• Zhouchao Hu, Nannan Yu, Heping Wang, Shibo Li, Jingang Yan, and Guoqiang Zhang.
• Interventional Diagnosis and Treatment Center.
• Medicine (Baltimore). 2020 Dec 24; 99 (52): e23924.

AbstractTo investigate whether MRI features could preoperatively predict local tumor progression (LTP) in patients with hepatocellular carcinoma (HCC) treated with radiofrequency ablation (RFA) as the first-line treatment and improve a novel predictive model through developing a nomogram including various conventional MRI parameters. 105 patients with HCCs who had received RFA were enrolled. All patients had undergone conventional MRI before RFA. Uni- and multivariable analyses for LTP were assessing using a Cox proportional hazards model. The developed MRI-based nomogram was further designed based on multivariable logistic analysis in our study and the usefulness of the developed model was validated according to calibration curves and the C-index. Rim enhancement (hazard ratio: 2.689, P = .044) and the apparent diffusion coefficient (ADC) values (hazard ratio: 0.055, P = .038) were statistically significant independent predictors of LTP after RFA at multivariable analysis. The performance of the nomogram incorporating two MRI parameters (with a C-index of 0.782) was improved compared with that based on rim enhancement and ADC alone (with C-index values of 0.630 and 0.728, respectively). The calibration curve of the MRI-based nomogram showed good conformance between evaluation and observation at 0.5, 1, and 1.5 years after RFA. The preliminary predictive model based on MRI findings including rim enhancement and ADC value could be used preoperatively to estimate the risk of LTP of HCC after RFA as the first-line treatment.Copyright © 2020 the Author(s). Published by Wolters Kluwer Health, Inc.

34 keywords...

hide…