Author + information
Background: Increased body motion, respiratory drive, and heart rate complicate cardiac MRI during in-scanner exercise. Real time (RT) imaging, with the benefit of shorter acquisition time, has been used to circumvent some of these issues. Re-binning reconstruction of RT data from multiple cardiac cycles using respiratory motion correction offers better reconstructed temporal resolution and signal to noise. This study compares feasibility of three RT imaging techniques during active exercise using an MRI-compatible bike ergometer.
Methods: RT acquisitions with linear (L), non-linear (NL) and re-binning (RB) reconstructions were collected across a range of heart rates, with exercise attempted at both moderate and strenuous levels. All techniques used an SSFP sequence with flip angle 50, FOV 360×270mm2, matrix 160×80, slice 8mm, TR 2.47ms, TE 1.04, parallel imaging factor 4. L used GRAPPA reconstruction with no partial Fourier (frame duration = 49.4ms). NL used 5/8 partial Fourier coverage in the phase encoding direction (frame duration = 32.1ms). RB acquired 8s of data per slide and was retro-gated to 30 phases per cardiac cycle using motion correction and non-linear reconstruction. Acquisition and processing times were recorded. Left ventricular end-diastolic (ED) and end-systolic (ES) volumes were measured using the summation of disks technique. Intraclass correlation coefficients were calculated to determine agreement of measurements among RT techniques.
Results: Twenty-five normal volunteers were enrolled: 72% female, age 10-59 (avg 23) years. Thirty-two full datasets were collected, with heart rate range 86-140 (avg 110) bpm. Intraclass correlation coefficients showed excellent agreement (ED = 0.955, ES = 0.892). Acquisition/processing times were L = 40/8, NL = 35/100, RB = 120/17s.
Conclusions: Exercise MRI using L, NL, and RB motion corrected RT imaging is feasible and all techniques yield similar ED and ES measurements. Trade off of the shorter acquisition time of traditional RT imaging should be weighed against improved temporal resolution and signal to noise of RB. Further work will focus on optimizing patient comfort and ECG fidelity while producing reliable image quality.
Poster Hall, Hall C
Friday, March 17, 2017, 10:00 a.m.-10:45 a.m.
Session Title: Imaging the Athlete's Heart
Abstract Category: 31. Non Invasive Imaging: Sports and Exercise
Presentation Number: 1119-223
- 2017 American College of Cardiology Foundation