Structural and functional evaluation (SaFE)-MRI of peripheral artery disease (PAD) using 3D double-echo steady-state and time-resolved velocimetry
Michael Langham1, Benoit Desjardins1, Erin Englund1, Emile R Mohler, III2, Thomas Floyd3, Jamal J Derakshan1, and Felix W Wehrli1

1Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States, 2Department of Medicine, University of Pennsylvania, Philadelphia, PA, United States, 3Department of Anesthesiology, SUNY Stony Brook, Stony Brook, NY, United States


An alternative unenhanced structural and functional evaluation (SaFE-MRI) MRI is proposed and described for assessing peripheral artery disease (PAD). The SaFE-MRI protocol interleaves 3D double-echo steady-state (DESS) acquisition and velocity mapping at each station. In SaFE-MRI velocity maps in all major conduit arteries (from abdominal aorta to runoff arteries) serve as a “road map” for locating hemodynamically significant stenoses and gray- and black-blood images are used to grade stenoses and detect vascular calcium. Initial results in patients with PAD agree with the finding from CTA and CE-MRA examinations as well as pressure volume recordings.

Introduction and Motivation

Contrast-enhanced angiographic evaluations of MRI and CT are the reference standards for assessing peripheral artery disease (PAD). Because PAD and diabetes often co-exist the prevalence of renal insufficiency is a major challenge to contrast-based angiography [1]. We introduce and describe an alternative unenhanced MRI method to assess PAD by integrating structural (high resolution gray- and black-blood images with double-echo steady-state (DESS) [2]) and functional (velocity waveforms in all of the major conduit arteries) evaluations, which we refer to as SaFE-MRI.


All studies were performed on a 3T system (Siemens TIM Trio) with the manufacturer’s peripheral artery coil system coupled with a body matrix coil. The SaFE-MRI protocol interleaves 3D DESS acquisition and velocity mapping at each station.

Simultaneous structural gray- and black-blood (GB, BB) imaging with water-selective 3D DESS

The water-selective 3D DESS pulse sequence that acquires two Fourier modes of unbalanced SSFP signal, also known as SSFP-FID and SSFP-Echo (Fig 1a). The unbalanced gradient pulse (crusher in Fig 1a) prevents mutual interference of FID and Echo signals. The same crusher also acts as a flow-sensitizing (signal attenuates for moving spins) gradient. As described previously [2] the blood signal is moderate or “gray,” (Fig 1b) on FID images, whereas on Echo images the blood signal is completely suppressed or ”black-blood” (Fig 1c), thereby providing complementary information on vessel patency and lesion characteristics.

Ungated time-resolved blood flow velocity quantification with velocity-encoded projections

Previously developed ungated1D PC-MRI [3] is able to resolve velocity waveforms bilaterally without gating in a fraction of time that is required for cine 2D PC-MRI. Method allows velocity waveforms in all of the major conduit arteries in successive slices spaced 45mm apart, from abdominal aorta down to infrapopliteal arteries in approximately 10 minutes. The basic design principle rests on collection of velocity-encoded projections that are free of background tissue signal (see [3]).

Human Subject Study

After acquiring 3D DESS images and velocity waveforms of the peripheral arteries in healthy subjects, SaFE-MRI was performed on two patients with PAD recently having undergone angiographic evaluation with CTA and CE-MRA, respectively.

3D DESS flip angle 8o+8o, TEFID/TEEcho/TR=2.2/7.8/10.6ms, bandwidth=744Hz/voxel (FID), 223 Hz/voxel (Echo), FOV=352×192×360mm3, spatial resolution=0.78×0.78×3mm3 zero-padded to 0.39×0.39×1.5mm3 , acquisition time (TA)=5.2mins for one station (typically three are needed).

1D PC-MRI flip angle 15o, TE/TR=5.2/10ms, bandwidth=142 Hz/voxel, FOV=352×192×10mm3, spatial resolution=1×1×10mm3, TA~15s/slice.


Fig 2 shows the typical triphasic velocity waveforms in the major segments of peripheral arteries of a healthy subject; in practice the velocity waveform is “recorded” every 45 mm. In contrast, severe PAD leads to monophasic waveforms with reduced peak forward velocity and absence of retrograde and late forward flow (Fig 3). The bilateral monophasic velocity waveforms in the distal external iliac arteries (Fig 3a) suggest hemodynamically significant proximal stenosis even though CTA report states only moderate atherosclerosis within abdominal aorta and minimal narrowing of the distal external iliac artery. However, according to the pulse volume recordings (PVR) report, the pressure waveforms at proximal thigh are abnormal bilaterally, a hallmark of aorto-iliac occlusive disease, thus give credence to the monophasic velocity waveforms. Figs 3b and c show the probable cause of the undetectable blood flow in distal right SFA. Fig 4 further demonstrates the potential clinical utility of velocity waveform analysis. In this patient normal waveforms are observed except for runoff vessels. Absence of flow in ATA and PTA is consistent CE-MRA findings (single run-off bilaterally) and PVR (abnormal at the bilateral ankles). In Fig 5a, the signal void within SFA on BB image suggests partially patent lumen but the same ROI is also hypointense on GB image (Fig 5b), i.e. calcium, which is invisible on MRA, is detectable with DESS as confirmed by CTA, Fig 5c.


SaFE-MRI is a new method for integrated structural and functional assessment of the peripheral vascular system comprising:

· Velocity maps serving as a “road map” for identifying and grading hemodynamically significant stenoses visualized on GB and BB images.

· Unlike cuff-based PVR velocity waveforms above the groin (aortoiliac arteries) and arteries with incompressible medial calcification can also be recorded.

· Method allows detection of vessel-wall calcification (Fig 5) and diagnostic quality images even in the presence of stents [2].

· Obviates both contrast material and exposure to ionizing radiation.


This work was supported by NIH Grants K25 HL111422, R01 HL075649, RO1 HL109545, EE received support from AHA pre-doctoral fellowship and JJD gratefully acknowledges research support from NIH T32 EB004311 Research Track Radiology Residency.


[1] Paraskevas et al., Annals of vascular surgery 2009.

[2] Langham et al, ISMRM 2015, p. 559.

[3] Langham et al, MRM 2010.


Simultaneous acquisition of gray- and black-blood images. a) Water-selective (fat signal attenuated with 1-1 binomial pulse) 3D DESS pulse sequence. The FID is read out with higher bandwidth to optimize Echo signal. Magnified view (inset) of SFA show distinct lumen contrast between b) FID (gray) and c) Echo (black) images.

Velocity waveforms in healthy peripheral arteries (velocity in cm/s vs. time in sec). a) Schematic diagram and b) waveforms in healthy vasculature, characterized by high forward velocity during systole, retrograde flow during early diastole, and slow forward flow during end-diastole. CFA=common femoral, SFA=superficial femoral, PTA=posterial tibial, PA=peroneal, ATA=anterior tibial.

62 year-old female with PAD (ABI, left/right=0.68/0.63). a) Bilateral velocity maps provide evidence of aorto-iliac occlusive disease. b) Sample BB image showing occlusion in medial right SFA (blue arrow, left inset) and patent left SFA (right inset), as confirmed by CTA (right SFA not visible), c).

Bilateral velocity analysis in a patient with a single patent run-off vessel in each leg. In this example, the “velocity report” has the potential to significantly reduce the image reviewing time since the reader only need to focus on the runoff vessels.

Visualization of stenosis and calcification in 66 year-old type-2 diabetic patient (ABI=0.57). a) BB and b) GB images of left SFA indicate calcium (solid arrow) and soft plaque (dashed arrow) confirmed by c) CTA with calcium appearing hypointense due to absence of contrast material.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)