Endocrine Abstracts

Introduction: Knowing that only in 5–15% of FNABs the pathological analysis shows malignancy, enhance diagnostic accuracy is needed. The use of the TIRADS score and a revised version proposed by the British Thyroid Association (BTA) intend to determine if a FNAB should be performed or not in the evaluation of nodular goitre.

Objective: To determine whether an adequate correlation is present between the risk category assigned to the nodules studied under TIRADS/BTA system and cytological results.

Material and methods: To assess the potential of the TIRADS/BTA systems and the degree of concordance with the results derived from FNAB, we designed a retrospective, observational study including 230 patients who underwent thyroid ultrasonography and FNAB. Nodules were classified following the BTA score as benign (U2), probably benign (U3), and suspicious/probably malignant (U4+5). FNAB samples were informed according to the Bethesda classification. Statistical analysis was assessed using the SPSS Statistical Suite (ver 21.0).

Results: Total sample n=230 (valid pairs, n=204); 71.3% females, mean age 59.2±5.6 years. U2 nodules were 76%; in one case (0.7%) the pathological analysis was consistent with malignancy/suspicious for malignancy while the rest of them (99.3%) were informed as benign. U3 nodules comprised 18 nodules, from which 55% had a benign pathological diagnosis, 16.66% probably benign, and 27.77% malignant or suspicious for malignancy. The high-risk group (U4+5) included n=31 (15.1%) nodules: FNAB results showed a malignant/suspicious for malignancy diagnosis in 80.6%, in 12.9% benign, and in 6.4% probably benign. TIRADS score sensitivity was 75% (95% CI: 57.79–87.85%); specificity was 97.62% (95% CI: 94.01–99.33%), with a positive predictive value of 87.1% and a negative predictive value of 94.8%. Cohen’s κ coefficient was 0.69 (P<0.001).

Conclusions: The TIRADS/BTA risk scores have proven to be a useful tool to assess thyroid nodules and the need for FNAB, enhancing diagnostic accuracy and resource allocation.