COMPARISON BETWEEN BETHESDA SYSTEM FOR REPORTING THYROID CYTOLOGY & ULTRASOUND BASED THYROID IMAGING REPORTING & DATA SYSTEM

Abstract

Background: Background: The use of ultrasonography and fine needle aspiration cytology (FNAC) is complementary when it comes to the assessment of thyroid nodules; however, the extent to which there is concordance between the two sets of risk stratification criteria in daily practice remains uncertain. Objective: To assess the correlation between The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC 2023) and the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) in patients with thyroid nodules who underwent ultrasound-guided FNAC. Methods: This cross-sectional analytical study was conducted in the Department of Pathology, Fatima Memorial Hospital, Lahore, over three months following synopsis approval (February 2026-May 2026). Sixty patients, both male and female, aged 18 years or above, presenting with thyroid nodules were included by non-probability consecutive sampling. Each nodule was assigned a TI-RADS category on ultrasound and, subsequently, a Bethesda category on FNAC. Patients with non-diagnostic smears or previously treated thyroid malignancy were excluded. Inter-system agreement was quantified using Cohen's kappa after collapsing both systems onto a common three-tier risk stratum (benign, indeterminate, malignant); the chi-square test was used to assess the association between the two systems. A p-value <0.05 was taken as significant. Results: The mean age was 42.9 ± 13.9 years (range 18-60 years) and out of 60 patients, 45 (75.0%) were female. On FNAC, 37 nodules (61.7%) were Bethesda II, 6 (10.0%) Bethesda III, 6 (10.0%) Bethesda IV, 5 (8.3%) Bethesda V and 6 (10.0%) Bethesda VI. On ultrasound, TI-RADS 3 was the commonest category (21 nodules, 35.0%), followed by TI-RADS 2 (16, 26.7%), TI-RADS 4 (14, 23.3%), TI-RADS 5 (8, 13.3%) and TI-RADS 1 (1, 1.7%). Case-defined concordance, applying the pre-specified category-correspondence criterion, was observed in 49 of 60 cases (81.7%). On the collapsed three-tier comparison, overall agreement was 56.7%, Cohen's kappa was 0.367 (95% CI 0.18-0.55), indicating fair agreement, and the association between the two systems was statistically significant (χ² = 30.85, df = 4, p<0.001). Five of the six Bethesda VI nodules and all five Bethesda V nodules fell in the higher-suspicion TI-RADS categories (TR4-TR5), whereas discordance clustered chiefly among Bethesda II nodules assigned TI-RADS 4. Conclusion: TBSRTC and ACR TI-RADS showed fair, statistically significant agreement, with the strongest correspondence at the extremes of the risk spectrum and the greatest divergence among cytologically benign nodules carrying higher ultrasound suspicion. Neither system can be regarded as an adequate substitute for the other, and combined interpretation is advisable for nodules that fall in the intermediate range of either classification.