The authors analyse the predictive diagnostic accuracy of fine-needle aspiration cytology (FNAC) and frozen section examination in adult patients operated for thyroid nodules.

The same pathologist performed macroscopic and cytological examination, followed by frozen section examination on each operative specimen. FNAC results were classified into three groups: benign, malignant or suspicious of malignancy. Frozen section examination was also classified into three categories: benign, malignant or suspicious of malignancy when not all criteria of malignancy were present.

One hundred and sixty-six (82%) of the 202 patients included in the study were females. Patients had a mean age of 51years. Thyroid carcinoma was diagnosed on final pathology in 22% of women and 25% of men. FNAC results were benign in 85% of cases, malignant in 9% of cases and atypical or suspicious in 6% of cases, with a specificity of more than 99% and a sensitivity, including and excluding microcarcinomas, of 36% and 48%, respectively. The diagnostic accuracy of FNAC was 84% and 89%, after excluding micro-carcinomas. Frozen section was benign in 85% of cases, malignant in 13% of cases and suspicious in 2% of cases, with a specificity of more than 99% and a sensitivity, including and excluding microcarcinomas, of 56% and 68%, respectively. The diagnostic accuracy of frozen section was 89% and 90%, after excluding microcarcinomas. The diagnostic accuracy of the combination of the two examinations was 94% after excluding microcarcinomas.

FNAC and frozen section have a comparable predictive diagnostic accuracy. Frozen section is requested by the surgeon not only on the basis of preoperative FNAC, especially when it is suspicious, or even indeterminate, but also in the light of the macroscopic surgical findings.