Discussion

Computed tomography (CT) has been used increasingly over the last decade in equine hospitals, and has been particularly useful for imaging the equine head and neck. In the last 3 - 4 years, its availability has grown tremendously in the UK with the installation of at least 7 CT machines in various private practices and universities. These new facilities are all capable of scanning the head (+/- proximal neck) of standing sedated horses. Whilst these 'standing' CT machines may necessitate a small compromise in image quality compared to scans acquired in anaesthetised patients, their greater accessibility and the relative reduced costs to owner and risks to the patient mean that they are the technique of choice in most cases for imaging the head.
Computed tomography has proven to be particularly useful in the diagnosis of equine dental disease in clinical cases where radiographic examinations are inconclusive, and also in complicated cases such as dental dysplasia, post extraction complications, occult apical infections and dental sinusitis where detailed cross-sectional images can provide the clinician with concrete evidence on which to decide between treatment options. Despite the fact that CT is being increasingly used for the diagnosis of equine dental disease, surprisingly little information has been published on the CT appearance of equine dental tissues in health and disease (Simhofer and Boehler 2010). There is much variability in the CT appearance of clinically normal equine cheek teeth (Windley et al. 2009a,b). The CT features of horses with periapical infection have been published in 2 papers (Henniger et al. 2003; Veraa et al. 2009), although clinical experience with the technique has increased markedly since their publication. Pathological findings associated with periapical infection include fragmentation of root structures, hypoattenuation (gas) within the pulp chambers, gross remodelling or lysis of apical and periapical structures (including dental roots and alveolar bone), subcutaneous soft tissue swelling and soft tissue attenuation within the dental alveolus. Limitations when assessing cheek teeth for the presence of apical infection include partial volume artefacts which 'blur' the edges of the pulp chambers and marked intra-horse and intra-tooth variation when comparing one root or apex to others as nondiseased 'controls'.
Anatomical and CT studies have now shown that hypoattenuation within the infundibulue of maxillary CT is a normal variation in many teeth and is not usually associated with a pathological process. Although CT has been used to image dental fractures, in my opinion such lesions should be easily diagnosed during detailed oral examination and do not require advanced imaging techniques. For horses with dental 'slab' fractures and concurrent clinical signs of apical infection, CT can be useful to help decide whether it is the fractured tooth or an adjacent tooth (or neither) that is the cause of these signs.

For sinus disorders, CT provides a much more accurate picture of which sinus compartments are involved, the extent of involvement and concurrent structures which may be affected by the disease process and can therefore prove invaluable for presurgical planning. Computed tomography allows for identification of cases with involvement of the sphenopalatine and ethmoidal sinuses and of intra-cranial and retro-bulbar extension of intra-sinus masses. These specific features are almost impossible to identify on radiographs and may markedly alter the surgical approach and prognosis for the case. Cases of dental sinusitis are usually easily identified with CT, whereas radiography has been shown to be relatively unreliable in such cases due to superimposition of radiopaque sinus contents on the dental alveolus.
There are, however, limitations of CT when investigating equine sinusitis: intra-sinus masses, pus, blood and thickened/oedomatous sinus mucosa can be very hard to differentiate between, even if the HU of the material is recorded. For instance, the HU range of sinus contents reported for horses with primary sinusitis is -30 - +40 HU (Henninger et al. 2003) and for sinus neoplasms is 33 - 47 HU (Cissell et al. 2011), thus there is considerable overlap. One study of CT features of equine sinonasal neoplasia found that 13/15 neoplasms could not be differentiated from adjacent soft tissue structures (Cissell et al. 2011). Although many tumours in this study showed evidence of destruction of the ethmoturbinates, overlying facial bones and the infra-orbital canal, such changes can also be observed in horses with sinus cysts and intra-sinus ethmoidal haematomas. Therefore in some cases, a sinusotomy or sinoscopically guided biopsy still provides the only means of making a definitive diagnosis.