Purpose: To develop a 3-dimensional free-breathing cardiovascular MR technique for quantifying the variation in left ventricular end-diastolic volume (LVEDV) during the respiratory cycle.
Methods: A 3-dimensional radial trajectory with double golden-angle ordering was used for free-running data acquisition during free breathing in healthy volunteers (N = 8). A respiratory self-gating signal was extracted from the center of k-space, and data were retrospectively binned into eight respiratory phases in end-diastole. Three-dimensional image volumes with 2-mm isotropic spatial resolution were reconstructed with conjugate-gradient sensitivity encoding for acquisition durations of 4.5, 9, and 25 min. The LVEDV was manually segmented for each respiratory phase and acquisition duration.
Results: Respiratory-induced variation expressed as minimum LVEDV (during mid-inspiration) relative to maximum LVEDV (during mid-expiration) was 5.9 ± 0.3% for 4.5 min, 5.3 ± 0.4% for 9 min (P = 0.64 versus 4.5 min), and 5.0 ± 0.4% for 25 min (P = 0.25 versus 4.5 min, and P = 1.00 versus 9 min).
Conclusions: The proposed technique enables high spatial-resolution quantification of the respiratory variation in LVEDV during free breathing in under 5 min, and was found to be 5 to 6% for healthy volunteers. Magn Reson Med 79:2693-2701, 2018. © 2017 International Society for Magnetic Resonance in Medicine.
Keywords: 3-dimensional; cardiac cine imaging; free breathing; golden angle; non-Cartesian; respiratory resolved.
© 2017 International Society for Magnetic Resonance in Medicine.