Szabo D, Sutherland-Smith J, Barton B, et al.
in Scientific Proceedings (Abstract). American College of Veterinary Radiology 2014.
IntroductionIPurpose: Computed tomography is increasingly being used in veterinary medicine as a means of investigating coughing and other pulmonary-related clinical signs. The purpose of this study was to assess the ability of a ratio comparing the bronchial wall thickness to adjacent pulmonary artery diameter to differentiate animals with bronchial wall thickening from unaffected animals, using animals with chronic bronchitis as a model for bronchial wall thickening .
Methods: Dogs with a clinical diagnosis of chronic bronchitis that had a thoracic CT were compared to animals receiving thoracic CT scans for reasons unrelated to a history of cough. Unaffected dogs with a diagnosis of pulmonary hypertension were excluded. The described ratio was measured using a second segmental bronchus in four locations (right cranial, right caudal, left cranial, and left caudal lung lobes). Statistical analyses were performed comparing the ratios in the two groups and also to determine whether there were any effects of age, weight, CT machine, bronchus location, or slice thickness.
Results: Sixteen dogs with a clinical diagnosis of chronic bronchitis were compared to 72 unaffected dogs. There was no significant difference based on age, weight, CT machine used, left VS. right side, or slice thickness.
Dogs with chronic bronchitis had significantly larger ratios than unaffected dogs (p<0.001). Affected and unaffected dogs had significantly larger ratios in the cranial lung lobes compared to the caudal lung lobes (p<0.001). When evaluated separately, the cranial and caudal lung lobe ratios remained significantly higher in affected dogs (p<0.001). however, a greater difference was noted in the cranial lung lobes.
A receiver operating characteristic curve using the results in the cranial lung lobes had an area under the curve of 0.912 indicating high accuracy in distinguishing between the two groups. A ratio of 0.6 had the best balance with a sensitivity of 77%, a specificity of 100%, a positive predictive value of 100%, and a negative predictive value of 95%.
Discussion/Conclusion: The ratio described here has high accuracy for discriminating dogs with bronchial wall thickening from unaffected dogs. We propose a cut off ratio of 0.6 in the cranial lung lobes, with higher ratios consistent with bronchial wall thickening.