Evaluating the reliability of computed tomography for the diagnosis of intervertebral disc extrusion in dogs

PICO question 
In dogs with an acute thoracolumbar myelopathy, is non-contrast computed tomography (CT) a reliable method for the diagnosis of intervertebral disc extrusion, compared to magnetic resonance imaging (MRI)? 
  
Clinical bottom line 
Category of research question 
Diagnosis 
The number and type of study designs reviewed 
Four papers were critically reviewed. Two were retrospective, cross-sectional studies, and two were prospective, observational cohort studies 
Strength of evidence 
Moderate 
Outcomes reported 
The current literature suggests that CT is often sufficient for the diagnosis of thoracolumbar intervertebral disc extrusion, with MRI superior to CT for lesion characterisation. Non-contrast CT is likely sufficient for the diagnosis and surgical planning for intervertebral disc disease in Dachshunds. However, MRI is recommended for diagnosis and surgical planning of thoracolumbar intervertebral disc disease in non-Dachshund breeds 
Conclusion 
Computed tomography is often sufficient for the diagnosis of thoracolumbar intervertebral disc extrusion. However, MRI is superior to CT for lesion characterisation and it is therefore recommended to use MRI for cases requiring surgical planning. Computed tomography is likely sufficient for the diagnosis and surgical planning of intervertebral disc disease in Dachshunds 
Regard for the reason of diagnostics performed (e.g. surgical planning) and the likelihood of other differential diagnoses (e.g. spinal neoplasms) should be taken before deciding to use one imaging modality over the other. Consideration into the risks associated with anaesthetising an animal with a spinal cord lesion, the ability of the practitioner to accurately localise the myelopathy and interpret the results, should be taken before pursuing diagnostics in these cases 
  
How to apply this evidence in practice 
The application of evidence into practice should take into account multiple factors, not limited to: individual clinical expertise, patient’s circumstances and owners’ values, country, location or clinic where you work, the individual case in front of you, the availability of therapies and resources. 
Knowledge Summaries are a resource to help reinforce or inform decision making. They do not override the responsibility or judgement of the practitioner to do what is best for the animal in their care. 
  



The evidence
Of the four papers reviewed, two were prospective and two were retrospective studies in nature. The strength of evidence for each paper is considered moderate, with the absence of a systematic review or meta-analysis for this topic.
Paper one (Cooper et al., 2014) is the only study to directly compare the sensitivity of MRI versus CT in the diagnosis of thoracolumbar intervertebral disc disease in dogs and this evidence was therefore considered stronger when reaching a clinical bottom line. However, the other three papers all provide insight into the use of both imaging modalities in the diagnosis of thoracolumbar myelopathies and provide evidence to concur with the conclusions drawn by Cooper et al. (2014).

Summary of the evidence 1. Cooper et al. (2014)
Population: Dogs with neurological signs localised to T3-S1.
Intervention details: All dogs had both a non-contrast CT and an MRI.
Study design: Observational cohort study.

Outcome studied:
• Sensitivity of non-contrast CT versus MRI for detecting intervertebral disc herniation. • Accuracy of non-contrast CT versus MRI for lesion localisation and lateralisation. • No evaluation of CT positive, MRI negative disc herniations and therefore an inability to fully compare MRI to CT for lesion characterisation.

Sample size: Eight dogs.
Intervention details: Three dogs had a CT myelogram and five dogs had an MRI.
Study design: Retrospective cross-sectional study.
Outcome studied: Presence of intradural herniation as confirmed at surgery.

Main findings: (relevant to PICO question):
• MRI misdiagnosed all cases in the study, failing to identify any intradural herniation and instead characterising the cases as standard intervertebral disc herniation. • The 'golf tee sign' and subarachnoid filling deficits were observed in 3/8 cases using CT myelography, suggesting that CT myelography may be more sensitive for detecting intradural disc herniations.

Limitations:
• MRI and CT examinations were not performed in the same animal and so could not be directly compared. • The MRI used was a low-field scanner and therefore findings may differ if a high-field scanner were used. • 5/8 cases evaluated were of the same breed and therefore the results may not be representative of every breed. • Both MRI and CT studies were not standardised, with different techniques and machines used (for example, not all studies included both T1-weighted and T2-weighted scans). • Small case numbers. • Descriptive analysis only performed.

Emery et al. (2018)
Population: Dogs with thoracolumbar myelopathies. • No statistically significant difference was noted in the need for further imaging when comparing chondrodystrophic to non-chondrodystrophic breeds. • The presence of a normal CT, or multiple intervertebral disc herniations identified on non-contrast CT (and therefore the difficulty distinguishing acute from chronic lesions) were both cited as common reasons for requiring additional imaging techniques. • 42/555 (7.6%) of dogs presenting with a thoracolumbar myelopathy required further advanced imaging beyond a plain CT scan. • CT scans performed with positive contrast media intravenously did not alter the need for additional imaging, compared to those where an intravenous contrast agent was not used.

Limitations:
• The accuracy of lesion localisation and lateralisation was not assessed in this study and therefore cannot be evaluated. • Retrospective study and therefore no standardisation of protocols used. • Referral population with patient pre-selection of imaging modality based on signalment and history. Therefore, not a true representation of the general case population.
Intervention details: All dogs underwent CT and MRI.
Study design: Prospective observational study.

Outcome studied:
• Influence of pre-operative planning considerations for hemilaminectomies based on CT versus MRI. • Factors studied include; location, lateralisation and extent.

Main findings: (relevant to PICO question):
• Large range in inter-observer agreement for both imaging modalities (43.5-66.6%) when assessing the site, side and size of laminectomy approach. • All observers planned a larger laminectomy defect based on MRI compared to CT imaging. • All lesions were identified in cases when using MRI, however lesions were not identified in four of these cases when assessing CT images alone.

Limitations:
• Confirmation of imaging findings was not evaluated (e.g. surgical visualisation of disc herniation). • No statistics therefore performed to directly compare the reliability of CT and MRI in detecting lesion localisation.

Appraisal, application and reflection
The literature search performed found four papers which addressed, or partially addressed the PICO.
MRI was found to have an increased sensitivity for the detection of intervertebral disc disease in dogs, compared to non-contrast CT (Cooper et al., 2014;and Noyes et al., 2017). However low-field MRI misdiagnosed all cases of thoracolumbar intradural disc herniation when compared with CT myelography in dogs (Tamura et al., 2015).
Non-contrast CT appears to be a sufficient imaging modality for intervertebral disc disease and surgical planning in the majority of Dachshunds (96.4%; 323/335) (Emery et al., 2018).
Regard for the reason for diagnostics performed (e.g. surgical planning) and the likelihood of other differential diagnoses (e.g. spinal neoplasms) should be taken before deciding to use one imaging modality over the other. Cost, procedure time, and need for a general anaesthetic are also factors that should be considered when assessing between MRI and CT examinations.

Methodology Section
Search Strategy