Effectiveness of diffusion-weighted magnetic resonance imaging (DW-MRI) in the differentiation of thyroid nodules

Abstract

Background: The aim was to investigate which of two different b values (b 500 s/mm² and b 800 s/mm²) are more effective in the differentiation of benign-malignant nodules using Diffusion-Weighted Magnetic Resonance Imaging (DW-MRI).

Materials and methods: Patients presenting with a preoperative diagnosis of nodular goiter or multinodular goiter were included in this study. These patients underwent neck MRI examinations, and their cases were analyzed retrospectively. A total of 26 patients were included in the study. A total of 46 nodules meeting the study criteria were examined. Measurements were performed on Apparent Diffusion Coefficient (ADC) maps of patients at two different b values (b 500 s/mm² and b 800 s/mm²), and the results were compared with histopathological findings.

Results: Out of a total of 46 nodules, 37 were identified as benign, and 9 as malignant based on histopathological analysis. The mean ADC value at b 500 was lower in malignant nodules (1259.65 ± 328.13) compared to benign nodules (19037.48 ± 472.74). Similarly, the mean ADC value at b 800 was lower in malignant nodules (1081.72 ± 200.23) compared to benign nodules (1610.44 ± 418.06). When a cut-off value of 1.1 × 10^{- 3} was accepted for the differentiation of pathology, the sensitivity for distinguishing pathology with ADC values at b 500 was 83.3%, with a specificity of 90.0%, and for ADC values at b 800, the sensitivity was 71.4%, with a specificity of 89.7%.

Conclusion: DW-MRI without the need for contrast agent administration is a useful method in the differentiation of benign-malignant thyroid nodules.

Keywords: Apparent diffusion coefficient; B values; Diffusion-weighted magnetic resonance imaging; Thyroid nodule.

Fig. 1

In a 77-year-old female patient diagnosed with anaplastic tumor, (A) Axial T2A image shows a solid nodule with hyperintense capsular invasion and vascular invasion in the left lobe, compressing the trachea. Pleural fluid is also observed in the left hemithorax in the images. (B) In the b 500 value ADC map, the ADC value of the nodule was measured as 1.0 × 10^− 3, (C) in the b 800 value ADC map, the ADC value of the nodule was measured as 1.0 × 10^− 3

Fig. 2

In a 47-year-old female patient with a benign nodule (A), a nodule is observed in the left lobe of the thyroid gland, compressing the trachea. (B) On the ADC map, the nodule was hyperintense and the ADC value of the nodule was measured as 1.8 × 10^− 3 at b 500 value. (C) The ADC value of the nodule was measured as 1.3 × 10^− 3 on the b 800 value ADC map

Fig. 3

Boxplot of ADC values in benign nodules, malignant nodules, and normal thyroid parenchyma

Fig. 4

Boxplot of ADC values in benign nodules, malignant nodules, and normal thyroid parenchyma

Fig. 5

ROC Curve for ADC500 value in malignant-benign nodule analysis. The ROC area of b 500 ADC values was found to be 0.880 ± 0.065 (0.753–1.006). The prediction point was determined as 1.1 × 10^-3. At this value, the sensitivity of b 500 ADC value is 83.3%; specificity 90.0%; The positive predictive value was 55.6% and the negative predictive value was 97.3%

Fig. 6

ROC curve for ADC800 value in malignant-benign nodule analysis. The ROC area of b 800 ADC values was found to be 0.916 ± 0.041 (0.835–0.996). The prediction point was determined as 1.1 × 10^-3. At this value, the sensitivity of the b 800 ADC value is 71.4%; specificity 89.7%; The positive predictive value was 55.6% and the negative predictive value was 94.6%