Reply to diagnostic criteria and risk‐adapted approach to indeterminate thyroid cytodiagnosis

Abstract

We thank Dr. Yang for his interest in our article1 and appreciate his comments. We fully agree that close communication between the clinician and cytopathologist is essential. Information on the clinical characteristics of the patient as well as the ultrasonographic characteristics of the nodule have the potential to profoundly influence both interpretation of the cytology and clinical management decisions. We further agree that improved selection of which nodules to biopsy will decrease the rate of indeterminate cytodiagnosis. However, all of the published guidelines, including the American Thyroid Association Guidelines2 revised in 2009, recommend biopsy in most nodules larger than 1-1.5 cm. Dr. Yang correctly points out that having a high rate of histological correlation of biopsy samples would be ideal. Indeed, his excellent article published in this journal in 20073 provided important information on the diagnostic accuracy of fine-needle aspiration biopsy of thyroid lesions. In his study of 4703 biopsies, histological correlation was obtained in 1052 patients. The rate of indeterminate thyroid cytodiagnosis varies with the reference cited. We cited 15% as a commonly accepted national average. Dr. Yang points out that 1 of our references indicated the rate to be as high as 22%. It is of interest that in Dr. Yang's series, the rate was only 11.6%. Further, in his patients with initial diagnosis of follicular neoplasm who underwent repeat biopsy, 40% were reclassified as benign. Our contention that the rate of diagnosis of follicular neoplasm is excessive is, in large part, based on our role to provide second opinions on biopsies previously categorized as follicular neoplasm. We consistently find that in a large number of these cases, the nodules are reclassified as benign, either on the basis of a review of the original biopsy sample or on the basis of a high-quality repeat biopsy when the initial sample was thought to be insufficient or of poor quality. Finally, Dr. Yang suggests that our preference for the term microfollicular neoplasm does not add significant clarification. We agree that the follicular neoplasm category in the Bethesda System includes “follicular cells...arranged predominately in microfollicular or trabecular arrangements.”4 However, we have frequently found that among cases we have reviewed for a second opinion, samples with orderly monolayer sheets were often misinterpreted as follicular neoplasm simply on the basis of hypercellularity. In his article,3 Dr. Yang correctly states that these should be classified as benign, and we agree completely. We believe that the description microfollicular neoplasm adds clarity to this category by emphasizing the need for architectural abnormality for this diagnosis. Those patients whose nodules demonstrate simple hypercellularity should not be misclassified as having a follicular neoplasm and placed at risk of unnecessary surgery. No specific funding was disclosed. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures.