The rates of sensitivity and specificity of fine needle aspiration biopsy (FNAB) for the diagnosis of thyroid malignancy differ considerably among various reported series. These values are influenced by three factors: (a) whether only clearly positive and negative results are considered, or whether the commonly encountered 10–20% of indeterminate/suspicious ones are included; (b) whether adenomas are considered as neoplasms in one group with carcinomas; and (c) whether only histologically proven cases are used in calculations or whether patients with benign clinical follow-up are included.
The aim of the study was to evaluate the sensitivity and specificity of FNABs performed at this institution in the last 7 years from the clinical point of view, considering only benign vs. suspicious/malignant FNAB results (indicating surgery), and benign (including adenomas) vs. malignant definitive histology.
Retrospective study comparing pre-operative FNAB results with definitive histological examination after operation.
A total of 2492 FNABs were performed in 2100 patients (1875 women and 225 men); their ages ranged from 9 to 85 years, with a median of 46 years. Clinical diagnosis was multinodular goitre in 1330, single nodule in 591, Hashimoto's thyroiditis in 147 and subacute thyroiditis in 32 cases. In 148 instances, the nodule was cystic. A history of previous treatment for carcinoma of the thyroid was present in 12 patients. Five hundred and thirty-six patients subsequently underwent thyroid surgery.
The values of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and diagnostic accuracy were calculated.
The sensitivity was 86%, specificity 74%, PPV 34%, NPV 97% and diagnostic accuracy 75%.
The specificity and positive predictive value are low when fine needle aspiration biopsy results are divided into two categories only (these being indication for surgery or not), and when only suspicious/malignant fine needle aspiration biopsies with subsequent malignant histology are considered to be true positive. Nevertheless, the ability to discriminate 11.7% of patients with a 34% probability of malignancy (suspicious/malignant cytology) from 81.2% of patients (benign cytology) with a probability of only 3% is very helpful.