Interobserver agreement and intraobserver reproducibility in thyroid ultrasound elastography

The objective of this study was to evaluate interobserver agreement and intraobserver reproducibility in thyroid ultrasound elastography.

Fifty-six patients (56 nodules) were enrolled for this prospective study between December 2010 and March 2011. All patients were scheduled for thyroid surgery because their nodules had been found by fine-needle aspiration biopsies to be malignant, suspicious for malignancy, or follicular/Hürthle cell neoplasm. On each patient, three endocrinologists performed ultrasound elastography examinations using a commercial ultrasound machine. No external compression with the transducer was applied because pulsation from the carotid artery was used. Each endocrinologist repeated an elastography examination on the same thyroid nodule. An elasticity contrast index, which quantifies the local strain contrast within a nodule, was interactively determined after a nodule's boundary was delineated by the endocrinologist. A large elasticity contrast index value indicates an increased probability of the nodule being malignant. Pearson correlation coefficient was used to evaluate the interobserver and intraobserver agreement in the measured elasticity contrast index values, and p values < 0.05 were considered statistically significant.

Significant interobserver and intraobserver agreement was found in thyroid ultrasound elastography. Pearson correlation coefficients between two observers were 0.79 (observers 1 and 2, p < 0.01), 0.77 (observers 1 and 3, p < 0.01), and 0.73 (observers 2 and 3, p < 0.01). Pearson correlation coefficients for intraobserver agreement were 0.87 (p < 0.01), 0.73 (p < 0.01), and 0.84 (p < 0.01) for observers 1, 2 and 3, respectively.

Good interobserver and intraobserver agreement exists in thyroid ultrasound elastography. Compared with external compression elastography, we found that thyroid elastography with intrinsic compression can produce reliable results.