Variability in Stiffness Assessment in a Thyroid Nodule Using Shear Wave Imaging

The aim of the study was to assess intrareader and interreader variability in placing a region of interest in the elastography color map in the thyroid nodule.

After institutional review board approval, elastograms were obtained by shear wave elastography from 38 individuals, with a total of 60 thyroid nodules. Two readers reviewed all images while blinded to the cytopathology or final histopathology results and to each other's measurements. For each image, readers first performed a qualitative assessment and recorded presence of stiff region(s) and number of stiff regions. Then, they placed a 3-mm circular region of interest on the stiffest region to quantify the mean and maximum stiffness in kilopascals. Intrareader and interreader agreement was summarized for categorical variables using Cohen κ and for continuous variables using the concordance correlation coefficient (CCC) and coefficient of variation.

There was good overall intrareader and interreader agreement for the presence of stiff regions and number of stiff regions after classifying them as 0, 1, and greater than 1 regions (κ = 0.78-0.81 and 0.74-0.79, respectively). There was also good overall agreement for the the following quantitative measurements: mean kPa (CCC = 0.97 and 0.93) and max kPa (CCC = 0.97 and 0.93). The intrareader and interreader coefficient of variation was 13% and 21% for mean kPa and 14% and 21% for max kPa.

There can be differences in measurement of stiffness by 2 different operators in softer nodules with shear wave imaging because of the lack of brighter areas on the color elastograms. A standardized technique is necessary to be able to compare results from 1 study to the next.