Dana Stoian1, Bogdan Timar2, Mihnea Derban3, Stelian Pantea4, Florian Varcus4, Marius Craina1, Mihaela Craciunescu5. 1. Endocrine Unit, Department of Obstetrics Gynecology, "Victor Babes" University of Medicine and Pharmacy, Timisoara, Romania. 2. Department of Medical Informatics, "Victor Babes" University of Medicine and Pharmacy, Timisoara, Romania. 3. Department of Pathology, "Victor Babes" University of Medicine and Pharmacy, Timisoara, Romania. mihneaderban@gmail.com. 4. Department of Surgery, "Victor Babes" University of Medicine and Pharmacy, Timisoara, Romania. 5. Department of Microbiology, "Victor Babes" University of Medicine and Pharmacy, Timisoara, Romania.
Abstract
AIMS: Due to the elevated prevalence of the solid thyroid nodules in a general population, an appropriate selection of cases referred to surgery is of paramount importance. The main aim of our study was to evaluate the performance of the Thyroid Imaging Reporting and Data System (TI-RADS) as a differential diagnosis tool for thyroid nodules. MATERIAL AND METHODS: We evaluated 174 nodules using TI-RADS risk stratification model, using conventional ultrasound and real time elastography parameters and linear multifrequency probe (Hitachi Preirus Machine, Hitachi Inc., Japan). All the nodules were classified using the TI-RADS system according to echogenicity, margins, shapes, calcification, lymph nodes, and increased strain ration. The results were compared with the pathology exam, which was considered the golden standard diagnosis. RESULTS: The prevalence of malignant nodules was 16.7% (29 cases). The differential diagnosis performance regarding the malignant tumor for TI-RADS is appropriate for clinical use, obtaining an area under ROC curve of 0.95761 [0.8424-0.989] 95% confidence interval. Combining TI-RADS 2, 3 and 4A as probably benign and TI-RADS 4B and 5 as probably malignant, the sensitivity, specificity, positive predictive value, and negative predictive value were 97.93%, 86.20%, 97.26% and 89.28% respectively. The overall accuracy of the method was 95.97%.According to the American Association of Clinical Endocrinologists Guidelines, 169 cases required FNAB examination; by applying the TI-RADS scoring system, the necessity for FNAB would decrease to 74 cases. CONCLUSION: Quantitative strain elastography, as the 6th parameter of TI-RADS system, adds diagnostic power to the risk stratification model.
AIMS: Due to the elevated prevalence of the solid thyroid nodules in a general population, an appropriate selection of cases referred to surgery is of paramount importance. The main aim of our study was to evaluate the performance of the Thyroid Imaging Reporting and Data System (TI-RADS) as a differential diagnosis tool for thyroid nodules. MATERIAL AND METHODS: We evaluated 174 nodules using TI-RADS risk stratification model, using conventional ultrasound and real time elastography parameters and linear multifrequency probe (Hitachi Preirus Machine, Hitachi Inc., Japan). All the nodules were classified using the TI-RADS system according to echogenicity, margins, shapes, calcification, lymph nodes, and increased strain ration. The results were compared with the pathology exam, which was considered the golden standard diagnosis. RESULTS: The prevalence of malignant nodules was 16.7% (29 cases). The differential diagnosis performance regarding the malignant tumor for TI-RADS is appropriate for clinical use, obtaining an area under ROC curve of 0.95761 [0.8424-0.989] 95% confidence interval. Combining TI-RADS 2, 3 and 4A as probably benign and TI-RADS 4B and 5 as probably malignant, the sensitivity, specificity, positive predictive value, and negative predictive value were 97.93%, 86.20%, 97.26% and 89.28% respectively. The overall accuracy of the method was 95.97%.According to the American Association of Clinical Endocrinologists Guidelines, 169 cases required FNAB examination; by applying the TI-RADS scoring system, the necessity for FNAB would decrease to 74 cases. CONCLUSION: Quantitative strain elastography, as the 6th parameter of TI-RADS system, adds diagnostic power to the risk stratification model.