To assess the value of conventional sonographic characteristics of thyroid nodules combined with internal coarse calcification in differentiating benign and malignant thyroid nodules.

The ultrasonographic data of 217 patients with 229 thyroid nodules confirmed by pathology at Changzhou First People's Hospital from January 2018 to July 2020 were retrospectively analyzed. Ultrasonography of the included nodules showed that there were coarse calcifications in the nodules. According to the gold standard of pathological results, the nodules were divided into either a benign (n=110) or a malignant (n=119) group. The conventional sonographic characteristics of nodules including maximum diameter, anteroposterior/transverse diameter ratio, echo intensity, growth pattern, composition, margin, and color Doppler flow imaging, and sonographic characteristics of coarse calcification including thickness, calcification interruption, morphology, margin of soft tissue outside the calcification, echo consistency, and location were analyzed retrospectively. Univariate analysis was used to compare the conventional sonographic characteristics of nodules and coarse calcification in nodules between the two groups. Binary logistic regression was used to establish prediction models of conventional nodule characteristics (model 1) and the comprehensive characteristics of nodules and coarse calcification (model 2). The diagnostic performance of each parameter and the combined model was compared by receiver operating characteristic curve (ROC) analysis and Z-test.

The maximum diameter of thyroid nodules ≤1.25 cm, anteroposterior/transverse diameter ratio >0.78, being solid, vertical growth, irregular margin or extra-thyroidal extension, as well as irregular thickness, interruption, echo inconsistency, soft tissue edge ,and central type of coarse calcification were significantly more common in the malignant group than in the benign group (P＜0.05). Model 1 included the maximum diameter, anteroposterior/transverse diameter ratio, and margin of the nodule. In model 2, the maximum diameter of nodule, anteroposterior/transverse diameter ratio, and calcification thickness, interruption, and location were included. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of model 2 were 81.85%, 82.73%, 82.29%, 80.18%, and 81.22%, respectively. The area under the curve was 0.877, which was significantly higher than that of model 1 (0.753; Z=4.197, P＜0.001).

Sonographic characteristics such as irregular thickness, interruption, and central calcification in nodules are helpful for the diagnosis of malignant nodules. While observing the sonographic characteristics of thyroid nodules, it is important to evaluate the characteristics of internal coarse calcification in order to better distinguish benign and malignant thyroid nodules.