Diagnostic accuracy of brush cytology in canine chronic intranasal disease

Abstract

Most cases of canine chronic intranasal disease cannot be differentiated based on clinical examination alone, and biopsy is often required for a definitive diagnosis. Nonsurgical cytologic and histologic biopsy techniques represent desirable diagnostic approaches. The aim of this retrospective study was to determine the diagnostic accuracy of brush cytology in differentiating non-neoplastic and neoplastic diseases in dogs with chronic intranasal disease. Cytologic samples of lesions in dogs with chronic intranasal disease were obtained by brushing over a 12-year period. All dogs had complete physical examinations as well as radiographic, rhinoscopic, and cytologic evaluation. Histologic diagnosis, follow-up clinical information, or both were used as the gold standard, and dogs free of disease or with no progression of disease at 1 year were considered negative for neoplasia. Indicators of performance of brush cytology in detecting neoplasia were calculated and included sensitivity, specificity, positive and negative likelihood ratios, and diagnostic odds ratio. Samples of nasal brushings from 138 dogs were evaluated. Of 62 cases of neoplastic disease, true-positive and false-negative diagnoses were made using cytologic evaluation in 44 (71.0%) and 18 (29.0%) cases, respectively. False-negative diagnoses of neoplasia were not attributed to low cellularity, but to the presence of inflammatory cells that masked neoplastic cells. Brush cytology had a sensitivity of 0.71, specificity of 0.99, positive likelihood ratio of 53.94, negative likelihood ratio of 0.29, and diagnostic odds ratio of 188.33. Brush cytology has good diagnostic accuracy for chronic intranasal lesions in dogs.