Comparison of diagnostic value between American College of Radiology thyroid imaging reporting and data system and 2015 American Thyroid Association guidelines in thyroid nodules

2018 
Objective To compare the diagnostic value between the recommended ultrasound characteristics by the American College of Radiology thyroid imaging reporting and data system (ACR TI-RADS) and the 2015 American Thyroid Association (ATA) guidelines in thyroid nodules. Methods We retrospectively analyzed the conventional ultrasound features of 295 thyroid nodules (76 benign and 219 malignant) in 218 patients that were confirmed by surgery and puncture biopsy pathology from September 2015 to July 2017 at the First Affiliated Hospital of Shanxi Medical University by the ACR TI-RADS classification and ATA risk stratification. The results were compared with the pathology results. The receiver operating characteristic (ROC) curve was then plotted. The χ2 test was used to compare the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the two methods to evaluate their diagnostic value. Results The areas under the ROC curves of the ACR TI-RADS classification and the ATA risk stratification were 0.889 and 0.902, respectively, and there was no significant difference between them (P=0.296).The best cut-off points for ACR TI-RADS and ATA were TR5 and high risk, respectively, and the corresponding sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 85.4% vs 90.0%, 88.2% vs 88.2%, 95.4% vs 95.6%, 67.7% vs 75.3%, and 86.1% vs 89.5%, respectively (P>0.05, Kappa=0.92). Conclusion Both ACR TI-RADS classification and ATA risk stratification have high diagnostic value in thyroid nodules. Using TR5 and high risk as the respective cut-off points of the ACR TI-RADS classification and ATA risk stratification has good consistency and highest value for the diagnosis of thyroid malignancy. Key words: Thyroid nodule; American College of Radiology; American Thyroid Association
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []