Can Diffusion-Weighted MR Imaging Be Used as a Tool to Predict Extrathyroidal Extension in Papillary Thyroid Carcinoma?

Shudong Hu, Heng Zhang, Xian Wang, Zongqiong Sun, Yuxi Ge, Jie Li, Weiqiang Dou
2020 Academic Radiology  
To evaluate whether quantitative diffusion-weighted MR imaging (DWI) with multi-b values can be used as a tool to predict the aggressiveness by using the histological feature of extrathyroidal extension (ETE) in papillary thyroid carcinoma (PTC). 238 patients were pathologically confirmed as PTCs underwent preoperative multi-b value DWI (300, 500, and 800 s/mm2) between January 2015 and December 2017. The patients were divided into three groups according to the degree of ETE: without ETE,
more » ... without ETE, minimal ETE, and extensive ETE. The apparent diffusion coefficients (ADCs) were evaluated for three different b values (b = 300 s/mm2, b = 500 s/mm2, and b = 800 s/mm2). The ADC values of the groups with and without ETE, minimal and extensive ETE were compared. The diagnostic relevance of the ADC values in terms of predicting ETE was compared using a receiver operating characteristic analysis. Differences between the areas under the curves (AUCs) were compared by using a Delong test. PTCs with ETE had significantly lower ADC300, ADC500 and ADC800 values than PTCs without ETE (p = 0.001, p < 0.001, and p < 0.001, respectively). The AUC of the mean ADC500 value (0.905) was higher than that of the ADC300 and ADC800 values (0.607 and 0.770, respectively) in differentiating ETE from without ETE (p < 0.001). The cut-off value of ADC500 to discriminate PTCs with and without ETE was determined at 1.407 × 10-3 mm2/s, with sensitivity of 80.7%, specificity of 86.7%, and an AUC of 0.905. The ADC value can be demonstrated an effective tool for evaluating the aggressiveness with the histological feature of ETE in PTC. In particular, ADC value at b = 500 s/mm2 showed the best performance for noninvasive preoperative evaluation of ETE.
doi:10.1016/j.acra.2020.03.005 pmid:32303443 fatcat:bcfp7eqvcngyfp5rp2xqnwevp4