Back to Search Start Over

Sonographic Criteria Predictive of Malignant Thyroid Nodules: Which Lesions Should be Biopsied?

Authors :
Carlos Miguel, Oliveira
Rui Alves, Costa
Miguel, Patrício
Amélia, Estêvão
Bruno, Graça
Filipe, Caseiro-Alves
Source :
Academic radiology. 25(2)
Publication Year :
2017

Abstract

The objective of this study was to evaluate the ultrasound features of thyroid nodules and their association with malignancy, focusing on establishing feature-oriented ultrasound criteria to determine proper management of a thyroid nodule.A sample of 379 thyroid nodules were biopsied (from a total of 357 patients aged 59.8 ± 14.8 years) and 300 were included in the final study (271 benign nodules and 29 malignant ones). Ultrasound features were recorded for each nodule: size, echogenicity, homogeneity, contours, shape, texture, peripheral halo, calcifications, and the presence of adenopathy. Statistical analysis of the data was performed using the Mann-Whitney U test and chi-square test. The sensitivity and the specificity of variables seen to have a statistically significant association with the malignancy of nodules were assessed and a logistic regression was performed.A taller-than-wide shape, an ill-defined contour, the presence of a halo, microcalcifications, and adenopathy were found to have a statistically significant relationship (P 0.05) with malignancy, although with a low sensitivity and a high specificity. The presence of at least one suspicious feature yields great sensitivity (89.7%) in detecting malignant disease.The ultrasound features of thyroid nodules alone do not allow the radiologist to make a confident diagnosis regarding the malignancy of a nodule without performing a biopsy. However, a nodule showing a taller-than-wide shape, microcalcifications, a peripheral halo, an ill-defined contour, or associated adenopathy should be considered for cytology.

Details

ISSN :
18784046
Volume :
25
Issue :
2
Database :
OpenAIRE
Journal :
Academic radiology
Accession number :
edsair.pmid..........5d75f3ac51779b4f4760a9456df1fd13