ABSTRACT
OBJECTIVES: To analyze the diagnostic accuracy of ultrasound-guided core-needle biopsy (CNB) of thyroid nodules. METHODS: Of 3517 CNBs performed using an 18G spring-loaded device in one institution, we retrospectively reviewed 676 nodules in 629 consecutive patients who underwent surgery. CNB and pathological examination were compared. CNB diagnosis was standardized in four categories: insufficient (I), benign (B), follicular lesion (FOL), and malignant (M). Main outcome measures were predictive positive values (PPV), false positives (FP), and false negatives (FN). RESULTS: CNB showed a low rate of insufficient and FOL diagnoses (5.8 % and 4.5 %). On surgery, there were eight FNs in 374 benign CNBs and three FPs in 148 malignant CNBs. The 154 nodules classified as FOL in CNB included, at surgery, 122 neoplasms; 28 of them malignant. PPV for malignancy of a malignant CNB was 98 %, and for a CNB diagnosis of FOL 18.2 %. Sensitivity for malignancy if CNB of FOL and M are considered positive was 95.6. Only one major complication was observed. CONCLUSIONS: CNB is reliable, safe, and accurate to evaluate thyroid nodules and can be an alternative technique to FNA. It has low rate of non-diagnostic and undetermined cases, with high sensitivity and PPV. KEY POINTS: Thyroid core-needle biopsy (CNB) has high sensitivity and PPV. Pitfalls of CNB are rare. Pitfalls are due to cystic cancer, histological heterogeneity, and mistakes in analysis. CNB is a reliable, safe, and accurate method to approach thyroid nodules. CNB can be used primarily or after insufficient or indeterminate FNA.
