Diagnostic accuracy of Thyroid Imaging and Data System (TIRADS) ultrasound classification of thyroid nodules compared with cytological findings among patients in Tanzania.

Chausiku Samwel Kisunte; Clement Frank Kalambo; Patrick Amsi; Felister Uisso; Adnan Sadiq

doi:10.5455/ajdi.20260114121548

Abstract
Background: Thyroid nodules are common worldwide, particularly among women and in iodine-deficient regions. Although most nodules are benign, a significant proportion may be malignant. Ultrasound is the primary imaging tool for evaluation, and the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) has shown strong performance in risk stratification. This study assessed the diagnostic accuracy of ACR TI-RADS compared with cytological findings at Kilimanjaro Christian Medical Centre (KCMC).Methods: A cross-sectional study was conducted among 255 patients who underwent thyroid ultrasound and fine-needle aspiration cytology (FNAC) between January 2022 and April 2025. For diagnostic accuracy analysis, Bethesda II–III were grouped as benign, Bethesda IV–VI as malignant, and Bethesda I cases were excluded. ACR TI-RADS categories TR1–TR3 were considered low risk (test-negative), while TR4–TR5 were considered high risk (test-positive). Sensitivity, specificity, predictive values, and overall accuracy were calculated, and ROC curve analysis was performed.Results: The median age was 47 years, and 86.3% of patients were female. Most nodules were mixed cystic/solid (58.8%), hyperechoic (60.4%), wider-than-tall (89.0%), and had smooth margins (67.5%). Based on ACR TI-RADS, 82% of nodules were low risk and 18% high risk. FNAC classified 83.9% as benign, 11% malignant, and 6.3% suspicious. Malignancy rates rose with higher TI-RADS categories, reaching 88% in TR5. ACR TI-RADS demonstrated high diagnostic accuracy with 90.6% sensitivity, 92.6% specificity, and an AUC of 0.951.Conclusion: ACR TI-RADS is a highly sensitive and specific, non-invasive tool for thyroid nodule evaluation. While FNAC remains the reference standard, this single-center study suggests TI-RADS has the potential to reduce unnecessary FNACs, particularly in benign nodules. Limitations include reliance on FNAC, single-center design, and operator dependence, which may affect reproducibility.

Key words: Thyroid nodules, Ultrasound, ACR-TIRADS, Cytology, Diagnostic accuracy,Tanzania