Fine-needle aspiration biopsy (FNAB) of thyroid nodules is an operator-interpreter-dependent procedure whose diagnostic accuracy has deficiencies, especially concerning indeterminate and negative results. There are reports in the literature of up to a 15% definitive diagnosis of carcinoma in the first category of cytology and an 11% false-negative rate in the second category. With these facts in mind, a prospective study was done at the Instituto Nacional de Cancerología, Mexico City, to determine whether one or several clinicoradiologic factors were associated with a final pathologic diagnosis of carcinoma, thereby increasing the diagnostic accuracy of thyroid FNAB. In total, 200 cases were reviewed, and 159 patients underwent surgery. Our study group comprises these 159 patients. The only clinical factor associated with false-negative results was size > or = 4 cm, with no statistical significance for the other factors analyzed. A comparison was done between clinical factors of patients who had indeterminate results in the aspirate and those of patients who had a definitive diagnosis by permanent section of the thyroid carcinoma, where an association was found between malignancy and tumor size > or = 4 cm, irregular borders, fixed lesions, and heterogeneity on ultrasound. We conclude that in the presence of a negative cytology result, with tumor size > or = 4 cm, close follow-up is indicated, and a new thyroid FNAB should be performed in 6 months. If the nodule persists or grows, a surgical procedure should be considered. In the case of an indeterminate cytology result and in the absence of those factors associated more frequently with a pathologic result of carcinoma, a conservative approach could be indicated with clinical follow-up and repeat thyroid FNAB in 6 months to 1 year.