Abstract
INTRODUCTION: Among thyroid malignancies, papillary thyroid carcinoma (PTC) is the most common, with the classical variant being the most common subtype. Some histological variants have aggressive behavior, advanced presentation stages, poor clinical outcomes, and may require additional therapy. Due to overlapping cytologic features and heterogeneity of lesions, the PTC classification is not adhered to in conventional reporting practice. This study aimed to classify the PTC cytology cases into a particular cytological variant by applying an algorithmic approach and correlating these variants with histology. MATERIALS AND METHODS: An analysis of all histopathologically confirmed cases of PTC who had previously been diagnosed with fine needle aspiration cytology (FNAC) from January 2014 to December 2019 was conducted. FNAC samples of thyroid nodules were blindly reviewed and classified into different morphological variants using a stepwise algorithmic approach based on architectural, nuclear, and cytoplasmic features. RESULTS: A review of 77 histologically proven cases of PTC variants or with a predominant area of variant histomorphology was done. One case was inadequate (TBSRTC I), nine cases were benign (TBSRTC II), two were follicular lesions of undetermined significance (TBSRTC III), and 65 cases were suspicious or definite for PTC (TBSRTC V/VI). Retrospective algorithmic cytopathological analysis of 65 cases that are suspicious or definite of PTC (TBSRTC V/VI) showed classical PTC (5), follicular variant-PTC (35), tall cell variant (20), diffuse sclerosing variant (1), warthin-like variant (2), and solid variant (2). Diagnostic accuracy of cytopathology in diagnosing the PTC variants when compared with histopathological diagnosis varied from 81.3% to 100% (mean 78.9%). Cluster analysis justified that our classification showed good agreement with the actual classification based on the cytopathological features identified by the cluster analysis. CONCLUSION: An awareness of cytomorphological features of aggressive variants may facilitate early and accurate diagnosis and appropriate clinical management with better patient outcomes. FNAC can subclassify PTC into different variants based on this algorithmic approach or aggressive and nonaggressive variants based on certain more frequently observed features.