Liver cirrhosis is a common condition that can lead to portal hypertension, formation of portosystemic shunts, and ultimately an array of complications. 4D flow MRI enables non-invasive characterization and quantification of such altered hemodynamics, which could identify patients at risk of cirrhosis-associated complications before they become clinically apparent. However, abdominal 4D flow MRI is currently limited by long acquisition times required to assess the wide range of velocities throughout the abdominal vasculature. Previously we have relied on separate 4D flow acquisitions with low and high velocity encoding gradients (venc) to characterize arterial and portal venous flow¹, but we hypothesize that similar abdominal flow data could be obtained in less time with a single 4D flow MRI acquisition with two different velocity encoding gradients (dual-venc 4D flow MRI).20 prospectively recruited patients (age=55±8yrs, 7 women) with liver cirrhosis and sequelae of portal hypertension (splenomegaly and/or portosystemic shunts) on diagnostic imaging fasted prior to undergoing non-contrast 4D flow MRI at 3T (MAGNETOM Skyra, Siemens Medical Systems, Erlangen, Germany). N=10 patients underwent separate low (50 cm/s) and high (100 cm/s) venc acquisitions with ECG- and respiratory navigator gating in an oblique axial imaging volume to include the hepatic and splenic vasculature with the navigator positioned at the lung-spleen interface. The other half underwent dual-venc 4D flow MRI² with vencs of 55/56 and 110/120 cm/s and navigator gating in an oblique coronal imaging volume. Demographics and pulse sequence parameters are provided in Figure 1. Scan data was first corrected for background phase offset errors in MatLab (the MathWorks, Natick, MA, USA). Anti-aliasing was performed only for high venc data. A PC-MRA was generated to enable 3D segmentation of the vasculature (Mimics, Materialise, Plymouth, MI) that was used to restrict the velocity field for visualization and quantification (Ensight, CEI, Apex, NC, USA) (Figure 2). High- and low-venc data were used to assess the arterial and venous systems, respectively³. Acquisition times were obtained from a single patient where both methods were applied. Flow quantification was compared using relative error in flow input and output in 3 areas: proximal v. distal main portal vein ± coronary vein, portal venous confluence, and branching of the celiac trunk.The dual-venc 4D flow MRI acquisition required 58% less scan time than two separate venc acquisitions (12min 44sec v. 30min 24sec); however, sequence parameters impacting scan time were not matched as the dual-venc should theoretically only be 25% faster². Post-processing required 18% less time for the dual-venc acquisitions (298min v. 364min). Both methods provided similar venous flow quantification with no significant difference in portal flow relative error [9±8% (Dual-venc) v. 13±8% (Single venc), p = 0.15]. See Figure 3 for single venc example and Figure 4 for dual-venc example. Arterial flow quantification was also comparable between the two methods [19±13% (Dual-venc) v. 10±7% (Single venc), p = 0.16)].The novel dual-venc 4D flow MRI sequence achieved comparable flow quantification accuracy as separate low- and high-venc acquisitions with 58% less scan time. Post-processing time was also reduced because the high and low venc data sets of the dual-venc acquisition are perfectly registered, allowing the same vessel segmentation to be used for both low and high-venc flow quantification. However, the non-significant trend towards higher relative error in arterial flow observed with the dual-venc sequence suggests future work is necessary to balance reductions in scan time with tradeoffs in temporal and spatial resolution to fully capture pulsatile arterial flow. Although we processed high and low venc data separately for the dual-venc acquisitions, these data sets could be combined in the future, using high venc data solely to estimate the aliasing in the low venc acquisition². This would allow a large range of velocities to be assessed while improving the velocity to noise ratio (VNR), addressing the principle limitation seen with the current approach. Overall, these results illustrate a critical step toward pairing the unparalleled clinical information provided by 4D flow MRI with an acquisition time more appropriate for routine clinical use.4D flow MRI has the potential to improve characterization of altered abdominal hemodynamics in liver cirrhosis and portal hypertension, and a dual-venc sequence can capture a wide range of velocities with reduced acquisition and post-processing times compared to single venc approaches. Furthermore, the processing of dual-venc data sets could be combined in the future to capture a large range of velocity data with a higher VNR. This is a critical advancement as shorter scan time and streamlined post-processing are important first steps to move 4D flow MRI into the realm of clinical feasibility.