Fine-needle aspiration biopsy (FNAB) has been widely accepted as the most accurate, safe, and cost-effective method for evaluation of thyroid nodules. The most challenging category in FNAB is atypia of undetermined significance (AUS) and follicular lesion of undetermined significance (FLUS). The Bethesda system (BS) recommends repeat FNAB in that category due to its low risk of malignancy. In our study, we aimed to investigate the malignancy rate of thyroid nodules of AUS and FLUS and whether there were different malignancy rates among the different patterns in this category, and to evaluate the presence of biochemical, clinical, and echographic features possibly predictive of malignancy related to AUS and FLUS. Data of 268 patients operated for AUS and FLUS cytology were screened retrospectively. Ultrasonographic features and thyroid function tests, thyroid antibodies, scintigraphy, and histopathological results were evaluated. Of the 268 patients' results, 276 nodules are evaluated. Malignancy rates were 24.3 % in the AUS group, 19.8 % in the FLUS group, and 22.8 % in both groups. In the evaluation of all nodules, the predictive features of malignancy are hypoechogenicity and peripheral vascularization of the nodule. We determined that the malignancy rates in these nodules are higher than that in the literature rate. This high ratio may be due to the fact that we studied only patients who underwent surgery. The ultrasonographic features alone may be insufficient to predict the malignancy; therefore, all the clinical and ultrasonographic features must be considered in the evaluation of the thyroid nodules. In addition, we think that the recommended management of repeat FNAB in these groups must be reconsidered with the clinical and ultrasonographic features.