Tirads classification and final diagnosis of thyroid nodules classified as bethesda-3 after fine-needle aspiration

Alhambra Exposito Maria Rosa; Martin Ana Barrera; Azpilcueta Pedro Segui; Galvez Moreno Maria Angeles

doi:10.1530/endoabs.63.GP93

Introduction: Ultrasound is the main diagnostic tool for imaging pathology of the thyroid gland. The TI-RADS classification allows the evaluation of a risk of malignancy in the case of thyroid nodules, indicating the need to perform a puncture for a histological study. Bethesda system classifies them according to the cytological malignancy risk. Bethesda category 3 (B3) comprises ‘follicular lesion of undetermined significance’ or ‘atypia of undetermined significance’.

Aim: Determine final diagnosis of B3 nodules and study associations between malignancy and other variables, as ACR TIRADS classification.

Methods: Retrospective study of thyroid nodules classified as B3 after FNA referred to our hospital between 2012 and 2018. Statistical analysis: SPSS v.22.0 (Student’s t-test to compare means and Chi-square/Fisher’s test for proportions).

Results: Two hundred two patients (80.7% female); mead (SD) age, 53.5 (13.7) years. Sonographic features of included nodules are shown in table 1. Of the nodules, 7.9% were TIRADS 2 (maximum mean diameter (DMM) 39.86 mm), 14.7% TIRADS 3 (DMM 34.32 mm), 56.5% TIRADS 4 (DMM 29.24 mm), 20.9% TIRADS 5 (DMM 25.66 mm). Eigteen (8.9%) lesions met the reference-standard criteria for malignancy: 13 papillary thyroid cancers, 4 follicular thyroid cancer, 1 thyroid metastases from other malignancies. Application of the ACR TIRADS system’s FNA criteria would have reduced the number of biopsies performed by 19.2%. Of the nodules without indication of FNA, only one was malignant; of those who had an indication for FNA, 16.5% were malignant. In our cohort, the sensitivity of this system was 94.4% and the specificity was 19.6%. The positive predictive value (PPV) of the test was 94% and the negative predictive value (NPV) 19.6%. A false negative rate of 5.6%. The area under the ROC curve was 0.735 (0.588–0.881, P 0.001).

Table 1 Sonographic features of included nodules.
Structure/composition		Echogenicity		Shape
	%		%		%
Cystic or almost completely cystic	1.7%	Anechoic	0%	Wider tan tall	87.6%
Spongiform	0.6%	Hyperechoic o isoechoic	27.2%	Taller tan wide	12.4%
Mixed	9%	Hypoechoic	56.4%
Solid	88.7%	Very hypoechoic	4%

Table 1 Continued.
Margin		Echogenic foci
	%		%
Smooth or defined	64.9%	None or large comet-tal artifacts	77.4%
Lobulated or irregular	21.8%	Macrocalcifications.	12.4%
Extra-thyroidal extension	1%	Periferical calcifications	2.5%
		Punctate echogenic foci	6.4%

Conclusions: Internationally-endorsed sonographic risk stratification systems vary widely in their ability to reduce the number of unnecessary thyroid nodule FNAs. The ACR TIRADS outperformed the others, classifying over half the biopsies as unnecessary with a FNR of 5.6%.

Endocrine Abstracts

GP93