Clinical Study Of Sonography In Thyroid Imaging Reporting And Data System For Thyroid Nodule

Background and Purpose:The widespread use of ultrasongraphy (US) has contributed to the increased detection of thyroid nodules. Thyroid US depicts nodules in up to67%of the population. However, less than10%of these nodules are malignant.Several thyroid ultrasound (TUS) findings have been associated with an increased risk for thyroid cancer. However, there is no consensus as to the format and style for reporting the results of TUS. The objective of this study was to explore the diagnostic value of sonography in thyroid imaging reporting and data system (TI-RADS) for thyroid nodules and to develop a practical TI-RADS with which to categorical thyroid nodules and stratify their malignant risk.Materials and Methods:â‘ What features were predicted a malignant classification by univariate and logistic regression analysis. The category1-5of353category of thyroid nodules were evaluated by using TI-RADS. All the nodules were confirmed by pathology. According to TI-RADS, a receiver operating characteristic (ROC) curve was drawn.â‘¢From January10,2011to January12,2012, surgery was performed in1132focal thyroid nodules in786consecutive patients. The study included the818category of thyroid nodules in653patients for which pathologic diagnosis or follow-up findings were available. Multinomial logistic regression was used to analyze the related factors and to evaluate odds ratio(OR).â‘¢We retrospectively reviewed213categories of nodules(158patients)confirmed by pathology. Univariate and multivariate analyses with generalized estimating equations were performed to investigate the relationship between suspicious US features and thyroid cancer. A score for each significant factor was assigned and multiplied by the coefficient obtained for each significant factor from multivariate logistic regression analysis. Scores for each significant factor were then added, resulting in an equation that fitted the probability of malignancy in thyroid nodules. The authors evaluated the fitted probability by using a regression equation; the risk of malignancy was determined according to the number of suspicious US features.Results:â‘ The following US features showed a significant association with malignancy: irregular margins, taller-than-wide shape,, posterior echo attenuation, and the lymph node with long axis/short axis<2, High-echo, cystic change, microcalcifications. According to TI-RADS, a receiver operating characteristic (ROC) curve was drawn. The area under the curve of TI-RADS were0.89.â‘¡Results of multinomial logistic regression showed that cystic component, mainly cystic of mixed echogenicity were related with TI-RADS2. Solid component, mainly solid of mixed echogenicity were related with TI-RADS3. The considerable overlap occurs was discovered among TI-RADS4(a,b) and TI-RADS5.â‘¢As the number of suspicious US feature increased, the fitted probability and risk of malignancy also increased. Positive predictive values according to the number of suspicious US features were significantly different(P<0.001).Conclusion:â‘ TI-RADS lexicon has an important guiding value for clinical diagnosis and treatment in ultrasound examination of thyroid nodule.â‘¡Multinomial logistic regression is more powerful in exploring the influence of Several US findings on TI-RADS lexicon, however, it can not different from TI-RADS4and TI-RADS5lexicon.â‘¢Risk stratification of thyroid malignancy by using the number of suspicious US features allows for a practical and convenient TI-RADS.