Diagnostic utility of computed tomography imaging in equine intracranial conditions.

REASONS FOR PERFORMING STUDY

The use of computer tomography (CT) and contrast-enhanced CT (CCT) to image the head is common. However, the validity of CT as a neurodiagnostic indicator of intracranial diseases in horses is unknown.

OBJECTIVE

To define the validity of CT and CCT in horses with suspected intracranial disorders.

METHODS

The validity of CT imaging was estimated by comparing clinical, clinicopathological and histopathological findings to CT findings in 15 horses presented for intracranial disorders, for which pre- and post contrast CT images and post mortem examination of the brain and skull were reviewed. Post mortem examination (gross and histopathological examination) was considered as the gold standard; and sensitivity, specificity, predictive values, likelihood ratios, and pre- and post test probabilities were calculated.

RESULTS

All horses had abnormal neurological examinations on admission. Computer tomography imaging identified intracranial lesions in 8 horses, and included masses (oligodendroglioma, adenocarcinoma and cholesterinic granulomas), acute haemorrhage and skull fractures. Computer tomography imaging failed to identify intracranial lesions in 6 cases, which included meningitis, meningoencephalitis and nonacute haemorrhage. Lesions not recognised by CT were also not evident on gross examination but were identified by histopathological examination of the brain. Post mortem examination of the brain and skull was unremarkable in one horse, for which cranial CT imaging was normal (specificity, 100%). Therefore, the odds of having an intracranial lesion after an abnormal CT were very high. In contrast, there was a moderate sensitivity (57.1%, 95% confidence interval: 29.6-81.2).

CONCLUSIONS AND POTENTIAL RELEVANCE

CT was an excellent neurodiagnostic tool in identifying skull fractures, intracranial space-occupying lesions (e.g. neoplasia) and acute haemorrhage and allows to rule in intracranial disorders. However, CT showed limited sensitivity in identifying inflammatory disorders and small parenchymal lesions in the equine brain, which was not further detectable after contrast administration.