Diagnostic Efficacy of Ultrasonographic Characteristics of Thyroid Carcinoma in Predicting Cervical Lymph Node Metastasis.

The goals of this study were to determine the ultrasonographic characteristics of thyroid carcinoma (TC) and to explore the diagnostic efficacy of these ultrasonographic characteristics in predicting cervical lymph node metastasis (LNM). From June 2012 to June 2014, a total of 186 TC patients were recruited from the Central Hospital of Chengde City, Hebei, China. We divided them into two groups: the metastatic group comprised 129 nodules (n = 86), and the non-metastatic group 117 nodules (n = 100). Univariate and multivariate analyses were used to evaluate the relationship between ultrasonographic characteristics and cervical LNM. Spectral Doppler ultrasound was employed to estimate peak systolic velocity, pulsatility index and resistive index. Receiver operating characteristic curves were drawn to evaluate the efficacy of ultrasonographic characteristics in predicting cervical LNM. The sensitivity, specificity, positive predictive value and negative predictive value of ultrasonographic diagnosis were 81.40% (105/129), 92.32% (108/117), 92.11% (105/114) and 81.82% (108/132), respectively. Cervical LNM in TC frequently occurred at the cervical level VI (37.98%) and was located mainly in the middle pole (46.51%) or lower pole (41.09%). Peak systolic velocity and resistive index values were significantly higher in the metastatic group than in the non-metastatic group (both p < 0.001). Multivariate analysis revealed that nodular diameter, capsular invasion, microcalcification and flow grade were risk factors for TC patients with cervical LNMs (all p < 0.05). Furthermore, receiver operating characteristic curve analysis revealed that nodular diameter, capsular invasion, microcalcification and flow grade had excellent accuracy in predicting cervical LNM. We conclude that ultrasonographic characteristics of TC, including maximum nodular diameter, capsular invasion, microcalcification and flow grade, may predict cervical LNM.