Characterization of thyroid nodules using ACR 2017 tirads classification system and correlation with the bethseda FNAC system at Kenyatta national hospital.

Abstract

Introduction: Thyroid gland pathology can be diffuse enlargement or focal nodules. Besides clinical examination, imaging and cytology play a key role in further evaluation of the varied thyroid pathology. All of these have resulted in improved evaluation of thyroid nodules, with the modality of first choice being high resolution ultrasound. Owing to this increased detection, an upsurge has been seen in the number of patients undergoing invasive procedures to discern benign from malignant nodules. Majority of these are found to be benign, leading to development of sonographic risk stratification models to help guide the clinician on which nodules require follow up or cytologic diagnosis. In Kenya, there is limited data comparing TIRADS and fine needle aspiration cytology (FNAC) in assessment of thyroid nodules and in most cases the TIRADS classification is not employed. Study Objective: To subject all patients diagnosed with thyroid nodules and referred for FNAC to sonographic stratification using the ACR 2017 TIRADS and correlate with the FNAC Bethesda Classification. Materials and Methods: Following authorization by the by the Kenyatta National Hospital/University of Nairobi Ethics and Research Committee a prospective cross-sectional study was conducted on 79 consenting patients who satisfied the inclusion criteria at Kenyatta National Hospital (KNH). The study ran from November 2019 to May 2020. Conventional B- mode and Color Doppler ultrasound of thyroid nodules was done by the principal investigator using a standard procedure and images saved and findings stratified. These were later verified by 2 consultant radiologists well versed with thyroid sonography and a score assigned according to the ACR TIRADS. Ultrasound guided FNAC of the identified nodules was done aseptically using standard ultrasound guided free hand FNA procedure. Diagnostic quality of the aspirate was confirmed, and slides fixed by a qualified cytotechnologist. The slides were sent to a qualified cytopathologist for evaluation and classification according to the Bethesda Classification. Data was captured using a standardized Microsoft Excel data collection tool. Demographic and clinical characteristics were summarized using proportions for categorical variables and means or medians for continuous variables. Prevalence of malignancy was summarized as a proportion and concordance rate defined as the number of classification agreement outcomes over the total number assessed. Analysis was done using Stata version 16. A comparison with other worldwide studies was made. Results A total of 79 patients were evaluated and of these, 72 (91.9%) were female and 7 (8.2%) were male. Majority of the patients were in the 4th decade of life accounting for 39.2% of the sample population, with a median age of 45yrs. Out of 108 nodules in our study, 106 were benign and 2xii were malignant. Of these nodules, 5 were categorized as TIRADS 1, 40 as TIRADS 2, 37 as TIRADS 3, 24 as TIRADS 4 and 2 as TIRADS 5. The percentage risk of malignancy based on TIRADS classification was found to increase with increase in TIRADS classification. TIRADS 1-3 had 0%, TIRADS 4 had 4.2% risk and TIRADS 5 had 50% risk of malignancy with a p value of <0.05. The overall risk of malignancy was 1.9%. The concordance of ACR TIRADS with Bethseda FNAC was 100% for TIRADS 1 to 3 with Bethseda II and 50% for TIRADS 5 with Bethseda V. Upon application of Cohen’s Kappa, slight agreement was shown which was not statistically significant (p value of 0.2514). The ACR TIRADS diagnostic performance was calculated using FNAC as the gold standard test and was found to have an overall sensitivity of 100%, specificity of 77.36%, positive likelihood ratio of 4.42, negative likelihood ratio of 0.0, positive predictive value of 7.69%, negative predictive value of 100% and overall diagnostic accuracy of 77.78%. (p value < 0.05) Conclusion ACR 2017 TIRADS as a stratification system can identify almost all benign lesions as per this study owing to its high specificity and high concordance between TIRADS 1-3 and Bethseda II. It is simple to use in day to day practice owing to its point based format, allowing all nodules to be classified and provides a tool that can be used in management of thyroid nodular disease to avoid unnecessary biopsies.