Purpose or Case Report: 4D flow MRI with full coverage of the heart and proximal vessels is a powerful tool to understand hemodynamic pathology arising from congenital heart disease (CHD) and subsequent surgery; however, scan times often exceed 10 minutes thereby limiting widespread adoption in pediatrics. Decreasing scan time using compressed-SENSE (CS), a combination of conventional parallel imaging (SENSE) and randomized spatial undersampling, shows promise over SENSE alone. However, both the feasibility and its effect on accurate flow quantification in pediatrics has not been evaluated. This study investigated the feasibility of CS for pediatric 4D flow MRI to quantify flow indices used in risk stratification of CHD such as coarctation or pulmonary regurgitation, in comparison with SENSE.
Methods & Materials: 15 patients (15.0±7.3y) with CHD were enrolled. All patients’ images were acquired on a Philips Ingenia 3T with CS=6 and separately with SENSE= 2 (phase encode) x 2 (slice), with online reconstruction. Images were processed to remove noise and velocity aliasing. To quantify blood flow, 4 anatomical landmarks were defined for the aorta and 3 were defined for the pulmonary arteries. Peak velocity, systolic flow rate, and net flow rate were computed at each landmark. In addition, based on this data, Qp/Qs was derived. For patients with pulmonary regurgitation or aortic coarctation, regurgitant fraction and maximum peak velocity at the coarctation were quantified. A two-tailed, paired t-test was used for statistical analysis.
Results: Of 15 patients, 3 patients had hemi-Fontan circulation or uncorrectable severe aliasing in their images and pulmonary measurements were not included. 2 patients had coarctation and 3 patients had pulmonary regurgitation. CS achieved a reduction in scan time of 30.2% over the SENSE approach (3:47±1:07mins vs. 5:25±1:32mins, respectively). No statistical differences were found for peak velocity, peak flow rate, and net flow (p > 0.08 for all measurements). Qp/Qs showed no statistical differences (p=0.14) with mean values of 1.12±0.24 (CS) and 1.05±0.24 (SENSE). Coarctation velocities were equivalent with a mean difference of 5.6% (subject 1: 1.56m/s vs. 1.48m/s; subject 2: 2.60m/s vs. 2.46m/s; CS vs. SENSE). Regurgitant fraction had a mean difference of 1.9% (23.3±4.6% vs. 21.4±8.1% CS vs. SENSE).
Conclusions: CS 4D flow can measure whole heart pediatric hemodynamics in less than 4 minutes, suggesting CS can be implemented for pediatric patients in a clinically relevant time window.