MRI and ultrasonography detection of cervical lymph node metastases in differentiated thyroid carcinoma before reoperation

Objective: The purpose of this study was to compare the diagnostic capabilities of magnetic resonance imaging (MRI) and ultrasonography (US) for cervical lymph nodal metastases in differentiated thyroid carcinoma (DTC) before reoperation. Material and method: From June 2011 to May 2013, preoperative MRI and ultrasound data were collected from differentiated thyroid cancer patients who underwent a reoperation. The following characteristics were assessed: the sensitivity, specificity, positive predictive value, negative predictive value and accuracy of MRI and US. The MRI and ultrasound findings were correlated with the histological diagnosis after reoperation. Results: One hundred and thirty-eight cases were included in the analysis. 88.4% (122/138) of which had evidence of residual thyroid cancer tissue or metastatic nodal involvement at final histology. Lymph nodal metastases were confirmed in the central compartment in 76.42% (81/106) of patients and in lateral compartment in 73.28% (85/116) of patients. The sensitivity, specificity and accuracy of MRI VS US for detecting central compartment metastases was 75% VS 41.67% (P=0.04), 90.91% VS 100% (P=1) and 80% VS 60% (P=0.618), respectively; For detecting lateral compartment metastases was 83.33% VS 77.78% (P=1), 25% VS 50% (P=0.606) and 65.38% VS 69.23% (P=1), respectively. There was statistically significant difference between the sensitivity of MRI and ultrasound for diagnose of central compartment metastases. The MRI features with the greatest correlation with positive lymph nodal metastases were fusion and enhancing lesions. The ultrasound features with the greatest correlation with positive lymph nodal metastases were hypoechoic and microcalcifications. Conclusion: MRI is more sensitive than ultrasonography in detecting central compartment metastases in papillary thyroid carcinoma. There is no significant difference in diagnosis of lateral neck node metastases between MRI and US.