• canine;
  • radiography;
  • temporomandibular

Investigation of temporomandibular joint (TMJ) disease requires a clear diagnostic image, which can be challenging to obtain using conventional radiography. The aim of this study was to compare five different oblique radiographic views with the head in lateral recumbency, assessing the clarity of visualization of the normal TMJ anatomy. The views under investigation were the laterorostral–laterocaudal oblique at a 10° and 20° rotation of the head (“nose-up” view), laterorostral–laterocaudal oblique with a rostrocaudal X-ray beam angulation of 10° and 20°, and a parallax view with the beam centered over C2 and collimated to include the TMJ region, using the divergence of the X-ray beam to project the TMJs separately on the radiograph. The views were performed on both TMJs of thirty canine cadavers and were graded independently by experienced and inexperienced observers. Grading was performed on the mandibular fossa, condylar process, joint space, retroarticular process, and the overall TMJ, and was based on a four-point scale. Mean grades for each component and for the overall joint were compared for each observer and each projection. Mean grades were significantly (P < 0.05) higher for the “Nose-up” projections than the angled beam or parallax projections, as was interobserver agreement, and both observers showed significantly higher (P < 0.05) mean grades for the 20o “Nose-up” angulation than the 10o “Nose-up” angulation. These results suggest that a latero 20o rostral–laterocaudal oblique gives the best representation of the anatomy of the TMJ of the dog of the projections assessed, and should be considered when investigating clinical cases of TMJ disease.