Abstract
Introduction Clinically significant cardiac valve pathology causes increased blood flow velocities across the regurgitant or stenotic valve orifice. Like echocardiography, cardiac magnetic resonance imaging (cMRI) can be reliably used for blood flow quantification to assess the severity of valvular heart lesions. Clinical scans can be limited by either long patient breath hold times or insufficient temporal resolution. New MRI parallel imaging techniques, such as iPAT (GRAPPA), use the inherent properties of the receiving coil to accelerate image acquisition without sacrificing temporal resolution. We compared the reproducibility of peak and mean blood flow velocities using cMRI with and without GRAPPA.
Methods Blood flow velocities across the mitral and aortic valves and the left ventricular outflow tract were measured in 7 research volunteers using a Siemens 1.5 T scanner. 3 subjects were referred for evaluation of known valvular disease, 3 for adenosine stress MRI and 1 subject for viability assessment. Regions of interest (ROI) were measured in a total of 119 individual frames and peak and mean flow velocities were determined using an on-site Siemens software package. Measurements were obtained during sequential breath-holds with and without GRAPPA. These data were compared to phantom studies with velocities obtained through a rigid, stationary tube at a flow rate of 2 L/min using similar imaging parameters with and without GRAPPA.
Results Peak and mean velocities ranged from 21 cm/sec to 444 cm/sec and from 12 cm/sec to 306 cm/sec, respectively. In the research volunteers, peak flow measurements with GRAPPA correlated strongly to the values obtained without GRAPPA (Pearson's correlation coefficient, r = 0.999; p<0.01). Mean velocities showed a similar relationship (correlation coefficient, r = 0.998, p<0.01). Breath hold times were reduced by 42%, from an average of 23.9 ± 2.3 secs to 14.1 ± 1.4 secs. In the flow phantom, the average velocities at a flow rate of 2.0 L/min, with and without GRAPPA (R = 2) were 50.9 cm/sec and 51.9 cm/sec (percent difference: 1.9%).
Conclusions cMRI for flow quantification using GRAPPA can be reliably used in the measurement of peak and mean blood flow velocities. Using GRAPPA, peak and mean valve velocities are reproducible and offer greater than 40% reduction in breath hold times. This substantial decrease in acquisition time makes a comprehensive evaluation in patients with significant cardiac disease more feasible.