Proposal for a scoring system to differentiate thyroid malignancies from benign lesions using selected demographic and sonographic data and the Bethesda system for reporting thyroid cytopathology

Background and objective: Differentiating benign from malignant thyroid nodules by using cytological findings alone can be misleading. Concurrent use of a reliable risk prediction tool can facilitate clinical decision-making and reduce the burden on patients and the health care system by minimizing unnecessary surgeries.
Methods: A cross-sectional study was conducted at Base Hospital, Avissawella, using data from patients with a confirmed histological diagnosis of thyroid nodules, who had ultrasonography and fine needle aspiration cytology (FNAC) prior to surgery. A scoring system was developed to predict the risk of thyroid nodules being malignant by incorporating age, sex, FNAC test results, and ultrasound findings. The benign or malignant status of the nodules determined using the scoring system was then compared with the histological diagnosis at various threshold settings. Receiver operating characteristics (ROC) analysis was used to determine an optimal cut-off threshold for the developed score analyses.
Results: A total of 138 cases were investigated. The scoring system was appropriate for clinical use as indicated by an area under ROC curve (AUC) of 0.752 (95% confidence interval = 0.622-0.882; p<0.001). A cut-off score of 8.5 had a sensitivity of 80% and a specificity of 54.5% for the diagnosis of malignancy. The proposed scoring system had a high negative predictive value at low score levels and a high proportion of cases were included in this category. At high score levels, it had a greater positive predictive value to differentiate benign from malignant nodules.
Conclusion: The proposed scoring system can be used as an adjunct to differentiate benign lesions from thyroid malignancy since the AUC was 0.752. We suggest a cut-off score of 8.5 as an acceptable trade-off between a sensitivity of 80% and a specificity of 54.5%. At scores of 7.5 and below, the nodules can be considered benign and safely followed up.