Background: The clinical management of repeatedly non-diagnostic thyroid nodules (RNDNs) via fine needle aspiration cytology (FNAc) is a matter of debate because current recommendations and clinical practice are not based on high-quality evidence. Our purpose was to characterize RNDNs and evaluate their clinical management in our centers.
Methods: This retrospective observational study included 319 consecutive patients who underwent ultrasound (US-)guided FNAc in two Italian academic hospitals between 2016 and 2020 and had previous cytology non-diagnostic result (TIR1). Clinical management and anamnestic data were retrieved, and the cytological specimens and US exams were double-blindly reviewed by two pathologists and endocrinologists.
Results: The risk of RNDNs was significantly greater in hypoechogenic nodules (Odds Ratio [OR]=1.727, 95% confidence Interval [CI]: 1.090-2.735, P=0.02) and lower in nodules that had been recognized less than 10 years before (OR=0.349, 95% CI: 0.153-0.796, P=0.01). Clinicians chose to directly perform surgery on multinodular, intermediate-risk nodules (as per AACE/AME guidelines and EUTIRADS class 4), while larger (P<0.0001) and uninodular (P=0.03) lesions were further investigated with a third FNAc. Only 16 RNDNs were sent to surgery. Twelve nodules turned out to be benign goiters with a high rate of fibrosis, while only 3 were definitively malignant. However, retrospectively, all the malignant ones exhibited higher-risk ultrasound features and had an undetermined result (TIR3B) at the third cytological evaluation.
Conclusions: In a real-life context, RNDNs exhibited a very low rate of malignancy and were mostly long-known goiters with regressive changes, as suggested by a hypoechoic pattern. Consequently, a clinical-US surveillance approach could be cautiously hypothesized, while greater attention could be given to larger and higher-US-risk (both as EUTIRADS 4-5 and AACE/AME intermediate- and high-risk classes) nodules.