The accuracy of fine-needle aspiration cytology and ultrasonography in assessing thyroid nodules in correlation with histopathology: a retrospective study.

Abstract

Background: Accurately diagnosing thyroid nodules is vital for preventing unnecessary surgeries and providing prompt therapy. Although fine-needle aspiration cytology (FNAC) and ultrasonography (US) are widely used diagnostic methods, their reliability is questioned. This study investigates the effectiveness of US and FNAC in thyroid nodule diagnosis and differentiates benign from malignant nodules in relation to final histopathological diagnosis.

Method: A retrospective study including 307 adult patients with thyroid diseases who underwent neck US and FNAC before surgery was conducted between April 2019 and May 2023. The diagnostic efficacy of US, FNAC, and their combination usage was compared to histopathological results.

Result: Histopathological findings revealed that 187 (61%) cases were benign, while 120 (39%) were malignant. The US features of 'taller-than-wider' forms and hypoechoic appearance had the highest diagnostic accuracy in characterizing malignant thyroid nodules, with 83 and 73% accuracy, respectively. The combination of US parameters demonstrated high sensitivity, specificity, positive predictive value, and negative predictive value (NPV) of 88.33, 63.10, 60.6, and 89.4%, with a statistically significant area under the ROC curve (AUC: 0.828, P<0.001) than individual parameters. FNAC's sensitivity, specificity, PPV NPV, and accuracy in detecting malignant lesions were 50, 95, 86, 75, and 77%, respectively, with acceptable discrimination and statistical significance (AUC: 0.723, P<0.0001). The combination of US parameters and FNAC significantly improved the AUC value (AUC: 0.878, P<0.0001), sensitivity (83.33%), and specificity (79.14%). Univariate analysis showed that hypoechoic appearance, heterogenicity, large mass size (>4 cm), 'taller-than-wider', infiltrative margins, and microcalcifications were risk factors for malignancy in thyroid nodules and were statistically significant (all P-values <0.05).

Conclusion: Combining US characteristics with FNAC results can afford the maximum analytical accuracy in distinguishing benign from malignant thyroid nodules. This strategy is practical due to its simplicity, minimal invasiveness, and cost-effectiveness, enabling robust management regimens and avoiding additional surgical procedures.