Quantification of turbulence and velocity in stenotic flow using spiral three-dimensional phase-contrast MRI

Abstract

Purpose: Evaluate spiral three-dimensional (3D) phase contrast MRI for the assessment of turbulence and velocity in stenotic flow.

Methods: A-stack-of-spirals 3D phase contrast MRI sequence was evaluated in vitro against a conventional Cartesian sequence. Measurements were made in a flow phantom with a 75% stenosis. Both spiral and Cartesian imaging were performed using different scan orientations and flow rates. Volume flow rate, maximum velocity and turbulent kinetic energy (TKE) were computed for both methods. Moreover, the estimated TKE was compared with computational fluid dynamics (CFD) data.

Results: There was good agreement between the turbulent kinetic energy from the spiral, Cartesian and CFD data. Flow rate and maximum velocity from the spiral data agreed well with Cartesian data. As expected, the short echo time of the spiral sequence resulted in less prominent displacement artifacts compared with the Cartesian sequence. However, both spiral and Cartesian flow rate estimates were sensitive to displacement when the flow was oblique to the encoding directions.

Conclusion: Spiral 3D phase contrast MRI appears favorable for the assessment of stenotic flow. The spiral sequence was more than three times faster and less sensitive to displacement artifacts when compared with a conventional Cartesian sequence.

Keywords: 4d flow; phase contrast mri; spiral; stenosis; turbulence mapping.

Figure 1

The velocity (a) and turbulent kinetic energy (b) along the centerline of the phantom together with cross‐sectional images (c) of the turbulent kinetic energy for Reynolds number 1000 (left) and 2000 (right) from Spiral and Cartesian 3D PC‐MRI, for two different orientations, coronal (COR) and transverse (TRA), as well as CFD data. The VENC was 35 and 10 cm/s for the velocity and turbulence mapping, respectively. X and Y denote distance from the center of the stenosis normalized by the unconstructed pipe diameter (14.6 mm). The principal flow direction is in the positive X‐direction.

Figure 2

a: Plots of the velocity along the centerline of the phantom of the increased‐flow case (left) and stenotic‐flow case (right), respectively. Both Cartesian and spiral velocity data from the coronal (COR) and transverse (TRA) orientations are shown. The principal flow direction is in the positive X‐direction. Images of magnitude (b) and speed (c) from the oblique (OBL) orientation of the increased‐flow case (left) and stenotic‐flow case (right). The VENC was 150 and 300 cm/s for the increased‐flow case and stenotic‐flow case, respectively. X denotes the distance from the center of the stenosis normalized by the unconstructed pipe diameter (14.6 mm). For the OBL orientation, frequency and slice encoding was carried out along the vertical and horizontal axis, respectively.

Figure 3

The volume flow rate from the increased‐flow case (a) and the stenotic‐flow case (b) for the different orientations. The VENC was 150 and 300 cm/s for these two flow cases, respectively. X denotes position of the cross sectional plane from which the flow rate was computed and is the distance from the center of the stenosis normalized by the unconstructed pipe diameter (14.6 mm). The principal flow direction is in the positive X‐direction. Nominal flow rate should be 56 mL/s in the increased‐flow case and 112 mL/s in the stenotic‐flow case, for all planes.