Objective: Conventional fine-needle aspiration biopsy (FNAB) for cystic thyroid nodules (CTNs) has a high rate of nondiagnostic and false-negative results. Ultrasound-guided FNAB (UG-FNAB) permits direct sampling of the wall and/or the solid portion of CTNs, increasing the possibility of a representative sample. In this study we evaluated the role of UG-FNAB in CTNs management. Methods: Five-hundred-seventy-five UG-FNAB of CTNs were performed. Thyroidectomy was carried out in 119 of these cases. The medical records of these 119 patients were reviewed and form the basis of this report. Results: The nondiagnostic smear rate was 9.2%. Cytological diagnosis was benign nodule in 42 cases, predominantly follicular lesion in 50 cases, and suspicious or malignant lesion in 16 cases. The final pathology revealed a benign nodule in 98 cases (82.4%) and a carcinoma in 21 (17.6%). The overall accuracy of UG-FNAB was 88.0%. No significant differences were found in age, sex, lesion size, or echographic pattern (p = NS) comparing patients with benign CTNs to patients with malignant CTNs. Conclusion: UG-FNAB has a low rate of nondiagnostic smears and a high overall accuracy in CTNs. All CTNs should undergo UG-FNAB to select patients for surgery, since the malignancy rate is not negligible and no clinical parameter can reliably predict it.