When assessing representative images of fine-needle aspirate cytological smears of canine mast cell tumours, interobserver agreement is influenced by educational attainment.

Abstract

OBJECTIVES: Cytology is a common diagnostic tool in veterinary practice, and canine mast cell tumours are presumed to be easy to diagnose. However, the potential influence of the observer's educational attainment on the interobserver agreement when interpreting representative images of cytological smears has not yet been investigated. MATERIALS AND METHODS: In total, 25 observers were equally divided into five groups based on educational attainment (board-certified diplomates of the European College of Veterinary Surgeons specialised in companion animals, board-certified of the European College of Veterinary Internal Medicine-Companion animals, general practitioners, interns and final-year veterinary medicine students) and asked to evaluate 36 distinct sets of images from cytological smears, of which 18 originated from canine mast cell tumours and 18 from non-mast cell tumours. No additional training was given. The interobserver agreement and kappa were calculated. RESULTS: For mast cell tumour diagnosis and mast cell presence, eosinophils and lymphocytes, specialists in internal medicine had the highest interobserver agreement, whereas interns or final-year students had the lowest. Differences in interobserver agreement were significant for mast cell tumour diagnosis and the presence of mast cells. CLINICAL SIGNIFICANCE: Our results emphasize the importance of training to improve the consistency of classifying cytological smears as mast cell tumours or not.