Optimal Plane Selection for Measuring Post-prandial Blood Flow Increase within the Superior Mesenteric Artery: Analysis Using 4D Flow and Computational Fluid Dynamics

Abstract

Purpose: 2D cine phase contrast (PC)-MRI is a standard velocimetry for the superior mesenteric artery (SMA); however, the optimal localization of the measurement plane has never been fully discussed previously. The purpose of this Institutional Review Board approved prospective and single arm study is to test whether flow velocimetry of the SMA with combined use of 2D cine PC-MRI and meal challenge is dependent on the localizations of the measurement planes and to seek optimal section for velocimetry.

Methods: Seven healthy volunteers underwent cardiac phase resolved ECG gated 2D cine PC-MRI pre- and 30 min post-meal challenge at three measurement planes: proximal, curved mid section and distal straight section of the SMA at 3T. 4D Flow using 3D cine PC-MRI with vastly undersampled isotropic projection imaging (PC VIPR) was also performed right after 2D cine PC-MRI to delineate the flow dynamics within the SMA using streamline analysis. Two radiologists measured flow velocities, and rated the appearances of the abnormal flow in the SMA on streamlines derived from the 4D Flow and the computational fluid dynamics (CFD).

Results: 2D cine PC-MRI measured increased temporally averaged flow velocity (mm/s) after the meal challenge only in the proximal (129.3 vs. 97.8, P = 0.0313) and distal section (166.9 vs. 96.2, P = 0.0313), not in the curved mid section (113.1 vs. 85.5, P = 0.0625). The average velocities were highest and their standard errors (8.5-26.5) were smallest at the distal straight section both before and after the meal challenge as compared with other sections. The streamline analysis depicted more frequent appearances of vertical or helical flow in the curved mid section both on 4D Flow and CFD (κ: 0.27-0.68).

Conclusion: SMA velocimetry with 2D cine PC-MRI was dependent on the localization of the measurement planes. Distal straight section, not in the curved mid section is recommended for MR velocimetry.

Keywords: 4D Flow; flow measurement; phase contrast image; post-prandial hyperemia; superior mesenteric artery.

Fig. 1

Three measurement planes were set perpendicular to the superior mesenteric artery (SMA). The scheme shows Pro: proximal section, CurvMid: curved mid section, DistStra: distal straight section on the MR angiography of the upper abdomen.

Fig. 2

The representative ratings for an individual on 4D Flow. (a) The representative ratings for flow disturbances assessed on 4D Flow. Proximal section was rated 0 (laminar), mid section 1 (mild vortex and/or helical flow) and distal section: 0 (laminar). (b) Proximal section: rated 1 (mild vortex and/or helical flow), mid section: 2 (severe vortex and/or helical flow), and the distal section: rated 1 (mild vortex and/or helical flow). The representative ratings for 2 different individuals on computational fluid dynamics (CFD). (c) The representative ratings for flow disturbances assessed on CFD. Proximal section was rated 1 (mild vortex and/or helical flow), mid section 2 (severe vortex and/or helical flow) and distal section: 1 (mild vortex and/or helical flow). (d) Proximal section: rated 0 (laminar), mid section: 1 (mild vortex and/or helical flow), and the distal section: rated 1 (mild vortex and/or helical flow).

Fig. 3

Graphs show phase resolved blood flow velocity measured at each plane with 2D cine PC MRI before and after meal challenge (Averaged values). Most of the significant increase of the flow velocities at each phase points are detected mainly at distal straight section, whereas least increase are seen at the curved mid section. ^*Significant increase of flow velocity (P < 0.05).

Fig. 4

Representative streamline in the superior mesenteric artery (SMA) at systole (a: left) and diastole (b: right) after meal challenge depicted by 4D Flow. Helical and/or vortex flow was dominant at the curved mid section at systole and diastole, whereas mostly laminar flow was observed at the proximal and distal straight section.

Fig. 5

The same trend of the flow dynamics within the superior mesenteric artery (SMA) was observed based on the computational fluid dynamics streamline analysis for the SMA model derived from the same volunteer as shown in Fig. 4 depicted at systole (a: left) and diastole (b: right) after meal challenge.

Fig. 6

The graphs show the average grades (0: laminar flow, 1: mild vortex and/or helical flow and 2: severe vortex and/or helical flow) of the abnormal flow dynamics rated by two observers at each section. Note highest frequencies of the appearances of the abnormal flow dynamics were at the curved mid section both before and after the meal challenge irrespective of cardiac phases. CFD, computational fluid dynamics.