Elastography of the thyroid

Abstract

Thyroid nodules are very common, while thyroid cancer is rare and has a very good prognosis. Thyroid nodule ultrasound characterization performed by experienced clinicians allows the selection of the tumours to be punctured and guiding fine needle aspiration (FNA). FNA provide cytology information able to differentiate benign tumours from cancer in approximately 80% of cases. However, it remains difficult to identify thyroid cancers with ultrasound imaging, as demonstrated by the very low rate of cancers detected in all of the carried out FNA (approximately 5%). As a majority of thyroid cancers are hard, the stiffness evaluation has become part of nodular characterization. Since 2005, elastography has been used for the evaluation of thyroid nodules; quasi-static elastography was the first technique available and used, at first, an external pressure induced by the probe, which was then replaced by carotid internal excitation allowing improvement in sensitivity. Semi-quantitative analysis allows comparison of tissue elasticities between tissue with elasticity anomalies and normal tissue and provides therefore useful analytic information. Shear wave elastography (SWE) provides a map of the elasticity in a region and allows stiffness quantification of lesions in kilopascals in order to reinforce the predictive value of malignancy. A tumour whose stiffness is greater than 65kPa or for which the stiffness ratio is greater than 3.7 compared to surrounding healthy tissue is highly suspicious. SWE may enable the detection of malignant follicular tumours that currently escape detection by the ultrasound-guided ultrasound/aspiration cytology couple. Lymph node metastasis of papillary thyroid cancer can also be detected by elastography due to its increased stiffness.

Ultrasound elastography in thyroid disease

“…Excellent interobserver agreement was also reported using a three score gray scale system [55]. Although most studies state that using the quality indicator while performing RTE aims to reduce variability, little is known about the actual figures for intraobserver variability of scoring, although at least one study reported acceptable intraobserver reproducibility [61].…”

“…As RTE displays relative elasticity within a ROI, it is highly advisable to have a ROI as large as possible and include in it as much "normal" reference thyroid tissue as available. If the whole ROI content is abnormal, the results are extremely unreliable [61]. On the other hand, one should avoid the inclusion of vessels, bones and, if possible, nonthyroid tissue.…”

“…As discussed above, larger than 3cm or lobar size nodules cannot be assessed with RTE [51]. Coalescent nodules are, also, unsuitable for elastography [24,38,61].…”

“…The nodules located in front of the common carotid artery are the most susceptible to pulsation interferences [51]. Nodules within residual parenchyma lack reference tissue for comparison [61].…”

Thyroid Ultrasound: State of the Art. Part 2 – Focal Thyroid Lesions

“…Different elastographic techniques were used by these authors and they found that all methods has both good sensitivity and specificity for predicting malignancy in thyroid nodules (fig 4) [12][13][14][15][16]. Also, inter-and intra-operator reproducibility of 'quasi-static' elastography is considered acceptable [17,18]. Thyroid nodules with VTQ values <2.16 m/s were all benign, which accounted for 60.2% of all benign nodules [19].…”