Diagnostic accuracy of conventional versus sonography-guided fine-needle aspiration biopsy of thyroid nodules

Abstract

Fine-needle aspiration biopsy (FNAB) is an accurate, slightly invasive, and safe method for the preoperative diagnosis of thyroid nodules. Recently, ultrasound guidance has been suggested as a valuable aid to enhance FNAB diagnostic performance. In this study, we have compared diagnostic accuracy of conventional FNAB (C-FNAB) versus sonography-guided FNAB (SG-FNAB) on a large sample population of 9683 patients with thyroid nodules. Over a 15-year period, 4986 patients were investigated by C-FNAB and 4697 underwent SG-FNAB. A valid cytological diagnosis was obtained in 85.9% of C-FNAB and in 91.5% of SG-FNAB cases, allowing detection of thyroid cancer in 1.6% and 2.1% of patients, respectively. The indeterminate pattern of follicular neoplasia was observed in 238 C-FNAB (5%) and in 272 (5.4%) SG-FNAB nodules. Specimens were cytologically inadequate in 433 C-FNAB (8.7%), but only in 167 SG-FNAB cases (3.5%). A total of 535 C-FNAB and 540 SG-FNAB nodules underwent surgery. False-negative results occurred in 7 C-FNAB nodules (2.3%), but only in 3 SG-FNAB cases (1%). Sensitivity, specificity, and global diagnostic accuracy of C-FNAB compared with SG-FNAB were 91.8% versus 97.1%, 68.8% versus 70.9%, and 72.6% versus 75.9%, respectively. Our results, based on a large population of thyroid nodules, demonstrate that SG-FNAB allows a more precise and adequate sampling of thyroid nodular lesions and is associated with a lower rate of false-negatives, thus improving global diagnostic accuracy in the preoperative selection of thyroid cancer.