US Diagnosis for Thyroid Nodules with an Indeterminate Cytology

ABSTRACT

PURPOSE: We wanted to assess the diagnostic efficacy of thyroid ultrasound (US) for evaluating thyroid nodules with indeterminate cytology. MATERIALS AND METHODS: Among 1865 nodules in 1278 patients who received a prospective US diagnosis of their thyroid nodule(s) and who subsequently underwent US-guided fine-needle aspiration, 130 nodules with indeterminate cytology were enrolled in the study. Each thyroid nodule was prospectively classified by a single radiologist into 1 of 5 diagnostic categories "benign", "probably benign", "indeterminate", "suspicious for malignancy" and "malignant." The solid nodules were classified using all 5 categories and the partially cystic nodules classified using 4 categories ("indeterminate" was omitted). We calculated the diagnostic efficacy of thyroid US by comparing the US diagnoses with the pathology results. RESULTS: Of 130 nodules with indeterminate cytology (130/1865, 7.0%), 62 nodules were surgically removed. Nineteen nodules were assigned to the indeterminate category on US. The malignantly rate of the US-indeterminate category was 56.5% (35/62). The sensitivity, specificity and positive and negative predictive values were 81.0%, 81.8%, 81.0%, 81.8% and 81.4%, respectively, when US-indeterminate nodules were excluded. There was no significant difference of diagnostic efficacy when these nodules were reclassified as malignant, but there was a significant difference of diagnostic efficacy when these nodules were reclassified as benign. CONCLUSION: Our US classification may be a feasible method for managing thyroid nodules with indeterminate cytology.