• Wen-Wen He, Jian-Guo Zhu, Dmytro Pylypenko, Fei Liu, Mei Wang, Yue-Fei Wu, Jun Tian, and Hai-Ge Li.
• From the Department of Radiology, The Second Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu Province, P.R. China.
• Medicine (Baltimore). 2022 Oct 7; 101 (40): e30861e30861.

AbstractTo synthetically evaluate the diagnostic accuracy of image features for differentiating benign from malignant gallbladder wall thickening disease with non-contrast MRI and establish the optimal diagnostic indicator. A total of 23 patients with wall thickening type gallbladder carcinoma and 61 patients with benign wall thickening disease were included. The diagnostic performance of six image features including the layered pattern on T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) images, T2WI signal intensity, papillary growth, the apparent diffusion coefficient (ADC) value, and the lesion to liver parenchyma ratio (LLR) of gallbladder were evaluated and compared. The receiver operating characteristic (ROC) curve and binary logistic regression analysis were used to construct the optimally combined indicator. All six indicators showed high diagnostic accuracy. The layered pattern on DWI and LLR had the highest area under the curve (AUC) value (0.904), followed by the layered pattern on T2WI (0.883), T2WI signal intensity (0.859), ADC value (0.836), and papillary growth (0.796). There was no statistically significant difference in the AUC among indicators for pairwise comparisons. A combination of layered patterns on DWI and papillary growth was shown to be the optimal indicator by binary logistic regression analysis. The AUC value of the combination (0.972) was higher than the layered pattern on DWI (0.904) and papillary growth (0.796) (P < .001). Non-contrast MRI provides several reliable indicators for differentiating benign from malignant gallbladder thickening disease. The combination of layered patterns on DWI and papillary growth is the optimal indicator.Copyright © 2022 the Author(s). Published by Wolters Kluwer Health, Inc.

11 keywords...

hide…