Impact of cytological pitfalls in the Bethesda System of Reporting Thyroid Cytopathology, on surgical decision-making of patients with thyroid nodules: Can these pitfalls be avoided?

INTRODUCTION

The purpose was to assess diagnostic accuracy of the Bethesda System of Reporting Thyroid Cytopathology, reasons for disagreement and cytological pitfalls with impact on surgical decisions in patients with thyroid nodules.

METHODS

Cases of thyroid cytology with histological follow up were included followed by cytological-histological correlation and were reviewed to look for reasons for discrepancies. The impact of disagreements and partial agreements on surgical decision was evaluated. Overall and complete diagnostic accuracy were calculated along with sensitivity, specificity, and positive and negative predictive values for malignant and neoplastic lesions.

RESULTS

Of 446 cases included in the study, there was complete agreement in 358 cases, partial agreement in 22 cases and disagreement in 66 patients. Overall diagnostic accuracy was 98.5% with sensitivity, specificity, positive predictive value and negative predictive value of 80%, 99.6%, 100%, 72.7% and 94.3% respectively for malignant lesions. Overall diagnostic accuracy was highest for the malignant category. Follicular patterned lesions, Hürthle cell-rich smears and overlapping cytological features between benign and malignant follicular neoplasm were the main reason of discrepancy. The discrepancy in cytological diagnosis altered the decision of type of surgery performed in 13.6% of the patients.

CONCLUSION

Fine needle aspiration cytology remains a powerful screening tool to aid decision-making in the majority of the patients. Histological-cytological discrepancy can adversely impact the management of patients. Factors causing an adverse impact on surgical management were rare and potential avoidable reasons for them were identified. Cytological pitfalls may be avoided through screening for minor components, clinico-radiological correlation and experience.