The American Thyroid Association (ATA) and the Thyroid Image Reporting and Data System (TI-RADS) provide some assistance for determining thyroid nodule malignancy in pediatric patients undergoing ultrasound readings, according to a retrospective study published in Pediatric Radiology.
In this retrospective study, investigators evaluated test characteristics of the adult-based ATA and TI-RADS assessments following ultrasound among 124 pediatric patients with thyroid nodules.
A total of 52 malignant and 71 benign thyroid nodules were assessed using ultrasound. After using pattern descriptions from the ATA, radiologists were able to classify approximately 80% of all malignant nodules as having a “high” (69%) or “intermediate” (11%) likelihood of malignancy. In addition, approximately 28% of benign nodules were given “high” or “intermediate” classifications under the ATA pattern descriptions.
When radiologists used the TI-RADS, they were able to classify malignant nodules as 2, 3, 4a, 4b, 4c, and 5, each with a malignancy rate of 0%, 0%, 13.5%, 13.5%, 61.5%, and 11.5%, respectively. Overall, positive and negative predictive values for TI-RADS were 71.7% and 80.0%, compared with 68.0% and 87.5% for ATA.
Potential selection bias of the radiologist may have existed in the original assessments, presenting a potential limitation with this study. Additionally, not all lymph nodes were assessed in each case, reducing the insight into the clinical utility of the ATA and TI-RADS strategies.
The investigators emphasize that the “different tumor biology in children, including higher risk for lymph node involvement, pulmonary metastasis, multifocal disease, and recurrence risk…must encourage prompt recognition of features of malignancy in children and distinction from nodules likely to be benign so as to appropriately triage patients for surgical intervention vs observation.”