Diagnostic accuracy of thyroid imaging reporting and data system (Ti-Rads) in distinguishing benign from malignant nodules, keeping histopathology as gold standard in a tertiary care hospital
Background: thyroid nodules are common particularly in iodine-deficient regions such as Pakistan and most are benign. Accurate, non-invasive risk stratification is therefore essential to avoid missed cancers and unnecessary procedures. The American college of radiology thyroid imaging reporting and data system (ti-rads) offers a standardized ultrasound framework. This study evaluated the diagnostic accuracy of ti-rads against histopathology and compared performance with fine-needle aspiration cytology (fnac). Methods: In a single-center cross-sectional study over 10 months, 150 adults (20–60 years) with a single thyroid nodule underwent high-resolution ultrasound with ti-rads scoring and ultrasound-guided fnac. A surgical/biopsy subset had histopathology as the reference standard. Data were analyzed in spss v25; continuous variables were summarized as mean±SD or median. Diagnostic metrics (sensitivity, specificity, positive predictive value (ppv), negative predictive value (npv), accuracy) were calculated from 2×2 contingency tables versus histopathology Results: women comprised 85.3% (128/150); mean age 42.4±14.2 years; right-lobe nodules were common (69.3%). ti-rads distribution favoured lower risk (tr3 62.0%, tr4 22.0%). in 32 cases with histopathology, ti-rads showed sensitivity of 70.0%, specificity 77.3%, ppv 58.3%, npv 85.0%, and accuracy 75.0% with 95%CI range (67.2% – 81.0%). FNAC performed better having sensitivity of 90.9%, specificity 81.0%, ppv 71.4%, npv 94.4%, and accuracy 84.4%. histopathology-based risk of malignancy rose stepwise with ti-rads category. tr2 33.3% (4/12), tr3 50.0% (4/8), tr4 70.0% (7/10), tr5 100% (2/2). Conclusion: Ti-rads showed moderate diagnostic performance with a graded increase in malignancy risk, suggesting its potential usefulness as a triage or screening framework. FNAC appeared to offer higher sensitivity and npv, highlighting its value in confirming or excluding malignancy.
Authors
- Ruqaiya Shahid (ORCID: https://orcid.org/0000-0001-5331-3893)
- Mariam Taufiq
- Nabil F. Saba (ORCID: https://orcid.org/0000-0003-4972-1477)
- Binish Rasheed (ORCID: https://orcid.org/0000-0002-8894-3729)
- Nida Rafiq
- Nasreen Naz
Institutions
- Dow University of Health Sciences (PK)
Publication Details
- Journal
- International Journal of Pathology
- Published
- 2025-09-29
- DOI
- https://doi.org/10.59736/ijp.23.03.984
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00