Exploratory 4D Flow MRI Study of Portal Venous Rotational Flow for High‐Risk Esophageal Varices in Cirrhosis

ABSTRACT Background Bleeding risk assessment of esophageal varices (EVs) is commonly performed in liver cirrhosis. However, existing approaches are often invasive or insufficient to fully capture bleeding risk. Purpose To investigate portal venous hemodynamics and vessel geometry associated with EVs severity in cirrhotic patients using four‐dimensional flow MRI as a noninvasive approach to support clinical risk assessment. Study Type Retrospective. Population One‐hundred and four cirrhotic patients consisting of patients without EVs (Group 1, n = 48), with low‐risk EVs (Group 2, n = 37), and high‐risk EVs (Group 3, n = 19). Field Strength/Sequence 3 T, a three‐directional, velocity‐encoded gradient echo sequence with Cartesian k‐space sampling. Assessment Flow rate, mean velocity, and fractional flow changes in main portal vein (PV); effective vessel diameter and vessel angle; and rotational flow including vorticity, helicity, helical flow intensity ( h 2 ), localized normalized helicity, and vortex volume. Statistical Tests One‐way ANOVA with Tukey's post hoc test or Kruskal–Wallis followed by Conover's post hoc test. Receiver operating characteristic (ROC) curve analysis. Significance level set at p < 0.05. Results Significant group differences were observed for flow rate at the proximal PV ( ε 2 = 0.05) and for mean PV velocity at both proximal ( ε 2 = 0.13) and distal locations ( η 2 = 0.11). In contrast, neither fractional portal venous flow nor vessel angle measurements differed across groups ( p = 0.294 and all p ≥ 0.475, respectively). Among rotational flow biomarkers, vorticity differentiated Group 3 from both Groups 1 and 2 (Group 1: 17.86 ± 3.26 s −1 , Group 2: 16.65 ± 4.58 s −1 , Group 3: 13.59 ± 3.23 s −1 ; η 2 = 0.15), with an area under the ROC curve of 0.77 (sensitivity = 0.79, specificity = 0.62). Data Conclusion Rotational flow biomarkers may support current risk assessment in cirrhotic patients with high‐risk EVs. Evidence Level 4. Technical Efficacy Stage 2.