Is Bethesda classification sufficient to predict thyroid cancer in endemic regions?

Abstract

Aim: Bethesda classification is widely used to determine the risk of malignancy of thyroid nodules and in many guidelines, treatment algorithms are determined according to this classification. We aimed to investigate the accuracy of malignancy predictions of Bethesda classification in patients who underwent surgery. Methods: In this retrospective cohort study, the medical records of patients who underwent thyroidectomy between 2013 and 2017 were analyzed. Patients’ demographic characteristics, fine needle aspiration biopsy (FNAB) results, ultrasonographic findings, number of nodules, diameter of nodules, type of surgeries performed, and the terminal pathology results were recorded. Malignancies that were detected in Bethesda 1-2 patients and in nodules other than the nodule to which FNAB was performed in other Bethesda categories were defined as incidental cancer. Results: Nine hundred sixty-seven patients were included in the study. The mean age of the patients was 46.9 (12.4) years and 82.4 % (n=797) were female. Mean nodule diameter was 29.7(13.9) mm and 64.3% (n=622) of the patients had 3 or more nodules. In our series, the rate of malignancy was 24.2% for Bethesda 1, 24.7% for Bethesda 2, 35 % for Bethesda 3, 52.1 % for Bethesda 4, 91.2 % for Bethesda 5 and 100 % for Bethesda 6. In all categories, the malignancy rate was higher than the expected range, but statistical significance was determined in Bethesda 1, 2, 4 and 5 groups. When compared to the whole series, rate of incidental cancer was 18.2%, whereas the incidental cancer rate was 49.5 % of all cancers. The tumor was multicentric in 34.6 % of malignant cases. Conclusion: In endemic regions, higher rates of malignancy are detected than that predicted by the Bethesda classification. Therefore, updates on guidelines in patient selection for surgery and in choosing the surgical technique may be necessary for endemic areas.