ORIGINAL RESEARCH

published: 09 June 2022

doi: 10.3389/fvets.2022.819627

Evaluation of the Abdominal Aorta

and External Iliac Arteries Using
Three-Dimensional Time-of-Flight,
Three Dimensional
Electrocardiograph-Gated Fast
Spin-Echo, and Contrast-Enhanced
Magnetic Resonance Angiography in
Clinically Healthy Cats
Minju Lee 1 , Minjung Ko 1 , Jisoo Ahn 1 , Jiyoung Ahn 1 , Jin Yu 2 , Jinhwa Chang 3 ,
Sukhoon Oh 4 and Dongwoo Chang 1*
1
Edited by: Section of Medical Imaging, Veterinary Medical Center, College of Veterinary Medicine, Chungbuk National University,
Haney Samir, Cheongju, South Korea, 2 College of Veterinary Medicine, Western University of Health Sciences, Pomona, CA,
Cairo University, Egypt United States, 3 Korea Animal Medical Center, Cheongju, South Korea, 4 Bio-Chemical Analysis Team, Korea Basic Science
Institute, Daejeon, South Korea
Reviewed by:
Katsuhiro Matsuura,
VCA Japan Shiraishi Animal Arterial thromboembolism is associated with high morbidity and mortality rates
Hospital, Japan
Ahmed S. Mandour,
in cats. Definitive diagnosis requires advanced imaging modalities, such as
Suez Canal University, Egypt computed tomography angiography (CTA) and contrast-enhanced (CE) magnetic
*Correspondence: resonance angiography (MRA). However, CTA involves exposure to a large amount
Dongwoo Chang of ionized radiation, and CE-MRA can cause systemic nephrogenic fibrosis.
[email protected]
Non-contrast-enhanced (NE) MRA can help accurately diagnose vascular lesions
Specialty section: without such limitations. In this study, we evaluated the ability of NE-MRA using
This article was submitted to three-dimensional electrocardiograph-gated fast spin-echo (3D ECG-FSE) and 3D
Veterinary Imaging,
a section of the journal time-of-flight (3D TOF) imaging to visualize the aorta and external iliac arteries in
Frontiers in Veterinary Science clinically healthy cats and compared the results with those obtained using CE-MRA.
Received: 22 November 2021 All 11 cats underwent 3D ECG-FSE, 3D TOF, and CE-MRA sequences. Relative
Accepted: 11 May 2022
signal intensity (rSI) for quantitative image analysis and image quality scores (IQS)
Published: 09 June 2022
for qualitative image analysis were assessed; the rSI values based on the 3D TOF
Citation:
Lee M, Ko M, Ahn J, Ahn J, Yu J, evaluations were significantly lower than those obtained using 3D ECG-FSE (aorta
Chang J, Oh S and Chang D (2022) 3D TOF: 0.57 ± 0.06, aorta 3D ECG-FSE: 0.83 ± 0.06, P < 0.001; external iliac
Evaluation of the Abdominal Aorta and
External Iliac Arteries Using
arteries 3D TOF: 0.45 ± 0.06, external iliac arteries 3D ECG-FSE:0.80 ± 0.05, P <
Three-Dimensional Time-of-Flight, 0.001) and similar to those obtained using CE-MRA (aorta: 0.58 ± 0.05, external
Three Dimensional
iliac arteries: 0.57 ± 0.03). Moreover, IQS obtained using 3D TOF were significantly
Electrocardiograph-Gated Fast
Spin-Echo, and Contrast-Enhanced higher than those obtained using 3D ECG-FSE (aorta 3D TOF: 3.95 ± 0.15, aorta
Magnetic Resonance Angiography in 3D ECG-FSE: 2.32 ± 0.60, P < 0.001; external iliac arteries 3D ECG-FSE: 3.98 ±
Clinically Healthy Cats.
Front. Vet. Sci. 9:819627.
0.08, external iliac arteries 3D ECG-FSE: 2.23 ± 0.56, P < 0.001) and similar to those
doi: 10.3389/fvets.2022.819627 obtained using CE-MRA (aorta: 3.61 ± 0.41, external iliac arteries: 3.57 ± 0.41).

Frontiers in Veterinary Science | www.frontiersin.org 1 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

Thus, 3D TOF is more suitable and produces consistent image quality for visualizing the
aorta and external iliac arteries in clinically healthy cats and this will be of great help in
the diagnosis of FATE.

Keywords: abdominal aorta, magnetic resonance angiography (MRA), arterial thromboembolism, feline, external
iliac arteries

INTRODUCTION studies have evaluated the aorta and external iliac arteries using
NE-MRA to confirm thrombi in patients with FATE.
Feline arterial thromboembolism (FATE) is a condition Electrocardiograph-gated three-dimensional fast spin-echo
associated with high morbidity and mortality rates that (3D ECG-FSE) sequence has been most frequently used during
is most commonly caused by underlying cardiomyopathy routine clinical examinations to evaluate thrombi and stenosis in
(1–4). The most frequent site of FATE is the terminal the lower extremity arteries of humans. The T2-weighted 3D FSE
aorta (saddle thrombus) (5). Clinical signs are insufficient technique using partial Fourier acquisition is based on cardiac
to produce a definitive diagnosis of FATE. A history of gating with either ECG or peripheral pulse gated to the image
predisposing conditions might support diagnosis, but FATE arterial bright blood, with MR using the difference in arterial flow
should be differentiated from other diseases such as spinal velocity during the systole and diastole (16–18). 3D ECG-FSE has
cord diseases, including intervertebral disk disease, neoplasia, been further improved to obtain a better display of slow-flowing
and embolism, and acute intracranial disorders, including blood, allow better visualization of collateral arteries, and reduce
embolism, trauma, and toxicity, that also cause acute loss of motion artifacts (10, 19, 20).
lower-extremity function. The 3D time-of-flight (3D TOF) method is still the dominant
Abdominal ultrasonography is useful for identifying large non-contrast method for neurovascular MRA. Stationary
aortic thrombi. However, it requires technical expertise and can tissues become magnetically saturated by multiple repetitive
easily miss relatively small thrombi, especially when performed radiofrequency (RF) pulses, whereas flowing blood does not
by an inexperienced sonographer. Thus, as immature thrombi experience these pulses. Therefore, initial magnetization is high.
appear hypoechoic and homogenous in acute conditions, FATE The signal from inflowing blood seems paradoxically bright
cannot be excluded even if a thrombus is not clearly observed compared to background tissues (16, 21).
(6, 7). Doppler flow assessments can usually help diagnose We evaluated the ability of NE-MRA to visualize the feline
appendicular FATE. However, it is not possible to evaluate the full aorta and external iliac arteries in clinically healthy cats and
extent of an aortic thrombus because of difficulties in remaining compared the results with those obtained using 3D ECG-FSE, 3D
perpendicular to the aorta and external iliac arteries. Moreover, TOF NE-MRA, and gadolinium CE-MRA using 1.5-T MRI. We
if the vessels are partially occluded, the presence of arterial blood hypothesized that NE-MRA techniques could be used to visualize
flow cannot be excluded (2). A definitive diagnosis thus requires the aorta and external iliac arteries of clinically healthy cats.
advanced imaging modalities such as computed tomography Furthermore, we hypothesized that 3D TOF, independent of the
and magnetic resonance imaging (MRI), which allow for the heart rate, would provide better a visualization of the aorta and
visualization and evaluation of the full extent of thrombi in the external iliac arteries than 3D ECG-FSE, which can be influenced
aorta and lower extremity arteries at high speed (8–10). by tachycardia.
However, computed tomographic angiography (CTA) has
the disadvantages of exposing the patient to a large amount of
MATERIALS AND METHODS
ionized radiation and causing contrast-induced nephropathy.
Gadolinium-contrast enhanced (CE) magnetic resonance Study Design
angiography (MRA) yields high sensitivity (81–99.5%) and high This prospective analytical study was approved by the
specificity (89–99%) for detecting significant artery stenosis Chungbuk National University Institutional Animal Care
in human patients (11, 12). Nevertheless, recent studies have and Use Committee. All protocols followed the Chungbuk
demonstrated a correlation between gadolinium-based contrast National University Guidelines for Animal Experiments.
media and nephrogenic systemic fibrosis in human patients Informed consent was obtained from all cat owners before
with impaired renal function, suggesting that gadolinium causes study enrollment.
damage by affecting the glomerular filtration rate (13, 14). From June to July 2020, 11 cats owned by client were
Recently, alternative non-contrast-enhanced (NE) MRA included in this study: one intact male, five castrated males,
techniques have been widely used to evaluate the entire lower and five spayed females. The mean age of the cats was
extremity arteries, with high accuracy for detecting vascular 41.9 (11–81) months, and the mean body weight was 4.50
stenosis and thrombi without side effects in humans. It is possible (3.1–5.85) kg. The breeds were Domestic Shorthair (n =
to evaluate only the arteries by separating or suppressing venous 10) and Norwegian Forest (n = 1) (Table 1). All cats were
signals from the artery signals. Furthermore, repetitive scans can confirmed to be clinically healthy after thorough physical
be performed as many times as necessary because no contrast examination, blood analysis including complete blood cell
media is used (15, 16). Nevertheless, no previous veterinary count, serum biochemical analysis (total protein, albumin,

Frontiers in Veterinary Science | www.frontiersin.org 2 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

TABLE 1 | Individual characteristics of eleven clinically healthy cats.

Subject Age (months) Sex Breed Body weight (kg) Used segments

Cat 1 72 SF DSH 4.2 Aorta, External iliac arteries

Cat 2 16 CM DSH 5.85 Aorta, External iliac arteries
Cat 3 61 CM DSH 4.25 Aorta, External iliac arteries
Cat 4 36 SF DSH 4.2 Aorta, External iliac arteries
Cat 5 11 CM NF 4.2 Aorta, External iliac arteries
Cat 6 66 SF DSH 5.1 Aorta, External iliac arteries
Cat 7 11 IM DSH 3.1 Aorta, External iliac arteries
Cat 8 54 SF DSH 5.2 Aorta, External iliac arteries
Cat 9 81 CM DSH 4.2 Aorta, External iliac arteries
Cat 10 22 CM DSH 5 Aorta, External iliac arteries
Cat 11 31 SF DSH 4.15 Aorta, External iliac arteries

SF, spayed female; CM, castrated male; IM, intact male; DSH, Domestic Short hair; NF, Norway Forest.

alanine aminotransferase, alkaline phosphatase, blood urea

nitrogen, creatinine, glucose, and C-reactive protein), analysis
of electrolytes (Na, K, and Cl), urinalysis, and diagnostic
imaging including thoracic radiography, abdominal radiography,
abdominal ultrasonography, and echocardiography.
All cats were sedated with two separate intravenous injections
of butorphanol (0.2 mg/kg; Butophan; Myung Moon Pharm,
Hwaseong, Korea) and midazolam (0.1 mg/kg; Bukwang
Midazolam; Bukwang Pharm, Ansan, Korea) and propofol
(4 mg/kg intravenous injection; Provive injection 1%; Myung
Moon Pharm) was subsequently injected for general anesthesia.
Inhalation anesthesia was maintained with 2.5 to 3.0% isoflurane
(Terrell; Piramal Critical Care, Inc., Bethlehem, PA, USA) and
oxygen (1.0–1.2 L/min). The cats were kept intubated with a
cuffed endotracheal tube and maintained under positive pressure
ventilation using 100% oxygen with a peak inspired pressure
of 8–12 mmHg at a respiratory rate of 7–15 breaths/min, tidal
volume of 15–20 mL/kg, and end-tidal CO2 concentration of
35–45 mmHg. During general anesthesia, the heart rate, non-
invasive systolic blood pressure, respiratory rate, end-tidal CO2
concentration, and isoflurane concentration were monitored
continuously using a patient monitoring machine (IntelliVue FIGURE 1 | An eight-channel human knee coil with a 1.5-T magnet was used
MP70, PHILIPS, Bothell, Washington). Decisions regarding for signal reception. Under inhalational anesthesia, the cat was placed in the
subject inclusion and exclusion for the study were made by two sternal recumbency position with the hind limbs caudally extended; the thigh
authors with diagnostic imaging expertise (M.J.L. and M.J.K.). region was included in the coil. To prevent the hind limbs from being disturbed,
the sponge and hind limbs were fixed together and blankets and hot packs
were applied over the cat to maintain its body temperature. MRI, magnetic
resonance imaging.
MRA Techniques
MRI scans of all cats were performed using a 1.5-T magnet
(SIGNA Creator; GE Healthcare, Milwaukee, WI, USA). All
cats underwent general anesthesia with butorphanol, midazolam in accordance with our parameters (Table 2). For 3D ECG-
and propofol were ventilated, and were placed in the sternal FSE, peripheral pulse gating was used to synchronize image
recumbency position with the feet first in the scanner and hind acquisition and cardiac cycle. To determine optimal trigger
limbs extended (Figure 1). An eight-channel knee coil was used delays, a dorsal single-shot fast spin-echo (SSFSE) sequence was
for signal reception. Subsequently, 3D ECG-FSE and 3D TOF performed in advance.
were performed in a random order immediately followed by
gadolinium CE-MRA. The field of view was adjusted to include Contrast-Enhanced MRA
the abdominopelvic region from the left kidney to caudal thigh. In 11 cats, Gadolinium CE-MRA was performed using the 3D
We performed 3D ECG-FSE, 3D TOF and CE-MRA analyses TOF fast spoiled gradient echo technique with the multiphase

Frontiers in Veterinary Science | www.frontiersin.org 3 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

TABLE 2 | Imaging parameters of 3D ECG-FSE and 3D TOF in non-contrast-enhanced magnetic resonance angiography (NE MRA) and contrast-enhanced magnetic
resonance angiography (CE MRA).

Imaging parameters 3D ECG-FSE 3D TOF CE MRA (3D TOF FSPGR)

Acquisition mode 3D 3D 3D
Acquisition time (minutes:seconds) 8–12 RR interval 8:24 1:24
TR (msec) 6,750 27 5.7
TE (msec) 60.4 6.8 1.6
Flip angle (degrees) 90 25 30
Field of View (mm) 240–260 240–260 240–260
Voxel size (mm) 1.0 × 0.9 × 1.2 0.8 × 1.1 × 1.2 1.0 × 0.9 × 1.2
Acceleration factor N/A Two Two
Bandwidth (Hz/px) 83.33 25 50
Fat suppression No No Yes
Thickness 1.2 1.2 1.2
Matrix 288 × 192 320 × 220 240 × 288

The 3D ECG-FSE acquisition time was dependent on the subject’s heart rate. RR interval: the time elapsed between two successive R waves of the QRS signal on the electrocardiogram.
N/A, not available; 3D ECG-FSE, three-dimensional echocardiograph-gated fast spin echo; 3D TOF, three-dimensional time-of-flight; FSPGR, contrast-enhanced 3D TOF fast spoiled
gradient echo; TE, echo time; TR, repetition time.

method. Each phase lasted 27 s, and four consecutive phases were

performed. Before injecting the contrast medium, a precontrast
scout scan was performed to check the anatomical coverage and
to obtain the subtraction image from the postcontrast study.
The scan was started at the same time as the intravenous bolus
injection of 0.2 mmol/kg of Gadolinium DTPA (Clariscan R
;
GE Healthcare AS, Oslo, Norway) with an auto-injector at a
rate of 2 mL/s, followed by a 10 mL saline flush at the same
rate. After subtraction processing using baseline images, 3D
reconstructed MRA images were acquired through maximum
intensity projection.

MRA Image Analysis

All digital images were assessed using the Digital Imaging
and Communications in Medicine (DICOM) viewer (RadiAnt
DICOM Viewer software; Medixant Co., Poznan, Poland).
Measurements were performed using electronic calipers and the
same imaging viewer. Post-processing and maximum-intensity
projection images of the subtracted data sets were analyzed
using a workstation (Volume viewer 14.0 Ext.2; GE Healthcare,
Milwaukee, WI, USA). Images obtained during all examinations
were evaluated twice, separately, and in random order by two
FIGURE 2 | Maximum-intensity projection reconstruction of non-contrast-
observers with diagnostic imaging expertise (M. J. L. and M. J. K.) enhanced 3D TOF magnetic resonance angiography images of the dorsal view
who were blinded to patient signaling and sequence information. (A) and lateral view (C), and contrast-enhanced 3D TOF fast spoiled
Disagreements were resolved through consensus. gradient-echo magnetic resonance angiography images of the dorsal view (B)
and lateral view (D) of the feline aorta and external iliac arteries at 1.5 Tesla.
Quantitative Image Analysis The solid arrow indicates the aorta, and the open arrows indicate the external
iliac arteries. 3D TOF, three-dimensional time-of-flight.
For the quantitative image analysis of clinically healthy cats, the
signal-to-noise ratio (SNR) could not be calculated because the
parallel imaging used in 3D TOF caused heterogeneous noise
distribution (22). intensity (rSI) was determined for the quantitative assessment of
Signal intensity (SI) was calculated at the center of the normal the relative contrast for each arterial segment, including the aorta
portion of each arterial segment with circular regions of interest and external iliac arteries of the bilateral extremities (22).
(ROI) of 0.01–0.02 mm2 . The rSI of the surrounding stationary
soft tissue was measured from the muscle next to the ROI in Artery SI − Adjacent Muscle SI
rSI=
the vessel with an area of 0.05–0.09 mm2 . The relative signal Artery SI+Adjacent Muscle SI

Frontiers in Veterinary Science | www.frontiersin.org 4 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

TABLE 3 | Image quality assessment in the qualitative image analysis.

Score Grade Description

1 Poor Non-diagnostic due to poor delineation of major arteries

or severe venous contamination
2 Fair Fair delineation of major arteries or some venous
contamination
3 Good Good delineation of major arteries or minor venous
contamination
4 Excellent Excellent delineation of major arteries and no venous
contamination

Qualitative Image Analysis

The image quality for the dorsal maximum-intensity projection
reconstruction of the aorta and external iliac arteries was
assessed by turning the image by 360◦ (Figure 2). Two observers
independently ranked quality using a four-point assessment
scale based on the delineation of artery borders, venous
contamination, and artifacts (Table 3). A grade score of 2–4 was
FIGURE 3 | Dorsal single-shot fast spin-echo (SSFSE) images for cat 6 (A)
deemed diagnostic (17). and cat 8 (B). Single-shot two-dimensional acquisition, which is a prerequisite
for performing 3D ECG-FSE, was not achieved because a clean peripheral
Statistical Analysis pulse gating waveform could not be obtained due to the fast heart rate of cat
The analyses were performed using the statistical software 6. However, single-shot 2D acquisition was normally achieved for cat 8. The
3D ECG-FSE images of D and F are compared in Figure 5. 3D ECG-FSE,
SPSS for Windows (version 21.0; SPSS Inc., Chicago, IL, USA) three-dimensional echocardiograph-gated fast spin echo magnetic resonance
and Prism 9.0 (GraphPad Software, San Diego, CA, USA). angiography; SSFSE, single-shot fast spin-echo.
Data for each segment of the aorta and external iliac arteries
are expressed as mean ± standard deviation and continuous
numerical variables. For all tests, statistical significance was set at
P < 0.05. The rSI values of the 3D ECG-FSE group and 3D TOF
group were obtained, and data were analyzed by one-way analysis trigger delays in 3D ECG-FSE MRA (Figure 3). Consequently,
of variance (ANOVA) procedure after evaluating normality via the total average scan time using 3D ECG-FSE was 1024.3 s
the Shapiro–Wilk test. The image quality scores (IQS) of 3D (879–1,255) s (Table 4).
ECG-FSE and 3D TOF were compared using non-parametric A total of 22 lower limbs and 33 anatomical arterial segments
tests, Spearman’s rank correlation and simple regression analysis. were quantitatively evaluated using rSI in 11 cats and qualitatively
Spearman’s rank correlation coefficients and simple regression evaluated using IQS in 10 cats. The 3D ECG-FSE MRA images
analysis were calculated for IQS and various variables, including of one cat were excluded because of poor image quality caused
heart rate, systolic blood pressure, respiratory rate, and scan by a non-detectable pulse gating signal. During the quantitative
time, to determine whether any individual factor associated with analysis, SNR could not be calculated because the parallel
general anesthesia correlated more closely with the IQS of the imaging technique, which resulted in a heterogeneous noise
3D ECG-FSE and 3D TOF at the time of scanning for each distribution, was used as an option for the 3D TOF protocol (23).
cat. For all analyses, a significant difference was defined as P < rSI values were measured at 33 arterial segments of the aorta and
0.05. To evaluate the reproducibility and consistency of rSI and external iliac arteries in 11 cats using 3D ECG-FSE and 3D TOF.
IQS between the two observers, intraclass correlation coefficients
were calculated using a two-way mixed model. Generally, it can
be evaluated as a good degree of consistency if it is 0.75 or higher. Comparison of Imaging Parameters of 3D
Lastly, Fisher’s exact test was used to confirm the association ECG-FSE and 3D TOF MRA
between sequence and IQS. The rSI values of 3D ECG-FSE MRA, 3D TOF, and CE-MRA
for the aorta were 0.83 ± 0.06, 0.57 ± 0.06, and 0.58 ± 0.05
RESULTS respectively. The rSI values of 3D ECG-FSE MRA, 3D TOF, and
CE-MRA for the external iliac arteries were 0.80 ± 0.05, 0.45
All eleven cats underwent NE-MRA using both 3D ECG-FSE ± 0.06, and 0.57 ± 0.03, respectively. The rSI for all segments
and 3D TOF techniques. The average scan times of 3D TOF observed using 3D TOF were significantly lower than those
and 3D ECG-FSE were 524 (508–528) s and 641.2 (559–762) s, observed using 3D ECG-FSE (P < 0.001) and similar to those
respectively. However, an additional dorsal single-shot fast spin- observed using CE-MRA. The highest segment rSI was observed
echo (SSFSE) sequence was required to determine the optimal in the aorta using 3D ECG-FSE (Figure 4).

Frontiers in Veterinary Science | www.frontiersin.org 5 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

TABLE 4 | General anesthesia variables for 3D ECG-FSE and 3D TOF.

Subject MRA Heart rate (beats per Systolic blood Respiratory rate Acquisition time
minute) pressure (mmHg) (breaths per minute) (minutes:seconds)

Cat 1 3D ECG-FSE 120 100 8 10:25

3D TOF 110 100 8 08:48
Cat 2 3D ECG-FSE 112 112 8 11:10
3D TOF 106 106 8 08:42
Cat 3 3D ECG-FSE 93 90 9 12:42
3D TOF 97 90 9 08:31
Cat 4 3D ECG-FSE 108 91 7 10:31
3D TOF 100 83 7 08:28
Cat 5 3D ECG-FSE 94 80 10 10:17
3D TOF 90 80 10 08:48
Cat 6 3D ECG-FSE N/A 80 12 01:48
3D TOF N/A 80 12 08:48
Cat 7 3D ECG-FSE 88 73 9 10:03
3D TOF 89 80 9 08:48
Cat 8 3D ECG-FSE 80 80 9 09:19
3D TOF 90 88 9 08:48
Cat 9 3D ECG-FSE 114 97 8 11:18
3D TOF 95 90 8 08:48
Cat 10 3D ECG-FSE 98 115 10 10:47
3D TOF 87 80 10 08:48
Cat 11 3D ECG-FSE 89 85 8 10:20
3D TOF 100 100 8 08:48

N/A, not available; 3D ECG-FSE, three-dimensional echocardiograph-gated fast spin echo; 3D TOF, three-dimensional time-of-flight; MRA, magnetic resonance angiography.

FIGURE 4 | Box and whisker plots of rSI for the aorta and external iliac arteries measured using 3D ECG-FSE, 3D TOF and CE-MRA for clinically healthy cats. Boxes
indicate the sample variation, and whiskers represent the standard deviation. It was confirmed that the rSI of 3D ECG-FSE was significantly higher than that of 3D TOF
and CE-MRA for all arterial segments. Although the rSI of 3D TOF was lower than that of CE-MRA for aorta and external iliac arteries, however there was no significant
difference. ****Statistically significant difference (P < 0.001) determined by an analysis of variance followed by the one-way ANOVA. rSI, relative signal intensity; 3D
ECG-FSE, three-dimensional echocardiograph-gated fast spin echo; 3D TOF, three-dimensional time-of-flight; FSPGR, contrast-enhanced 3D TOF fast spoiled
gradient-echo; ns, no significance.

Frontiers in Veterinary Science | www.frontiersin.org 6 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

TABLE 5 | Image quality assessment of the aorta and external iliac arteries.

Anatomic location Sequence Score

Excellent Good Fair Poor

Aorta 3D TOF(n)/3D FSE(n)/FSPGR(n) 10/0/5 1/3/5 0/7/0 0/1/0

External iliac arteries. 3D TOF(n)/3D FSE(n)/FSPGR(n) 10/0/4 1/2/6 0/8/0 0/1/0

3D ECG-FSE, three-dimensional echocardiograph-gated fast spin echo; 3D TOF, three-dimensional time-of-flight; FSPGR, contrast-enhanced 3D TOF fast spoiled gradient echo.

(2.23 ± 0.56) (P < 0.001) and similar to those obtained using

CE-MRA (3.57 ± 0.41; using one-way ANOVA).
When using 3D ECG-FSE in the aorta and external iliac
arteries, venous contamination was confirmed in all 11 cats
(100%), and filling defect-like artifacts were clearly identified in
8 out of 11 cats (73%) and 9 out of 11 cats (82%), respectively.
These artifacts appeared to be signal dropout artifacts, but there
was a difference in their degrees. Generally, the arterial vessel
margins were blurry and irregular. However, when using 3D
TOF, the arterial delineation was excellent and there was little
contamination of the veins around the aorta and external iliac
arteries (Figures 5, 6).
Repetitive scanning for each sequence to observe the
reproducibility and consistency of images indicated no noticeable
changes when using 3D TOF. However, changes in blood vessel
diameters and artifacts, such as seemingly disconnected blood
vessels, filling defects (9 of 11 cats; 81.8%), and loss of blood
vessels (3 of 11 cats; 27.3%), were clearly identified when using 3D
ECG-FSE. Thus, differences were observed in the reproducibility
and consistency of images obtained using the two main NE-MRA
techniques (Figure 6).
FIGURE 5 | Maximum-intensity projection reconstruction of magnetic CE-MRA was performed for 11 cats. Pure arterial contrast
resonance angiography images of each dorsal and oblique views obtained
using 3D TOF (A–C), 3D ECG-FSE (D–F) and FSPGR. The aorta and external
images of the aorta (four cats) and external iliac arteries (five cats)
iliac arteries were excellently delineated with no venous contamination and were acquired; the remaining cats were confirmed to have venous
were scored as “Excellent” (A,B,G). In (C,D,H), it was scored as “Good” due contamination. CE-MRA was confirmed to have abilities and IQS
to minor venous contamination. Considering the margin of the arterial wall, the similar to those of 3D TOF (Figure 5).
venous contamination overlapping the aorta and external iliac arteries, and the
overall diagnostic qualities, image reconstruction was scored as “Fair” (E), and
“poor” (F), respectively. N/A, not available; 3D ECG-FSE, three-dimensional
Correlation Between General Anesthesia
echocardiograph-gated fast spin echo; 3D TOF, three-dimensional Variables and Image Quality Scores
time-of-flight; FSPGR, contrast-enhanced 3D TOF fast spoiled gradient-echo. No variable correlated with the IQS of 3D TOF in both the
aorta and external iliac arteries (heart rate in aorta and external
iliac arteries, P = 0.516, rs = 0.234; systolic blood pressure
in aorta and external iliac arteries, P = 0.871, rs = 0.059;
Quantitative Image Analysis respiratory rate in aorta and external iliac arteries, P = 0.614,
The subjective IQS for the aorta were rated as either “excellent” rs = −0.183; and acquisition time in aorta and external iliac
(10/11, 91%) or “good” (1/10, 9%) using 3D TOF, and as “good” arteries, P = 0.552, rs = −0.214). Spearman’s rank correlation
(3/11, 27%) using 3D ECG-FSE. The subjective IQS for the showed moderately negative correlations between the heart rate
external iliac arteries were rated as either “excellent” (10/11, in the aorta (P = 0.017; rs = −0.798), heart rate in the external
91%) or “good” (1/10, 9%) using 3D TOF, and as “good” (2/11, iliac arteries (P = 0.005; rs = −0.830), systolic blood pressure
18%) using 3D ECG-FSE (Table 5). A non-diagnostic image was in the aorta (P = 0.042; rs = −0.686), systolic blood pressure in
obtained when 3D ECG-FSE was used (Table 5 and Figure 5). the external iliac arteries (P = 0.027; rs = −0.709), acquisition
The IQS for the aorta obtained using 3D TOF (3.95 ± 0.15) time in the aorta (P = 0.033; rs = −0.722), acquisition time in
were significantly higher than those obtained using 3D ECG-FSE the external iliac arteries (P = 0.013; rs = −0.779), and the IQS
(2.32 ± 0.60) (P < 0.001) and similar to those obtained using of the 3D ECG-FSE (Table 6). Multiple linear regression analysis
CE-MRA (3.61± 0.41; using one-way ANOVA). The IQS for the (aorta: F = 6.471, p = 0.033/external iliac arteries: F = 5.318, p
external iliac arteries obtained using 3D TOF (3.98 ± 0.08) were = 0.048) contained four predictors that together accounted for
also significantly higher than those obtained using 3D ECG-FSE over 70.9% of the variance (adjusted R2 = 0.709) in image quality

Frontiers in Veterinary Science | www.frontiersin.org 7 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

FIGURE 6 | Cat 7. The aorta and external iliac arteries were not the same during the first [(A), left] and second [(A), right] trial using 3D ECG-FSE. Moderate-to-severe
venous contamination is observed [(A), arrow], and intermittent filling defect-like artifacts are observed in the aorta [(A), arrowhead]. A visible difference in the
diameters (A) of the vessels in the aorta and the caudal vena cava was observed [(A), left and right, empty arrow]. The first trial and second trial resulted in
observations similar with those obtained using 3D TOF (B). 3D ECG-FSE, three-dimensional echocardiograph-gated fast spin-echo magnetic resonance angiography;
3D TOF, three-dimensional time-of-flight; MRA, magnetic resonance angiography.

TABLE 6 | Correlation between variables and image quality scores for 3D ECG-FSE and 3D TOF in general anesthesia.

Value Heart rate Systolic blood pressure Respiratory rate Acquisition time

3D ECG-FSE rs value −0.846/−0.675 −0.435/−0.304 0.284/0.041 −0.846/−0.778

(aorta/external iliac arteries) P-value 0.002*/0.032* 0.210/0.393 0.427/0.911 0.002*/0.008*
3D TOF rs value 0.234/0.234 0.059/0.059 −0.183/−0.183 −0.214/−0.214
(aorta/external iliac arteries) P-value 0.516/0.516 0.871/0.871 0.614/0.614 0.552/0.552

Spearman’s correlation coefficient (rs value) and statistical significance (P-value) for correlations between variables in general anesthesia and image quality scores for 3D ECG-FSE MRA
and 3D TOF.
*Statistically significant. An alpha level of 0.05 was used to determine statistical significance.
3D ECG-FSE, three-dimensional echocardiograph-gated fast spin echo; and 3D TOF, three-dimensional time-of-flight.

TABLE 7 | Multiple linear regression analysis for aorta and external iliac arteries between variables and image quality scores for 3D ECG-FSE in general anesthesia.

Variables 95% confidence interval β p

3D ECG-FSE (aorta/external iliac arteries) Heart rate (−0.074, −0.009)/(−0.062, −0.004) −0.976/−0.923 0.021/0.034
Systolic blood pressure (−0.021, 0.036)/(−0.020, 0.032) 0.189/0.180 0.526/0.575
Respiratory rate (−0.408, 0.250)/(−0.433, 0.161) −0.161/−0.333 0.565/0.292
Acquisition time (−0.008, 0.005)/(−0.008, 0.004) −0.144/−0.214 0.645/0.531

F = 6.471/5.318, P = 0.033/0.048, R2 = 0.838/0.900; adjusted R2 = 0.709/0.810.

β, standardized coefficients; 3D ECG-FSE, three-dimensional echocardiograph-gated fast spin echo.

score within aorta and for over 65.7% of the variance (adjusted R2 r = 0.938). Regarding IQS, an almost perfect agreement was
= 0.657) in image quality score within external iliac arteries. As confirmed between the two observers for 3D ECG-FSE in the
reported in Table 7, heart rate contributed explaining the image aorta (r = 0.916), 3D ECG-FSE in the external iliac arteries (r
quality score. = 0.996), 3D TOF in the aorta (r = 1.000), 3D TOF in the
external iliac arteries (r = 0.809), CE-MRA of the aorta (r =
Inter-observer Agreement 0.890), and CE-MRA of the external iliac arteries (r = 0.991).
Inter-observer agreement for rSI and IQS in the aorta and The measurements obtained by the two readers showed high
external iliac arteries was assessed. Excellent agreement was inter-observer agreement, thereby indicating the reproducibility
achieved for rSI between the two observers using the two main of these measurements.
NE MRA techniques, as determined via the intraclass correlation
coefficient test (3D ECG-FSE in aorta, r = 0.999; 3D ECG-FSE Cross Analysis
in external iliac arteries, r = 0.994; 3D TOF in aorta, r = 0.956; A Fisher’s exact test was conducted to determine the relationship
and 3D TOF in external iliac arteries, r = 0.955; CE-MRA of between sequence and IQS. As an expected frequency of <5
the aorta, r = 0.965; and CE-MRA of the external iliac arteries, accounted for 5(56%) of 9 cells, the accuracy of Fisher’s exact test,

Frontiers in Veterinary Science | www.frontiersin.org 8 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

TABLE 8 | Fisher’s exact test between MRA and image quality scores in the aorta. However, 3D TOF is prone to artifacts from flow saturation in
the horizontal direction due to the rapid and repeated application
Imaging quality score Fisher’s exact test (P)
of the excitation RF pulse and tracking of the venous saturation
Excellent Good Fair of the RF pulse, which is typically placed just downstream of the
imaging region (26). To compensate for this, vein contamination
3D TOF 10 1 0 <0.001
was prevented by adding a fat saturation band in the inferior
3D ECG-FSE 0 2 8 direction at the location where the vein descended from the
FSPGR 5 5 0 heart. The intravascular mottled artifacts due to a signal dropout
Total 15 8 8 were subsequently reduced by providing overlapping in the slab.
MRA, magnetic resonance angiography; 3D ECG-FSE, three-dimensional In this study, 3D TOF had excellent image quality, less venous
echocardiograph-gated fast spin echo; 3D TOF, three-dimensional time-of-flight; contamination, and better arterial delineation than 3D ECG-FSE.
FSPGR, contrast-enhanced 3D TOF fast spoiled gradient echo. Furthermore, a common disadvantage of TOF is the long
scanning time in humans. However, in our study, the average
scanning time of 3D TOF was ∼9 min shorter than that of 3D
TABLE 9 | Fisher’s exact test between MRA and image quality scores in the
external iliac arteries.
ECG-FSE. It is because 3D TOF required only post-processing
after the sequence, while 3D ECG-FSE needs optimal SSFSE
Imaging quality score Fisher’s exact test (P) sequence and an additional processing step to determine the
Excellent Good Fair optimal trigger delays. Also, scanning time of 3D ECG-FSE can
be prolonged depending on cat size. Additionally, 3D TOF had
3D TOF 10 1 0 < 0.001 high consistency and reproducibility compared with 3D ECG-
3D ECG-FSE 0 3 7 FSE when visualizing the aorta and external iliac arteries in
FEPGR 4 6 0 healthy cats.
Total 10 4 7 Second, 3D ECG-FSE is known as an excellent NE-MRA
technique that allows visualization of the lower arteries without
MRA, magnetic resonance angiography; 3D ECG-FSE, three-dimensional
echocardiograph-gated fast spin echo; 3D TOF, three-dimensional time-of-flight;
motion artifacts in humans (10, 19, 20). However, in the current
FSPGR, contrast-enhanced 3D TOF fast spoiled gradient echo. study, the image quality obtained using 3D ECG-FSE to visualize
the aorta and external iliac arteries in clinically healthy cats was
not as good as that obtained using 3D ECG-FSE in humans.
In principle, 3D ECG-FSE involves the voxel-wise subtraction
not the Pearson chi-square test, was identified. The sequence and
of two 3D datasets of fast arterial flow during the systole and
IQS were related to each other (P < 0.001) (Tables 8, 9).
slow venous flow during the diastole obtained through ECG
gating or pulse gating. Consequently, the signals of the veins and
DISCUSSION surrounding tissues are eliminated, and only the signals of the
arteries are obtained (17). However, when the difference in the
In cats with cardiomyopathies, the diagnosis of FATE using arterial signals of the systole and diastole is minimal, poor arterial
abdominal ultrasonography requires technical expertise, and it contrast images are obtained. Sensitivity to cardiac arrhythmia
may be difficult to scan the thrombi when they are small or and spatial blurring due to T2 decay during rapid FSE echo
located in the peripheral arteries. Advanced imaging modalities training also results in contaminated venous images (9, 16).
such as CTA and CE-MRA can be applied; however, adverse Generally, the heart rate of cats is approximately 100 beats per
effects may occur with the use of contrast media, as reported minute during anesthesia (just before scanning), which is higher
previously (24, 25). No previously published veterinary medicine than the 60 bpm average in humans (29, 30). Nevertheless, a heart
study has reported the use of NE-MRA to visualize the aorta rate of approximately 100 bpm is still too fast to capture an image
and external iliac arteries in cats. The findings of this study of a small subject such a cat. According to a previous study in
demonstrated that NE-MRA can be used to visualize the aorta humans, the image quality scores for subjects with heart rates
and external iliac arteries in clinically healthy cats without the use ≤65 bpm were significantly better than those for subjects with
of contrast media. This can thus provide anatomical information higher heart rates (31).
and support FATE diagnosis. In this study, a fast and shallow peripheral pulse was
Three important findings were observed in this study. First, confirmed during pulse monitoring, despite general anesthesia.
TOF is one of the oldest NE-MRA techniques and is rarely This made it difficult to achieve sufficient relaxation and
used for lower-extremity peripheral vascular diseases because contraction during cardiac cycles and resulted in the degraded
other protocols, such as quiescent-interval signal-shot (QISS), quality of the images with venous and surrounding tissue
provide better image quality (26). Instead, non-subtractive, contamination and artifacts. The IQS of 3D ECG-FSE had a
inflow-dependent 3D TOF methods are widely used to evaluate moderately negative correlation with the heart rate, systolic blood
the extracranial carotid arteries and circle of Willis due to its pressure, and scan time (Table 6).
isotropic resolution in humans. Moreover, TOF is mainly used Technical problems could be considered the cause of artifacts
for the evaluation of brain arterial diseases in veterinary medicine and noise in 3D ECG-FSE. The bandwidth of 3D ECG-FSE,
(27, 28). which is the range of frequencies involved in the transmission

Frontiers in Veterinary Science | www.frontiersin.org 9 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

or reception of an electronic signal, had to be higher than that applied based on the guidelines stipulated for humans. It was
of lower-extremity MRA in humans to acquire more systolic difficult to modify and apply the fluoroscopic trigger method
and diastolic data of remarkably fast cardiac cycles in cats. to detect the contrast agent peak in the abdominal aorta of the
The disadvantage of a higher bandwidth is a larger amount of evaluated cats due to their smaller body size and higher heart
noise and lower SNR owing to the larger frequency range (32). rate compared with those of humans. Therefore, it is necessary
Consequently, the possibility that the high bandwidth caused to study the CE-MRA protocol for evaluation of arteries in the
more artifacts in the 3D ECG-FSE image was also considered. lower extremities of cats specifically.
Another drawback of 3D ECG-FSE is that it requires
additional time to obtain a preparatory single-shot two- CONCLUSION
dimensional acquisition sequence to find the specific trigger
delays for the systole and the diastole (16, 17). Each single-shot The ability of NE-MRA to visualize the feline aorta and
2D acquisition is performed with two or three RR intervals to the external iliac arteries was examined in this study. Unlike
produce separate images for each phase in humans (16, 33, 34). human medicine, 3D ECG-FSE was confirmed to result in the
However, the cats in this study required at least 8–12 RR intervals deterioration of image quality (venous contamination, artifacts,
because of their shallow cardiac rhythm. Additional 2–4 min and decreased arterial delineation) in cats, and repetitive and
were required to obtain the SSFSE sequence for one trial. If the reproducible images were not obtained because of their fast and
sequence is not acquired immediately, more time is required for shallow cardiac cycle. 3D TOF is more reliable and reproducible
repeated scans. If specific trigger delays could not be acquired than the 3D ECG-FSE technique because it results in good image
despite repeated retakes, then the image quality of 3D ECG-FSE quality that is sufficient for visualizing the aorta and external
was poor (Figures 3, 5). The longer the RR interval, the longer iliac arteries in clinically healthy cats. Therefore, it is expected
the acquisition time, and the lower the image quality (Table 6). to be used in the future to diagnose thromboembolism, assess
Third, our findings showed that a high signal strength did disease severity, choose appropriate therapy, and determine the
not indicate good image quality for the aorta and external iliac treatment progress and prognosis of patients with FATE.
arteries when using NE MRA. The rSI of 3D ECG-FSE was
significantly higher for all segments, including the aorta and
external iliac arteries, compared with those of 3D TOF and CE DATA AVAILABILITY STATEMENT
MRA. It was confirmed that rSI does not have a significant effect
The datasets presented in this study can be found in online
on image quality.
repositories. The names of the repository/repositories
Based on the complete occlusion of the middle cerebral
and accession number(s) can be found in the
artery observed through 3D TOF in a previous study that
article/supplementary material.
induced canine ischemic stroke model, we inferred that arterial
thromboembolism occurring in feline aorta and external iliac
arteries could be visualized using 3D TOF (35). However, since ETHICS STATEMENT
this study involved only clinically healthy cats, it is not known
whether artifacts that can mimic thrombi, which is one of the The animal study was reviewed and approved by CBNUR-1507-
common disadvantages of 3D TOF, will be noticeable in actual 21. Written informed consent was obtained from the owners for
patients with FATE and whether there could be better results the participation of their animals in this study. Written informed
with 3D ECG-FSE than with 3D TOF sequences. Further studies consent was obtained from the minor(s)’ legal guardian/next of
comparing the image quality of 3D TOF and 3D ECG-FSE in cats kin for the publication of any potentially identifiable images or
with FATE are warranted to evaluate their clinical significance. data included in this article.
Regarding the overall condition of cats with FATE, the
application of contrast medium and long-term general anesthesia AUTHOR CONTRIBUTIONS
can further worsen the patient’s condition. In contrast, when
using 3D TOF as a diagnostic tool in cats with FATE, adverse ML, MK, JisA, JiyA, JY, JC, SO, and DC contributed to the case
effects caused by contrast media could be avoided. management. ML wrote the first draft of the manuscript. ML,
There were some limitations to this study. First, our sample MK, JisA, JiyA, JC, SO, and DC participated in the revision of the
size was small, and the statistical power was insufficient to detect manuscript. All authors have read, commented on, and approved
associations. Although all 11 cats were scanned simultaneously the final manuscript.
with administration of the contrast agent, the contrast patterns
were different for each cat. We confirmed that approximately half FUNDING
of the cats were in the venous phase before the arteries filled.
Therefore, the average IQS obtained via CE-MRA was slightly This work was supported by Institute for Information &
lower than that obtained via 3D TOF, although this difference was communications Technology Promotion (IITP) grant funded by
not statistically significant. the Korea government (MSIP) (No. 2021-0-00490, Development
Moreover, the dose and injection rate of the contrast medium of Precision Analysis and Imaging Technology for Biological
and the contrast agent tracking methods used in this study were Radio Waves).

Frontiers in Veterinary Science | www.frontiersin.org 10 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

REFERENCES pulses - initial experience. Rofo. (2010) 182:861–7. doi: 10.1055/s-0029-12

45328
1. Smith SA, Tobias AH, Jacob KA, Fine DM, Grumbles PL. Arterial 18. Miyazaki M, Sugiura S, Tateishi F, Wada H, Kassai Y, Abe H. Non-
thromboembolism in cats: acute crisis in 127 cases (1992-2001) contrast-enhanced MR angiography using 3D ECG-synchronized
and long-term management with low-dose aspirin in 24 cases. J half-Fourier fast spin echo. J Magn Reson Imaging. (2000) 12:776–83.
Vet Intern Med. (2003) 17:73–83. doi: 10.1111/j.1939-1676.2003.tb doi: 10.1002/1522-2586(200011)12:5&lt;776::AID-JMRI17&gt;3.0.CO;2-X
01326.x 19. Atanasova IP, Kim D, Storey P, Rosenkrantz AB, Lim RP, Lee VS. Sagittal
2. Smith SA, Tobias AH. Feline arterial thromboembolism: an fresh blood imaging with interleaved acquisition of systolic and diastolic
update. Vet Clin North Am Small Anim Pract. (2004) 34:1245–71. data for improved robustness to motion. Magn Reson Med. (2013) 69:321–8.
doi: 10.1016/j.cvsm.2004.05.006 doi: 10.1002/mrm.24576
3. Borgeat K, Wright J, Garrod O, Payne JR, Fuentes VL. Arterial 20. Storey P, Otazo R, Lim RP, Kim S, Fleysher L, Oesingmann N, et al.
thromboembolism in 250 cats in general practice: 2004– Exploiting sparsity to accelerate noncontrast MR angiography in the context
2012. J Vet Intern Med. (2014) 28:102–8. doi: 10.1111/jvim. of parallel imaging. Magn Reson Med. (2012) 67:1391–400. doi: 10.1002/mrm.
12249 23132
4. Schoeman JP. Feline distal aortic thromboembolism: a review of 44 cases 21. Saloner D. The AAPM/RSNA physics tutorial for residents. An introduction
(1990–1998). J Feline Med Surg. (1999) 1:221–31. doi: 10.1053/jfms.1999. to MR angiography. Radiographics. (1995) 15:453–65.
0049 22. Zhang N, Fan Z, Luo N, Bi X, Zhao Y, An J, et al. Noncontrast MR
5. Lunsford KV, Mackin AJ. Thromboembolic therapies in dogs and cats: an angiography (MRA) of infragenual arteries using flow-sensitive dephasing
evidence-based approach. Vet Clin N Am Small Anim Pract. (2007) 37:579– (FSD)-prepared steady-state free precession (SSFP) at 3.0 Tesla: comparison
609. doi: 10.1016/j.cvsm.2007.01.010 with contrast-enhanced MRA. J Magn Reson Imaging. (2016) 43:364–72.
6. Sharpley J, Thode H, Sestina L, Park R, Monnet E, Kraft SL. doi: 10.1002/jmri.25003
Distal abdominal aortic thrombosis diagnosed by three-dimensional 23. Asbach P, Meade MD, Sattenberg RJ, Klessen C, Huppertz A,
contrast-enhanced magnetic resonance angiography. Vet Radiol Heidenreich JO. Continuously moving table aortoiliofemoral run-
Ultrasound. (2009) 50:370–75. doi: 10.1111/j.1740-8261.2009.0 off contrast-enhanced magnetic resonance angiography: image
1552.x quality analysis in comparison to the multistep acquisition.
7. Drost WT, Bahr RJ, Henry GA, Campbell GA. Aortoiliac thrombus Acta Radiol. (2014) 55:266–72. doi: 10.1177/02841851134
secondary to a mineralized arteriosclerotic lesion. Vet Radiol 98535
Ultrasound. (1999) 40:262–6. doi: 10.1111/j.1740-8261.1999.tb0 24. Scarabelli S, Cripps P, Rioja E, Alderson B. Adverse reactions following
0359.x administration of contrast media for diagnostic imaging in anaesthetized
8. Urata J, Miyazaki M, Wada H, Nakaura T, Yamashita Y, Takahashi M. dogs and cats: a retrospective study. Vet Anaest Analg. (2016) 43:502–10.
Clinical evaluation of aortic diseases using nonenhanced MRA with ECG- doi: 10.1111/vaa.12335
triggered 3D half-Fourier FSE. J Magn Reson Imaging. (2001) 14:113–20. 25. Kirberger RM, Cassel N, Carstens A, Goddard A. The effects of repeated
doi: 10.1002/jmri.1160 intravenous iohexol administration on renal function in healthy beagles-
9. Mathew RC, Kramer CM. Recent advances in magnetic a preliminary report. Acta Vet Scand. (2012) 54:47. doi: 10.1186/1751-014
resonance imaging for peripheral artery disease. Vasc 7-54-47
Med. (2018) 23:143–52. doi: 10.1177/1358863X187 26. Offerman EJ, Hodnett PA, Edelman RR, Koktzoglou I. Nonenhanced methods
54694 for lower-extremity MRA: a phantom study examining the effects of stenosis
10. Hanrahan CJ, Lindley MD, Mueller M, Kim D, Sommers D, Morrell G, and pathologic flow waveforms at 1.5T. J Magn Reson Imaging. (2011) 33:401–
et al. Diagnostic accuracy of noncontrast MR angiography protocols at 3T 8. doi: 10.1002/jmri.22457
for the detection and characterization of lower extremity peripheral arterial 27. De Marco JK, Nesbit GM, Wesbey GE, Richardson D. Prospective
disease. J Vasc Interv Radiol. (2018) 29:1585–94.e2. doi: 10.1016/j.jvir.2018. evaluation of extracranial carotid stenosis: MR angiography with
06.015 maximum-intensity projections and multiplanar reformation compared
11. Meaney JFM, Ridgway JP, Chakraverty S, Robertson I, Kessel D, Radjenovic with conventional angiography. Am J Roentgenol. (1994) 163:1205–12.
A, et al. Stepping-table gadolinium-enhanced digital subtraction MR doi: 10.2214/ajr.163.5.7976902
angiography of the aorta and lower extremity arteries: preliminary 28. Morelli JN, Gerdes CM, Schmitt P, Ai T, Saettele MR, Runge VM,
experience 1. Radiology. (1999) 211:59–67. doi: 10.1148/radiology.211.1.r99a et al. Technical considerations in MR angiography: an image-based
p1859 guide. J Magn Reson Imaging. (2013) 37:1326–41. doi: 10.1002/jmri.
12. Steffens JC, Schäfer FKW, Schäfer S, Link J, Jahnke T, Heller M, et al. 24174
Bolus-chasing contrast-enhanced 3d MRA of the lower extremity 29. Seymour C, Duke-Novakovski T, Mendenhall V. BSAVA Manual of Canine
comparison with intraarterial DSA. Acta Radiol. (2003) 44:185–92. and Feline Anaesthesia and Analgesia. 2nd ed. Gloucester: British Small
doi: 10.1034/j.1600-0455.2003.00044.x Animal Veterinary Association (2008). p. 176.
13. Kooiman J, Pasha SM, Zondag W, Sijpkens YWJ, van der Molen 30. Moitra VK, Gabrielli A, Maccioli GA, O’Connor MF. Anesthesia
AJ, Huisman M v., et al. Meta-analysis: serum creatinine changes advanced circulatory life support. Can J Anaesth. (2012) 59:586–603.
following contrast enhanced CT imaging. Eur J Radiol. (2012) 81:2554–61. doi: 10.1007/s12630-012-9699-3
doi: 10.1016/j.ejrad.2011.11.020 31. Wu YW, Tadamura E, Yamamuro M, Kanao S, Nakayama K, Togashi
14. Buhaescu I, Izzedine H. Gadolinium-induced nephrotoxicity. Int J Clin Pract. K. Evaluation of three-dimensional navigator-gated whole heart MR
(2008) 62:1113–18. doi: 10.1111/j.1742-1241.2007.01582.x coronary angiography: the importance of systolic imaging in subjects with
15. Edelman RR, Koktzoglou I. Noncontrast MR angiography: an high heart rates. Eur J Radiol. (2007) 61:91–6. doi: 10.1016/j.ejrad.2006.
update. J Magn Reson Imaging. (2019) 49:355–73. doi: 10.1002/jmri. 08.013
26288 32. McRobbie DW, Moore EA, Graves MJ, Prince MR. MRI From Picture to
16. Miyazaki M, Takai H, Sugiura S, Wada H, Kuwahara R, Urata J. Peripheral Proton. Cambridge: Cambridge University Press (2006).
MR angiography: separation of arteries from veins with flow-spoiled gradient 33. Miyazaki M, Akahane M. Non-contrast enhanced MR angiography:
pulses in electrocardiography-triggered three-dimensional half-Fourier fast established techniques. J Magn Reson Imaging. (2012) 35:1–19.
spin-echo imaging. Radiol. (2003) 227:890–96. doi: 10.1148/radiol.22730 doi: 10.1002/jmri.22789
20227 34. Lim RP, Hecht EM, Xu J, Babb JS, Oesingmann N, Wong S, et al. 3D
17. Lanzman RS, Blondin D, Schmitt P, Orzechowski D, Godehardt E, Nongadolinium-enhanced ECG-gated MRA of the distal lower extremities:
Scherer A, et al. Non-enhanced 3D MR angiography of the lower preliminary clinical experience. J Magn Reson Imaging. (2008) 28:181–9.
extremity using ECG-gated TSE imaging with non-selective refocusing doi: 10.1002/jmri.21416

Frontiers in Veterinary Science | www.frontiersin.org 11 June 2022 | Volume 9 | Article 819627

Lee et al. Abdominal Aorta and External Iliac Artery MRA in Cats

35. Kang BT, Jang DP, Gu SH, Kim YB, Lim CY, Lee JH, et al. Three-dimensional the publisher, the editors and the reviewers. Any product that may be evaluated in
time-of-flight magnetic resonance angiography of intracranial vessels in a this article, or claim that may be made by its manufacturer, is not guaranteed or
canine model of ischemic stroke with permanent occlusion of the middle endorsed by the publisher.
cerebral artery. Comp Med. (2009) 59:72–7.
Copyright © 2022 Lee, Ko, Ahn, Ahn, Yu, Chang, Oh and Chang. This is an open-
Conflict of Interest: The authors declare that the research was conducted in the access article distributed under the terms of the Creative Commons Attribution
absence of any commercial or financial relationships that could be construed as a License (CC BY). The use, distribution or reproduction in other forums is permitted,
potential conflict of interest. provided the original author(s) and the copyright owner(s) are credited and that the
original publication in this journal is cited, in accordance with accepted academic
Publisher’s Note: All claims expressed in this article are solely those of the authors practice. No use, distribution or reproduction is permitted which does not comply
and do not necessarily represent those of their affiliated organizations, or those of with these terms.

Frontiers in Veterinary Science | www.frontiersin.org 12 June 2022 | Volume 9 | Article 819627