Contrast Enhanced MRA

Wednesday 14 May 2014

Jeffrey H. Maki¹, Toshimasa J. Clark¹, and Gregory J. Wilson^1
1Radiology, University of Washington, Seattle, WA, United States

Image quality and resolution in CE-MRA depend in part on the rate, quantity and type of Gd contrast agent administered. To evaluate different injection strategies, a physiologic model of bolus delivery from the antecubital vein to the artery of interest was used to predict contrast agent concentration vs. time. Contrast agent relaxivities (T1 and T2*) in oxygenated whole blood were then used to predict signal intensity modulation during the CE-MRA acquisition, which ultimately leads to blurring in CE-MRA. Blurring was evaluated in image space for varying injections using modulation transfer function analysis.

Stanislas Rapacchi¹, Yutaka Natsuaki², Gerhard Laub², John Paul Finn¹, and Peng Hu^1
1Radiology, UCLA, Los Angeles, California, United States, ²Siemens Healthcare, Los Angeles, California, United States

In this work, we aim to study the temporal and spatial blurring effect of view-sharing on dynamic CE-MRA using phantom and in vivo clinical data and its impact on the assessment of blood vessels. Furthermore we propose to reduce artifacts and improve the accuracy of dynamic CE-MRA by limiting the use of view-sharing and reconstruct under-sampled k-spaces with compressed sensing enhanced with parallel imaging.

Mahdi Salmani Rahimi¹, James H Holmes², Kang Wang², Alejandro Roldan³, and Frank Korosec^3,4
1Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States, ²Global MR Applications and Workflow, GE Heatlhcare, Madison, WI, United States, ³Radiology, University of Wisconsin-Madison, Madison, WI, United States, ⁴Medical Physics, University of Wisconsin-Madison, Madison, WI, United States

Dual-echo fat-water separated techniques have been recently used for MR angiography of the lower extremities. Bipolar readout gradients regularly used in multi-echo acquisitions can cause an extra phase shift in flowing spins that is not typically accounted for in fat-water signal separation models. This work investigates the effects of these readout gradients on the flowing spins using a controlled flow pump and a stenosis phantom. Results from the phantom experiment show the degree of stenosis may be overestimated at high flow velocities due to water inappropriately being mapped to the fat image when using bipolar readout gradient acquisitions. Non-bipolar readout gradients are shown to mitigate these effects.

J. Paul Finn¹, Sarah N Khan¹, Aarti Luhar¹, Theodore Hall¹, Stanislas Rapacchi¹, Fei Han¹, Peng Hu¹, Yutaka Natsuaki², and Isidro Salusky^3
1Radiology, UCLA, Los Angeles, California, United States, ²Siemens Healthcare, California, United States, ³Nephrology, UCLA, California, United States

Purpose: We report our initial findings with ferumoxytol for non-Gd CEMRA at 3.0T in children with renal failure. Methods: 9 patients aged 6 days to 14 years were studied on a Siemens TIM Trio system. Multiple CEMRA phases were acquired up to 30 minutes following injection and measurements of SNR and CNR in the thoracic aorta and inferior vena cava (IVC) were recorded at each phase. Phantom measurements of T1 and T2* relaxivity were made at 3.0T. Results: The T1 relaxivity was 9.0 mM-1s-1 and the T2 relaxivity was 90 mM-1s-1. CNR measurements in patients confirmed that the intravascular signal in the ferumoxytol group remained high and stable to the last measurements, up to 35 minutes post injection, whereas the signal in the control group fell off with time. Conclusion and Discussion: Initial results with ferumoxytol are highly encouraging for CEMRA in children with renal failure.

Jianing Pang^1,2, Qi Yang³, Kuncheng Li³, Yi He⁴, Zhanming Fan⁴, Bin Sun⁵, Fabio S Raman⁶, Mark A Ahlman⁶, David A Bluemke^6,7, Jing An⁸, Xiaoming Bi⁹, Daniel S Berman², and Debiao Li^2,10
1Radiology and Biomedical Engineering, Northwestern University, Chicago, IL, United States, ²Biomedical Imaging Research Institiute, Cedars-Sinai Medical Center, Los Angeles, CA, United States, ³Department of Radiology, Xuanwu Hospital of Capital Medical University, Beijing, China, ⁴Department of Radiology, Anzhen Hospital of Capital Medical University, Beijing, China, ⁵Fujian Medical University Union hospital, Fuzhou, Fujian, China, ⁶Radiology and Imaging Sciences, National Institutes of Health Clinical Center, Bethesda, MD, United States, ⁷Molecular Biomedical Imaging Laboratory, National Institutes of Biomedical Imaging and Bioengineering, Bethesda, MD, United States, ⁸MR Collaborations NE Asia, Siemens Healthcare, Beijing, China, ⁹MR R&D, Siemens Healthcare, Los Angeles, CA, United States, ¹⁰Bioengineering, University of California, Los Angeles, CA, United States

We have developed a 3DPR based contrast-enhanced coronary MRA technique that delivers good image quality at (1.0 mm)³ spatial resolution with scan time of 5 minutes. Further investigations are warranted on subjective image quality evaluation, protocol optimization, and tests on patient population with suspected CAD.

Peter Bannas^1,2, Mark L Schiebler¹, Utaroh Motosugi¹, Christopher J Francois¹, Scott B Reeder^1,3, and Scott K Nagle^1,3
1Department of Radiology, University of Wisconsin, Madison, Madison, WI, United States, ²Department of Radiology, University Medical Center Hamburg-Eppendorf, Hamburg, Hamburg, Germany, ³Department of Medical Physics, University of Wisconsin, Madison, Madison, WI, United States

The aim of our study was to establish a quantitative and objective approach to differentiate Gibbs artifact from true pulmonary emboli, in order to improve the diagnostic performance of pulmonary MRA for diagnosis of pulmonary embolism. The percentage of signal loss between the vessel lumen and the central dropout was calculated on both first-pass and steady state MRA from 65 signal drops in 28 patients. Gibbs artifacts revealed a significantly lower signal drop as compared to pulmonary embolism. ROC analyses suggest an optimum threshold value of 53% (first-pass) and 42%-signal drop (steady state) to differentiate between Gibbs artifact and PE.

Sébastien Roujol¹, Murilo Foppa¹, Tamer A. Basha¹, Mehmet Akçakaya¹, Kraig V Kissinger¹, Beth Goddu¹, Sophie Berg¹, Warren J. Manning^1,2, and Reza Nezafat^1
1Department of Medicine, Beth Israel Deaconess Medical Center / Harvard Medical School, Boston, MA, United States, ²Department of Radiology, Beth Israel Deaconess Medical Center / Harvard Medical School, Boston, MA, United States

Contrast-enhanced pulmonary vein MR-angiography (CE-PV MRA) is commonly performed before and after pulmonary vein isolation procedures to assess PV anatomy to detect potential post-procedural complications such as PV stenosis. CE-PV MRA is clinically acquired within a prolonged breath-hold at contrast arrival in the PVs. This sequence is not ECG triggered and can lead to motion-induced blurring artifacts and over-estimation of the PV size. In addition, this sequence requires an accurate initiation at contrast arrival in the PVs which may fail in some patients. Therefore, the development of improved PV-MRA protocol is desirable. In this study, we sought to investigate the feasibility of a highly-accelerated ECG-triggered CE-PV MRA with isotropic spatial resolution using compressed sensing.

Adrian Lam¹, Luis F Mora-Vieira², Michael Lloyd², and John N Oshinski^2
1Georgia Institute of Technology, Atlanta, GA, United States, ²Emory University, GA, United States

In Cardiac Resynchronization Therapy (CRT), the location of the left ventricular (LV) lead plays an important role in dictating patient benefit. The LV lead is delivered through the coronary veins and should be ideally implanted at the latest contracting site that is not predominantly myocardial scar. Since the coronary veins are not imaged until CRT procedure, it is unknown whether a desired implantation location has coronary vein access. The objective of this study is to evaluate the potential of MRI to image the coronary vein branch used for LV lead implantation with validation against the gold standard, retrograde x-ray venography.

Yutaka Natsuaki¹, J Paul Finn², Randall Kroeker³, and Gerhard Laub^4
1Siemens Healthcare, Los Angeles, CA, United States, ²Radiology, UCLA, Los Angeles, CA, United States, ³Siemens Healthcare, Winnipeg, MB, Canada,⁴Siemens Healthcare, San Francisco, CA, United States

AIn thoracic ECG-gated CEMRA, the steady-state triggering plays a key role in maintaining the magnetization and minimizing cardiac pulsatile motion. The current gated CEMRA assumes steady R-R interval with the fixed scan window (= trigger delay + segment acquisition). When R-R interval is disrupted due to heart rate variation and ECG failure, scan time substantially increases, which causes problems in timing-sensitive CEMRA scans. The current study introduces the Adaptive Steady-State Triggering, a novel prospective trigger adjustment approach to compensate the early trigger loss and to force the late trigger scan completion within an acceptable range.

Gabriel Acevedo-Bolton¹, Farshid Faraji¹, and David Saloner^1,2
1Radiology and Biomedical Imaging, UCSF, San Francisco, CA, United States, ²VAMC San Francisco, CA, United States

Image quality in patients undergoing CE-MRA is sensitive to correct synchronization between imaging initiation and the contrast bolus. A patient specific flow model was used to study the effects of scan initiation relative to contrast injections on 3D CE-MRA images. We found unless the center of k-space was in after the contrast upswing poor image quality resulted. With the high temporal resolution available, we were able to resolve the transport of contrast into the model. More importantly, the use of a flow model provides a robust and repeatable test bed to study timing effects.