis as Part of the Neurological Examination

The most important clinical feature affecting the prognosis or outcome following spinal cord injury is still the initial severity grading at the time of presentation. Severity grading in animals is complicated by difficulty in assessing a number of the sensory pathways within the spinal cord, including temperature discrimination, pressure, touch, etc. However, there are still enough spinal cord pathways of varying susceptibility that we can assess to allow us to accurately grade spinal cord injuries. The critical prognostic feature is the presence or absence of conscious perception of a painful stimulus applied to the affected limbs. Unfortunately, accurate discrimination of a conscious response to pain is often still poorly distinguished from the withdrawal (or pedal) reflex by many veterinarians in practice and it is therefore often difficult to give accurate prognostic advice over the telephone in many of these cases.

5-Point Scale: Originally Developed by Ian Griffiths, but Modified by Wheeler and Sharp


 0: Normal

 1: Pain: not severe enough to result in any neurological dysfunction.

 2: Paresis with or without pain: as the lesion becomes more severe the degree of paresis and/or proprioceptive deficits becomes more severe.

 3: Plegia: total loss of voluntary movement in the affected limbs (and/or tail).

 4: Plegia with loss of voluntary urinary function.

 5: Plegia with loss of voluntary urinary function and loss of conscious perception of painful stimuli ('deep pain perception') in the affected limbs (and/or tail).

In particular, the 5-point grading system is useful for predicting the outcome following intervertebral disc disease. The table modified from Wheeler and Sharp summarises the major outcome predictors:


The outcome predictors have now been modified for intervertebral disc disease following decompressive surgery in a variety of studies. We are now able to predict that for grades 1 to 4 the outcome is very good, with most dogs making an acceptable recovery, and for grade 5 intervertebral disc disease following decompressive surgery there is a 50 to 60% chance of a successful recovery in the first 3 months, with a further 10% demonstrating a delayed return to ambulation over the next 18 months. Approximately half of the dogs that never recover deep pain sensation will demonstrate ambulation at 18 months (so-called 'spinal walking'), however in these cases the gait is obviously abnormal and the dogs never recover voluntary urination.

14-Point Spinal Grading Scale Developed by Natasha Olby (Journal of Neurotrauma 2003; 21:49-59)

Stage 1:

 0: No pelvic limb movement and no deep pain sensation.

 1: No pelvic limb movement with deep pain sensation.

 2: No pelvic limb movement but voluntary tail movement.

Stage 2:

 3: Minimal non-weight-bearing protraction of the pelvic limb (movement of one joint).

 4: Non-weight-bearing protraction of the pelvic limb with more than one joint involved less than 50% of the time.

 5: Non-weight-bearing protraction of the pelvic limb with more than one joint involved more than 50% of the time.

Stage 3

 6: Weight-bearing protraction of pelvic limb less than 10% of the time.

 7: Weight-bearing protraction of pelvic limb 10-50% of the time.

 8: Weight-bearing protraction of pelvic limb more than 50% of the time.

Stage 4

 9: Weight-bearing protraction 100% of time with reduced strength of pelvic limb. Mistakes 90% of the time.

 10: Weight-bearing protraction of pelvic limb 100% of time with reduced strength. Mistakes 50-90% of the time.

 11: Weight-bearing protraction of pelvic limb 100% of time with reduced strength. Mistakes 50% of the time.

Stage 5

 12: Ataxic pelvic limb gait with normal strength, but mistakes made 50% of the time.

 13: Ataxic pelvic limb gait with normal strength, but mistakes made 50% of the time.

 14: Normal pelvic limb gait.

What Factors may Therefore Indicate that Treatment is Less Likely to be Successful?

Consistent with the information presented above and with what most of us use as a guide in practice, factors that indicate treatment is less likely to be successful include:

 Prolonged loss of deep pain perception, particularly if it is greater than 48 hours.

 Spinal fracture with loss of deep pain, irrespective of how long there has been loss of deep pain perception. In one study by Olby and others (JAVMA 2003; 222: 762-769), outcome in dogs with paraplegia and loss of deep pain was compared between intervertebral disc disease and spinal trauma. None of the spinal trauma cases regained deep pain perception, while 58% of the dogs with intervertebral disc disease regained deep pain perception and the ability to walk.

In addition to the severity score and duration of the lesion, studies have indicated some further considerations that we should be aware of, but in all of these the first consideration in establishing the prognosis is still the severity score and duration of the lesion. The main limitation for studies trying to interpret these changes is the lack of a fine grading scale in order to accurately measure outcome. While the 5-point scale is useful for a rapid determination of outcome in the clinic, it is not suited to fine discrimination of outcome in prospective trials and, in particular, not for retrospective studies.

Some of the documented additional outcome measures include:

 Extensive and severe spinal cord lesions, in particular extending over more than 2 vertebrae in length. The prognosis with extensive lesions due to haemorrhage secondary to intervertebral disc prolapse in one study was shown to be reasonable, with a return to ambulation in 21 out of 23 dogs with lesions extending over 3 to 7 vertebrae (JSAP 2005; 46: 485-490). A further consideration in extensive lesions with loss of deep pain perception is the development of ascending myelomalacia. If there is clinical evidence for myelomalacia then obviously the prognosis is hopeless.

 Severe lesion with involvement of lumbar or cervical intumescences. A study evaluating intervertebral disc extrusions affecting the lumbar intumescence (JAAHA 1999; 35: 323-331) concluded that this was not associated with a worse outcome. However, closer examination of the study only reveals that, in the authors' population, intervertebral disc disease that truly affects the lumbar intumescence is rare, and their population did not allow adequate comparison. The major spinal cord disease in canine practice that regularly affects the intumescences if fibrocartilaginous embolisms (FCE). In one study, a correlation was demonstrated between involvement of the intumescence and a poor clinical outcome (JSAP 2003; 44: 76-80).

 MRI signal change within the spinal cord. In the majority of cases there is no change in signal intensity in the spinal cord following injury and this is generally associated with a good outcome. However, if spinal cord signal intensity does alter (typical hyperintensity on T2-weighted images, with either iso- or hypointensity on T1-weighted images) then this is associated with a variable outcome, with some patients making a good recovery and some not demonstrating any recovery (Radiat Med 1993; 11: 127-138). In dogs demonstrating an area of hyperintensity of the spinal cord this was correlated with a poor outcome (JAVMA 2005; 227: 1454-1460).

 Degree of spinal cord compression and spinal canal stenosis. The degree of spinal cord compression, expressed either as maximum spinal canal compromise or maximal spinal cord compression (Spine 2007; 32: 2083-2091), has been associated with outcome in acute spinal cord injury for a number of different clinical settings in human patients (for example, in traumatic cervical spinal cord injury). In an experimental model in the rat (Spine 1999; 24: 1623-1633) the prognosis for neurological recovery was adversely affected by the degree of spinal cord compression; histological examination of the spinal cord revealed more severe spinal cord damage as the degree of compression increased. The reason there does not appear to be a good correlation in the dog (JSAP 2006; 47: 644-650) between the degree of spinal cord compression and outcome in Hansen type-1 intervertebral disc disease is most likely related to study design and the absence of a fine grading system for severity, and spinal cord compression and spinal canal stenosis have therefore not been excluded as a factor in dogs.

 Unfavourable underlying disease. Needless to say, the identification of an unfavourable underlying disease process indicates a poor prognosis.

 Age and body weight. The juvenile nervous system has a tremendous degree of plasticity, expressed as reserve capacity following injury and, therefore, the prognosis for equivalent injuries in the juvenile dog are better as compared to older dogs. In the study by Olby and others (JAVMA 2003; 222: 762-769) the speed of recovery of ambulation was significantly correlated to both age and body weight. The ease of management of recumbent large breed dogs and the effect of this on outcome should also be considered and, in the study by Cauzinille and Kornegay (JVIM 1996; 10: 241-245), dog size was potentially a deciding factor in electing for euthanasia. A separate consideration is the mortality rates in older human patients and how this may affect our canine patients. In older human patients (> 70 years of age) the 1-year mortality rate is 27.7% as compared to 3.2% in younger patients.

 Duration of the clinical signs prior to treatment. Although the contribution of duration of clinical signs has already been discussed in relation to deep pain absent dogs with intervertebral disc disease, this factor also plays a role in less severe spinal cord injury. In an experimental model of spinal cord injury in the dog (J Bone Joint Surg Am 1995; 77: 1042-1049) there was a direct correlation with the speed of decompression and clinical outcome (measured either as a return to ambulation or by means of somatosensory evoked potentials), and a similar situation has been demonstrated in the rat (Spine 1999; 24: 1623-1633). Even in the short term, the duration of spinal cord compression has an effect in the dog, with dogs having spinal cord compression of only 90 minutes having a lower severity score and better functional recovery than dogs having spinal cord compression of 180 minutes duration (J Bone Joint Surg Am 2--3; 85: 86-94). However, in the Olby study (Journal of Neurotrauma 2003; 21: 49-59) and an additional study (JSAP 2002; 43: 158-163) the outcome in dogs with a less severe severity was good irrespective of the duration of the clinical signs, although a longer duration did correlate with a longer recovery period. The rate of onset of clinical signs was correlated with a worse outcome. In a separate retrospective study of thoracolumbar disc extrusion there was no correlation between the duration of clinical signs and outcome (JAVMA 2005; 227: 1454-1460).

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Jacques Penderis, BVSc, MVM, PhD, CVR, DECVN, MRCVS
Institute of Comparative Medicine, Faculty of Veterinary Medicine
University of Glasgow
Glasgow, United Kingdom

MAIN : Neurology : Spinal Cord Injury
Powered By VIN