3D Cardiac Imaging Raja Muthupillai, PhD Department of Diagnostic and Interventional Radiology St. Luke s Episcopal Hospital Houston, TX Disclosures Research Support: Philips Healthcare This presentation contains information regarding the use of Gadolinium based CA which is not FDA approved for CV imaging 1
2D Imaging Versus 3D Imaging k y k y k z k x k x 3D Cardiac Imaging : Why? 2D-Imaging 3D-Imaging Scan Time N y x N x TR N y x N x TR x N z SNR/voxel N y Typical Slice Thk > 3 mm < 1 mm (possible) Blood-Muscle CNR High (Inflow effect) FoV sl Dependent 2
3D Cardiac Imaging Morphologic Imaging Functional Imaging Large Coverage Rapid Imaging High Spatial Resolution Coronary MRA High Temporal Resolution Diastolic Function Volumetric Cine Systolic Function Outline A systematic approach toward a coronary MRA protocol Two common approaches for cmra Typical Problems and Solutions 3
Coronary MRA: Requirements Small Tortuous Motion Coronary Flow Epicardial Fat Muscle High Resolution (< mm) Large Coverage Gating (Cardiac/Respiratory) Acq. Duration Blood to Fat Contrast Blood to Muscle Contrast Goals of Coronary MRA Extend Acquisition Time Maximize Blood (Arterial) Contrast 4
I. Respiratory Motion and cmra Breath-holding? Multiple Breathholds? BH duration: 16-22 heartbeats Acq. Duration: 2200-3000 ms Num PEs : 550-750 (TR=4 ms) One or two high resolution slices / breath hold tion of diaphragm Posit BH1BH2 BHn Pencil-Beam Respiratory Motion Tracking RF G 1 G 2 end-inspiration end-expiration Freely Movable Low Flip Angle excitation Real time tracking A(z) z 5
Real-Time Respiratory Tracking/Gating : Accept /Reacquire Algorithm ver Liv Lung Time Navigators extend Acquisition time beyond breathholding capacity II. Contrast Preparation for cmra Epicardial Fat Fat Suppression Cardiac Muscle Arterial Blood M xy /M 0 a.u-> 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 02 0.2 Cardiac Muscle Arterial Blood Venous Blood 0 0 50 100 150 200 250 T 2 Prep Duration (msec) 0.1 Venous Blood 6
Effect of T 2 Preparation: 0.7 Sig gnal Difference (a.u)-> 0.6 0.5 0.4 0.3 02 0.2 Art-Ven Art-Mus 0.1 0 0 50 100 150 200 250 T Prep Duration (msec) 2 Typical cmra Protocol Nav Track 50 ms 30 ms 15 ms 75 ms T2 - prep NAV Fat Sat 3-D TFE / SSFP 7
Targeted cmra protocol Aarhus Berlin Boston Leiden N = 109 AO LCA LV RCA Cologne St. Luke s Leeds Zürich Single Vendor Sub Lingual ISDN Prevalence of disease by invasive angiography > 50% diameter stenosis (%) Overall 59 Left Main 5 Left anterior descending 28 Left circumflex 23 Right coronary 36 One vessel disease 28 Two vessel disease 18 Three vessel disease 12 % LM LAD LCx RCA Any LM / 3 VD Sensitivity 67 88 53 93 93 100 Specificity 90 52 70 72 42 85 Prevalence 89 65 67 80 43 15 PPV 30 56 29 69 70 54 NPV 98 86 86 94 81 100 Kim et al. NEJM, 2001 8
Whole Heart 3D cmra Similar to prior cmra SSFP readout In-plane resolution 1.0 mm x 1.0 mm SENSE factor x 2 CB Higgins et al. MRM 2003;50:1223-1228 Sakuma, et al. JACC, 2006 Sakuma et al. : n = 131 pts; Scan Time: 12.9 ± 4.3 min (5.8 28.8 min) 86% Completion rate; Mean weight: 65 kg 9
Advances in Whole Heart SSFP cmra 1.5 T scanner; 32 Channel coil SENSE factor 2 x 2 Voxel 1 x 1 x 1.5 mm Magnetization prepared, fat sat 3-D SSFP Mean scan time 240 ± 40 secs K Nehrke et al. JMRI 2006;23:752-756 Whole Heart coronary MRA 32 Channel coil/sense factor 4; Scan Time: 5 min @ 70% efficiency Courtesy: Dr. Benjamin Cheong, SLEH. 10
Clinical Applications : Coronary MRA A Ao 3 LA Tissue Characterization Proximal Coronaries Coronary Anomalies Typical Problems and Solutions Significant Motion artifacts in Image Check if Data Acquisition is at correct cardiac phase ms 15 Displacement mm/10 00 10 5 RCA LAD 0 0 200 400 600 800 Trigger Delay Sodickson, et al, ISMRM 1997 Acquire a High Temporal Resolution Cine to select Appropriate cardiac phase 11
Typical Problems and Solutions Patient Falls asleep, and Breathing position has changed Enable Navigator Drift Correction (if available) Speed up Acquisition (Use Parallel Imaging) w/o SENSE: 850 hb with SENSE: 400 hb Typical Problems and Solutions Navigator efficiency too low Enable Motion Adapted Gating (if available) Muthupillai, et al. AJR, 2006 Restrict coverage/use SENSE with SENSE: 410 hb with SENSE+MAG: 360 hb 12
Typical Problems and Solutions Coronary MRA at 3.0T Use TFE readout versus SSFP Sensitive to off-resonance; Long TR due to spatial res demands T2 prep less robust; IR-prep post-contrast Stuber et al. MRM 2002 13
Thank you! 14