Why the ACR TIRADS Outperforms Other Systems: Application of Higher Dimensional Thresholds to Indicate Biopsy Would Reduce the Number of Biopsies

  • Background/Purpose. The American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TIRADS) was recently proposed with the aim of reducing thyroid biopsies. It was reported to outperform other risk-stratification systems. The aim of this study was to explore whether this superior performance is caused by the higher size thresholds for biopsy indication.

    Methods. All thyroid nodules referred for biopsy between November 2015 and September 2018 were included. Sonographic features were collected to evaluate the indication to biopsy according to ACR TIRADS. Then we classified the nodules under other three risk-stratification systems (the ATA Guidelines, the European [EU-TIRADS], and the Korean TIRADS [K-TIRADS]) but applying the same size cutoffs proposed by the ACR for similar risk classes. Nodules <1 cm and without a definite pathological diagnosis were excluded. Indication to biopsy according to each system was considered as test positivity.

    Results. The final cohort included 553 nodules (42 malignant). When ACR size thresholds were applied to the other systems, the number of avoidable biopsies increased in the K-TIRADS (291, 52.6%) to be at least comparable with the ACR TIRADS (287, 51.9%), whereas in the ATA system (405, 73.2%) it was significantly higher than in the ACR TIRADS (p<0.001). The original ACR TIRADS confirmed the highest negative predictive value (NPV) (97.2%, 95% confidence interval 94.6-98.8%).

    Conclusion. The excellent NPV of the ACR TIRADS can be attributed to the structure of risk classes, whereas the main drivers of biopsy reduction are the higher size cutoffs.

 

Leave a Reply