Diagnostic performance of gray-scale US and elastography in solid thyroid nodules

Abstract

Purpose: To evaluate the diagnostic performance of gray-scale ultrasonography (US) and elastography in differentiating benign and malignant thyroid nodules.

Materials and methods: This was an institutional review board-approved retrospective study with waiver of informed consent. A total of 703 solid thyroid nodules in 676 patients (mean age, 49.7 years; range, 18-79 years) were included; there were 556 women (mean age, 49.5 years; range, 20-74 years) and 120 men (mean age, 50.7 years; range, 18-79 years). Nodules with marked hypoechogenicity, poorly defined margins, microcalcifications, and a taller-than-wide shape were classified as suspicious at grayscale US. Findings at elastography were classified according to the Rago criteria and the Asteria criteria. The diagnostic performances of gray-scale US and elastography were compared. For comparison between the diagnostic performances of gray-scale US and the combination of gray-scale US and elastography, three sets of criteria were assigned: criteria set 1, nodules with any suspicious grayscale US feature were assessed as suspicious; criteria set 2, Rago criteria were added as suspicious features to criteria set 1; and criteria set 3, Asteria criteria were added as suspicious features to criteria set 1. The diagnostic performances of gray-scale US, elastography with Rago criteria, and elastography with Asteria criteria, and odds ratios (ORs) with 95% confidence intervals for predicting thyroid malignancy were compared using generalized estimating equation analysis.

Results: Of 703 nodules, 217 were malignant and 486 were benign. Sensitivity, negative predictive value (NPV), and OR of gray-scale US for the 703 nodules were 91.7%, 94.7%, and 22.1, respectively, and these values were higher than the 15.7% and 65.4% sensitivity, 71.7% and 79.1% NPV, and 3.7 and 2.6 ORs found for elastography with Rago and Asteria criteria, respectively. Specificity, positive predictive value, and accuracy for criteria set 1 were significantly higher than those for criteria sets 2 and 3 for most of the nodule subgroups that were considered.

Conclusion: Elastography alone, as well as the combination of elastography and gray-scale US, showed inferior performance in the differentiation of malignant and benign thyroid nodules compared with gray-scale US features; elastography was not a useful tool in recommending fine-needle aspiration biopsy.