Reliability of fine-needle aspiration for thyroid nodules greater than or equal to 4 cm

Background: Fine-needle aspiration (FNA) is considered the diagnostic test of choice in the evaluation of thyroid nodules. Some practice recommendations, however, suggest surgical resection of larger thyroid nodules due to concerns of FNA unreliability in the diagnosis of thyroid malignancy. The purpose of this study was to determine the reliability of FNA in thyroid nodules ≥4 cm. Methods: Retrospective review of prospectively collected data of 1068 consecutive patients who underwent FNA and thyroidectomy at a single tertiary medical center from 2003 to 2010 was performed. Patients were divided into two groups: those patients with a dominant thyroid nodule ≥4 cm (n = 212) and those patients with a dominant thyroid nodule <4 cm (n = 856). Sensitivity, specificity, and negative and positive predictive values were calculated for FNA results and final histopathology after thyroidectomy. Results: Of 212 patients with lesions ≥4 cm, 35% had thyroid malignancy on final pathology. Conversely, 54% of 856 patients with dominant thyroid nodules <4 cm had a final diagnosis of thyroid cancer after thyroidectomy. FNA demonstrated similar test characteristics among patients with lesions ≥4 cm and <4 cm, with a specificity of 99% (CI: 96%-100%) and 98% (CI: 96%-99.0%), respectively, and a sensitivity of 35% (CI: 23%-49%) and 42% (CI: 37%-46%), respectively. The positive predictive value of FNA was 82% (CI: 75%-100%) for nodules ≥4 cm and 96% (CI: 92%-98%) for nodules <4 cm. Negative predictive value was significantly different, with a value of 82% (CI: 75%-87%) for lesions ≥4 cm and only 59% (CI: 55%-63%) for lesions <4 cm. Conclusions: The reliability of FNA as a diagnostic test is not affected by the size of thyroid nodules. Routine surgical resection for all thyroid nodules ≥4 cm should not be used as the only independent factor in determining need for surgical resection.