Abstract
Thyroid nodule is a common disease in clinical practice. The diagnosis of malignant thyroid tumors determines the treatment strategy. Among a number of methods have claimed to help evaluating thyroid nodules, ultrasound is a usable one in spite of several disadvantages (dependent on the physician/technician, incomparable, etc.) and magnetic resonance imaging (MRI) accompanied by quantitative apparent diffusion coefficient (ADC) is a promising diagnostic tool. This study was designed to investigate the usefulness of ADC cut-off values and the protocol of thyroid MRI derived from quantitative diffusion weighted imaging (DWI) in differentiating benign and malignant thyroid nodules. The study was conducted on 93 patients with 128 thyroid nodules, diagnosed and underwent surgery at Hanoi Medical University Hospital. All the patients took thyroid MRI with different b levels (from 200 to 800). ADC value was calculated to each b level, and the statistical tests were conducted with the Statistical Package for Social Sciences (SPSS-Windows and Mac version 20) and STATA 12. The mean ADC with all the b ranging from 200 to 800 of malignant groups was significantly higher than the group of benign lesions (p from <0.001 to 0.01). We chose b = 500 as a standard b-value in the protocol of thyroid MRI. The ADC cut-off point for distinguishing malignant from benign thyroid lesions: 1.7 × 10(-3) mm(2)/s with high accuracy (87.1%, 95% CI: 79.59-92.07%). The study revealed that quantitative diffusion weighted MRI with ADC measurement could potentially quantitatively differentiate between benign and malignant thyroid nodules.