IMAGING & INTERVENTION
Original Article

Should We Use Magnetic Resonance Imaging for Thyroid Nodules with Indeterminate Cytology?

1.

The University of Texas MD Anderson Cancer Center, Houston, TX, USA

2.

Department of General Surgery, Istanbul University Cerrahpaşa Hospital, Istanbul, Turkey

3.

Department of Radiology, Marmara University Hospital, Istanbul, Turkey

4.

Department of Internal Medicine, Istanbul University Cerrahpaşa Hospital, Istanbul, Turkey

5.

Department of Radiology, Istanbul University Cerrahpaşa Hospital, Istanbul, Turkey

IMAGING & INTERVENTION 2021; 1: 52-60
DOI: 10.5152/iai.2021.21028
Read: 1410 Downloads: 589 Published: 01 December 2021

Background: Although most thyroid nodules with indeterminate cytology (Bethesda III/IV) are benign, they represent a manage- ment dilemma possibly leading to unnecessary surgery. This study aims to evaluate the role of magnetic resonance imaging inthe risk stratification of thyroid nodules with indeterminate cytology.

Methods: Patients scheduled for thyroidectomy for thyroid nodules with indeterminate cytology were prospectively enrolled.Diffusion-weighted images and T2-weighted images were obtained. Apparent diffusion coefficient values and T2-weighted sig- nal intensities of thyroid nodules, normal thyroid parenchyma, muscle tissue, and spinal cord were recorded.

Results: In the final analysis, 17 nodules (5 malignant and 12 benign) were included. Apparent diffusion coefficient values, nodule- to-spinal cord apparent diffusion coefficient ratio, and nodule-to-muscle T2 signal intensity ratio were significantly lower inmalignant nodules (P = .019, P = .019 and P = .037, respectively). Using a threshold of 1.570 ×10−3 mm2/s, nodule apparent diffusioncoefficient value had a sensitivity of 100%, specificity of 75%, positive predictive value of 62.5%, negative predictive value of 100%, and an accuracy of 82.4%. At a threshold of 3.831, nodule-to-muscle T2-weighted signal intensity ratio had a sensitivity of 100%, specificity of 50%, positive predictive value of 45.5%, and negative predictive value of 100%. If falling under both of thesethreshold values was required for the diagnosis of malignancy, then the sensitivity and negative predictive value would remainunchanged at 100%, but there would be an increase in specificity, positive predictive value, and accuracy to 83.3%, 71.4%, and 88.2%, respectively. Furthermore, 83.3% (10/12) of benign nodules could have been spared from unnecessary surgery.

Conclusion: Magnetic resonance imaging is a very sensitive tool for discriminating benign from malignant thyroid nodules with indeterminate cytology and it may spare patients from unnecessary surgery.

Cite this article as: Awiwi MO, Teksöz S, Akbaş S, Gjoni M, Kızılkılıç O. Should we use magnetic resonance imaging for thyroid nodules with indeterminate cytology? Imaging Interv. 2022;1(3):52-60.

Files
EISSN 2791-805X