Correlation of ultrasound thyroid imaging reporting and data system with cytology among patients at Mulago national referral and teaching hospital

Abstract

Background: Ultrasonography (US) is noninvasive modality for initial assessment of thyroid nodule. Thyroid Imaging Reporting and Data System (TI-RADS) classify the risk for malignancy has demonstrated good performance in differentiating malignant thyroid nodules. However, the combination of TI-RADS categories and cytology had not been studies extensively. Objective: To correlate Thyroid Imaging Reporting and Data System (TI-RADS) with cytology among patients referred for US guided-FNNA cytology at Mulago hospital. Methodology: This was a descriptive cross-sectional study involving patients who were referred for US guided-FNNA cytology of thyroid nodules at Mulago NRTH. Nodule sonographic appearance was documented and categorized into five TI-RADS levels. For Nodules ≥1cm, FNNA cytology was done. A standardized questionnaire was used to obtain data. Spearman’s correlation coefficient was used to establish correlation between TI-RADS and FNNA cytology findings. The Sensitivity, Specificity, Positive predictive values (PPV), Negative predictive values (NPV), Positive Likelihood ratios (PLR) and Negative likelihood ratios (NLR) was determined by comparing ACR-TIRADS with Bethesda system of thyroid classification as a gold standard. Results; Majority of the participants were females 117/130 (90%). The mean age was 41 years with a standard deviation of 13 years. The majority were solid or almost solid 87(54.7%), shaped wider than tall 154(96.9%), with nodules having smooth margins 37(57.2%), 133(83.7%) hyperechoic or isoechoic and 141(88.7%) having none or large comet tail artifact. TIRADS 3 was the commonest (42.9%). The proportion of malignancy for TIRADS 4, and TIRADS 5 was 73.3% and 85.7%. The correlation between ACR-TIRADS and the Bethesda system of thyroid classification scores was r=0.577 and this was statistically significant (p<0.001). The Sensitivity, Specificity, PPV, NPV, PLR and NLR of ACR-TIRADS to detect malignancy was 94.4% with 95% CI of 0.944, 96.5% with 95% CI of 0.965, 77.3%, 99.3%, 27 and 0.06 respectively. Conclusion and Recommendation: We found that ACR-TIRADS classification is appropriate and non-invasive method for assessing thyroid nodules in routine practice. Also, ACR-TIRADS can safely reduce number of unnecessary FNNA in a significant proportion of benign thyroid lesions. Thyroid nodules classified as TIRADS 3 should be followed routinely. ACR-TIRADS should be standardized as the screening tool in resource limited areas.