A steady-state free precession sequence can be combined with an inversion-recovery prepulse after the injection of Gadofosveset-trisodium for imaging of cardiovascular structures.
GdT in combination with IR-SSFP significantly improves vessel wall sharpness, contrast to noise ratios and image quality compared to the commonly used T2-SSFP (T2-SSFP) sequence after the injection of Gadopentetate-dimeglumine (GdD) and GdT.
Implications for Patient Care:
GdT in combination with IR-SSFP can be effective in imaging complex cardiovascular anatomy in patients with congenital heart disease (CHD).
IR-SSFP with GdT improved image quality allowing the exclusion and confirmation of cardiovascular pathologies in patients with CHD with higher diagnostic certainty compared to T2-SSFP with GdD and GdT and respective contrast enhanced first-pass angiographies.
Purpose: To compare image quality and diagnostic performance of a contrast agent specific inversion-recovery SSFP sequence (IR-SSFP) in combination with an intravascular contrast agent (Gadofosveset-trisodium, GdT) to a commonly used T2prepared SSFP sequence (T2-SSFP) with an extravascular (Gadopentetate-dimeglumine, GdD) and intravascular (GdT) contrast agent in patients with congenital heart disease.
Get your grade
or your money back
using our Essay Writing Service!
Materials and Methods: The local ethics committee and the MHRA approved this study. Patient informed consent was obtained. Twenty-three patients with CHD were assessed on a 1.5T MR-scanner using a 32-channel coil. GdD and GdT were used in the same patient on consecutive days. Vessel wall sharpness, contrast to noise ratios, image quality and diagnostic performance of IR-SSFP with GdT were compared to T2-SSFP with GdD and GdT and respective CEMRAs. The Wilcoxon rank sum test was used to compare categorical variables; t-tests were used to compare continuous variables.
Results: IR-SSFP with GdT significantly improved vessel wall sharpness, contrast to noise ratios and image quality (all p<0.05) of all investigated intra- and extracardiac structures compared to T2-SSFP with GdD and GdT and respective CEMRAs. Using IR-SSFP with GdT new unsuspected pathologies (22%, n=5/23, e.g. bicuspid aortic valves, ventricular septum defect) were diagnosed, while other pathologies could be excluded (65%, n=15/23, e.g. sinus venosus defects, coronary artery anomalies). Information available from echocardiography (n=23), catheterization (n=4) or surgery (n=1) was used to confirm all diagnoses.
Conclusion: IR-SSFP with GdT improved image quality and diagnostic performance allowing for a more accurate and complete assessment of cardiovascular anatomy in patients with CHD compared to T2-SSFP with GdD and GdT and respective CEMRAs.
Due to modern surgical and medical treatments there have been major improvements in the prognosis of patients with congenital heart disease (CHD) exposing to increasing prevalence amongst adults . Serial lifelong follow-up is essential in this patient population. Echocardiography is widely available, but image acquisition is operator and acoustic-window-dependent. Cardiac catheterization is a two-dimensional (2D) invasive procedure, which is associated with radiation exposure. Even though procedure related mortality and complication rates were reduced to very low rates over the past decades, further risk reduction using non-invasive alternatives would be desirable .
Cardiovascular magnetic resonance imaging (MRI) is an established radiation-free non-invasive tool for the diagnosis and follow-up of patients with CHD . For example, non-ECG-gated breath-hold first-pass contrast-enhanced magnetic resonance angiography (CEMRA) using extravascular contrast agents, like Gadopentetate-dimeglumine (GdD), has become widely accepted in the diagnosis of vascular disease over the past years . Therefore diagnostic cardiac catheterization can frequently be avoided . First-pass techniques are limited due to the relatively fast diffusion of currently used contrast agents into the extravascular space. Alternatively, free-breathing high-resolution three dimensional (3D) datasets of the thorax and upper abdomen can be acquired using navigator gated and ECG-triggered 3D steady-state free precession sequences with a T2 preparation prepulse (T2-SSFP) . As these sequences are corrected for cardiac and respiratory motion they are an attractive diagnostic approach in this group of patients. They have been previously successfully used without contrast agents . However, this technique would benefit from increased blood pool contrast as detailed imaging of small cardiovascular structures such as coronary arteries or parts of the pulmonary vasculature is hampered by limited contrast between the blood pool and extravascular tissue/fluid . Gadofosveset-trisodium (GdT) is an intravascular MR contrast agent, potentially allowing imaging of vascular structures with higher contrast to noise ratios .
The aim of this study was to test the following hypothesis: A contrast agent specific sequence design (IR-SSFP) in combination with an intravascular contrast agent (GdT) brings together advantages of first-pass imaging with high signal from the intravascular contrast bolus and motion compensated high-resolution isotropic MR imaging using T2-SSFP sequences.
Always on Time
Marked to Standard
The purpose of this study was to compare image quality and diagnostic performance of a contrast agent specific inversion-recovery SSFP sequence (IR-SSFP) in combination with an intravascular contrast agent (Gadofosveset-trisodium, GdT) to a commonly used T2prepared SSFP sequence (T2-SSFP) with an extravascular (Gadopentetate-dimeglumine, GdD) and intravascular (GdT) contrast agent in patients with congenital heart disease.
Materials and Methods
The study was partly supported by Bayer-Schering-Pharma including provision of contrast agents and scan time. Bayer-Schering-Pharma had no control of inclusion of any data and information that might present a conflict of interest.
Study design and population
The study was approved by the local ethics committee (Guy's NHS Research Ethics Committee. London, UK) and was registered with the Medicines and Healthcare-Products-Regulatory-Agency (MHRA-Study-No. 28482/0002/001-0001, EudraCT-No. 2006-007042). Twenty-seven patients with CHD were prospectively enrolled (male: age 21-60 years, median 32 years; female: age 24-48 years, median 33 years, all: age 21-60 years; median 32 years, no significant difference in age was found between the male and female patient population using an unpaired t-test (p > 0.05)). Written consent was obtained and medical history was assessed in all subjects. Only individuals with a history of CHD listed for a routine MRI examination and therefore with no contraindication to MRI were included in this study. All subjects had no contraindications for the use of MR contrast agents. Patients were given a take-home informed consent for MR contrast media. Within a follow-up period of more than one year no side effects were noted in any patient. Four patients declined to take part in the second MRI examination without stating a reason.
Patients were investigated twice. On day 1, Gadopentetate-dimeglumine (GdD, MagnevistÂ®, Bayer-Schering-Pharma, Germany; max 0.2 mmol/kg; max volume 40 ml) and on day 2 (> 24 hours apart but within 6 days after administration of GdD) Gadofosveset-trisodium (GdT, VasovistÂ®, Bayer-Schering-Pharma, 0.03 mmol/kg) was injected. All examinations were performed on a 1.5T clinical MR scanner (Achieva, Philips Healthcare, Best, The Netherlands) with a 32-element cardiac coil.
After contrast agent injection (day 1 GdD, day 2 GdT, flow: 2 ml/s) CEMRA was performed requesting the patient to suspend breathing (end-expiration) for first-pass data acquisition (Table 1). This scan was followed by a respiratory navigator-gated and ECG-triggered T2-SSFP sequence (Table 1). On day 2 the same scans were repeated and an additional inversion-recovery (IR) preparation-prepulse SSFP sequence (IR-SSFP) was applied, comparable to sequences previously reported for coronary MRA .
The optimal inversion-time (TI, 260-280 ms) to suppress extravascular tissue was determined using a Look-Locker-sequence . The Look-Locker technique is a MRI pulse sequence for the measurement of spin-lattice T1 relaxation times. It is used in combination with inversion recovery sequences (clinically used for myocardial infarction imaging) to optimize the contrast between the signal from the contrast agent and the surrounding tissue. In this study it was used to determine the optimal inversion-time (TI, 260-280 ms) to minimize signal from extravascular tissue.
Image processing and reformatting was performed (>10 years experience, G.F.G) using commercially available analysis software (View Forum, Philips Healthcare, Best, The Netherlands).
Quantitative image analysis
Contrast to noise Ratio
Regions of interest (ROIs) were defined to determine the signal (S) from blood (SBlood) and reference tissue (SMyocardium). Noise (N) was determined by the standard deviation in the respective ROIs, as parallel imaging was used (sensitivity encoding for fast MRI (SENSE) (vendor-specific) ). ROIs areas were adjusted to the analyzed vascular bed at 7 levels for the aorta and the main pulmonary artery (measurements were performed by M.R.M.). CNR with corresponding ROI areas (LPA: 7.4Â±1.6 cm2, Aorta: Values given in Supplementary Table 2).
CNR was defined as described in equation 1 .
Vessel wall sharpness-, length and area
In order to compare data from vascular structures obtained with different scanning techniques a custom made analyzing tool ("Soap-Bubble") was used . This tool allows comparison of datasets for objective quantitative analysis of contrast to noise ratio (CNR), vessel length, sharpness and area. As earlier described by Botnar et al. , the local vessel sharpness can be obtained utilizing a Deriche algorithm . In brief, this algorithm calculates an edge image using a first-order derivative of the input image. The local value in a Deriche image represents the magnitude of local change in signal intensity. A vessel sharpness of 100% refers to a maximum signal intensity change at the vessel border. A lower edge value is consistent with inferior vessel sharpness. For the identification of the vessel edges along the path, a semi-automatic "vessel tracking" algorithm is used. Using this algorithm, the location of the vessel border and sharpness of the user-specified vessel segment is defined semi-automatically.
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
For quantitative assessment of the pulmonary artery and the aorta images were first reformatted using the "soapbubble" tool . The aorta and left pulmonary artery were chosen for analysis because these vascular structures were imaged with all imaging sequences (CEMRA, IR-SSFP, T2-SSFP) and contrast agents (GdD, GdT) investigated in this study. Additionally, the aorta and the pulmonary vasculature are frequently imaged in patients with acquired and congenital heart disease.
Vessel wall sharpness of the aorta and the left-pulmonary-artery (LPA) were measured as described . Furthermore the maximal true visible length of the LPA was assessed in each subject. Cross-sectional areas were measured at 7 different corresponding levels of the aorta for each sequence used.
Qualitative Image Analysis
Consensus reading was performed for image quality scoring by 2 readers. Prior to the analysis, a trial assessment of 5 separate MR images (for all MR sequences) for quality assurance was performed together. Subsequently, the 2 readers (>10 and >5 years experience in pediatric cardiology and pediatric cardiac MR) analyzed all images independently in a blinded and random order. Disagreements were discussed before a single final grade was given. The image grading system was adopted from McConnell et al. (Table 2, ). Visual inspection of image quality was performed for intracardiac, extracardiac arterial and venous structures. The presence of any cardiac or vascular abnormality, artifact, or incidental finding was recorded for each individual cardiac segment.
Consensus reading, as described above, was also performed to determine diagnostic information. Diagnostic information derived from IR-SSFP with GdT was compared to the currently used T2-SSFP with GdD and GdT and respective CEMRAs.
Two groups were defined:
1. Exclusion of pathologies: All clinically relevant pathologies that could be excluded with diagnostic image quality (Image quality score 3-5, see Table 2).
2. New unsuspected pathologies: All new diagnoses with diagnostic image quality (Image quality score 3-5, see Table 2) relevant to the patient's assessment regarding CHD.
Information available from echocardiography (n=23), catheterization (n=4) or surgery (n=1) was used to confirm all diagnoses.
Variables are reported as meanÂ±standard deviation. Paired and unpaired t-tests were used to compare continuous variables, as appropriate. The Wilcoxon rank sum test was used to compare categorical variables. A p-value <0.05 was considered statistically significant.
Twenty-seven patients were enrolled in the study protocol. Twenty-three patients completed both scan sessions. Four patients declined to take part in the second MRI examination without stating a reason.
Quantitative image analysis
Using IR-SSFP with GdT, allowed for imaging with an improved CNR and vessel wall sharpness (Figure 1, Supplementary Figure 1, Supplementary Table 2) compared to T2-SSFP with GdD and GdT. Using IR-SSFP with GdT image acquisition was independent from bolus timing, as needed for the CEMRAs (Figure 1, Supplementary Figure 1). Cross sectional areas of the aorta were compared between all 3D-imaging techniques (CEMRA, T2-SSFP and IR-SSFP). Comparable results were found for T2-SSFP with GdD and GdT and IR-SSFP with GdT using respiratory-navigator-gating and ECG-triggering (Supplementary Table 2). Areas were larger in breath-hold non-ECG-triggered CEMRAs (Supplementary Table 2, all p<0.05).
Qualitative image analysis
All results are summarized in Supplementary Table 3. Comparing all 3D imaging techniques, lowest values for image quality were found using breath-hold first-pass non-ECG-triggered CEMRAs without significant difference between GdD (day 1) and GdT (day 2) (p=n.s.). Comparing T2-SSFP with GdD and GdT no significant differences could be found either.
In contrast, IR-SSFP with GdT resulted in a significantly higher image quality for all evaluated structures compared to T2-SSFP with GdD and GdT and respective CEMRAs (Supplementary Table 3, all p<0.05). If IR-SSFP is applied in combination with a extravascular contrast agent with only a short blood half-life, like GdD , the compound will rapidly extravasate into the extravascular space during image acquisition (Figure 3).
The diagnostic information gained using IR-SSFP with GdT was compared to T2-SSFP with GdD and GdT and respective CEMRAs (Figure 1-2, Supplementary Figure 1-2, Table 3).
Exclusion of relevant pathologies:
In fifteen patients relevant pathologies regarding CHD could be excluded for optimal treatment planning (Figure 1-2, Supplementary Figure 1-2, Table 3). An example of this category was a patient in whom severe aortic coarctation was suspected (Figure 1). IR-SSFP with GdT could not only delineate a complete obstruction of the aortic arch, but also showed a continuity of the aorta in this area (Figure 1). This reassured the interventionalist to perforate the aortic arch obstruction with a guidewire in the cardiac catheterization laboratory and to insert a covered stent. T2-SSFP with GdD and GdT and respective CEMRAs were not able to provide this information (Figure 1). IR-SSFP with GdT also showed advantages (Figure 1, Table 3) in judging the pulmonary vascular system due to better discrimination of pulmonary arteries and veins. The spatial relationship between arterial and venous structures and adjacent nonvascular structures such as myocardium or pericardial fluid was clearly delineated by using an intravascular contrast agent with an IR-prepulse. This improved image quality particularly for exclusion of sinus venosus defects (Figure 2, 65%, n=15/23). Other examples include imaging of the semilunar valves (39%, n=9/23).
Extravascular fluids and blood pool both result in bright signal in T2-SSFP with GdD and GdT. Therefore, in some patients a clear diagnostic delineation of the origin and course of the coronary arteries was only possible using IR-SSFP with GdT (Figure 2, 39%, n=9/23) as pericardial fluid was fully suppressed. If the ventricular septum was evaluated, IR-SSFP with GdT allowed for imaging with diagnostic image quality, due to the full suppression of the myocardium or membranous part of the ventricular septum while generating strong signal from the blood pool. Image quality provided by T2-SSFP with GdD and GdT was in some cases not diagnostic and did not allow excluding VSDs (Supplementary Figure 2, 52%, n=12/23). Similar results were observed in patients after Mustard or Fontan-operation for assessment of the interatrial tunnel (Table 3, 8%, n=2/23). Information available from other modalities (echocardiography (n=23), catheterization (n=4), surgery (n=1)) was used to confirm all diagnoses.
New unsuspected pathologies:
In five of the investigated patients (22%) a previously undetected diagnosis was found using IR-SSFP with GdT (Table 3). Previously unexpected bicuspid aortic valves (13 %, n=3/23) and a VSD (Supplementary Figure 2, VSD, 4%, n=1/23) were imaged by IR-SSFP with GdT but could not be delineated using T2-SSFP with GdD and GdT and respective CEMRAs. Results were consecutively confirmed by echocardiography. A partial anomalous pulmonary venous return (PAVPR) of the right and left upper pulmonary veins was diagnosed using IR-SSFP with GdT but was missed by first-pass CEMRA due to suboptimal timing of the contrast bolus in a patient having difficulties to hold her breath. Cardiac catheterization also missed the diagnosis as it was targeted to image an aortic arch obstruction. T2-SSFP with GdD and GdT raised suspicion of the diagnosis of PAPVR but due to limited delineation of the intrapulmonary vasculature (Figure 1) definitive diagnosis was only provided using IR-SSFP with GdT. Repeated CEMRA with optimal contrast bolus timing to the pulmonary phase and subsequent cardiac catheterization confirmed the diagnosis (Figure 1, 4%, n=1/23).
Information available from other modalities (echocardiography (n=23), catheterization (n=4), surgery (n=1)) was used to confirm all diagnoses.
In this prospective study, we proposed a single injection of an intravascular contrast agent (GdT) in combination with an IR-SSFP sequence for improved assessment of cardiovascular morphology in patients with CHD. IR-SSFP with GdT improved image quality and diagnostic performance compared to T2-SSFP with GdD and GdT and respective CEMRAs. No qualitative or quantitative differences were found between the two contrast agents in the same patient using T2-SSFP and CEMRA sequences.
First-pass CEMRA is an approach used in clinical practice to image intrathoracic vessels. One of its main advantages is the high contrast generated by the intravascular contrast agent bolus while extravascular tissue is suppressed. However, there are several limitations. 1) Acquisition time is restricted to one breath hold, limiting the spatial resolution (usually non-isotropic voxels) and field of view . This is also relevant if multiphase imaging is used. 2) CEMRA relies on patients holding their breath to compensate for respiratory motion. If patients cannot fully comply with breathing commands artifacts from chest motion can be a result. 3) Due to the requirement for optimal patient specific bolus timing while investigating highly complex cardiovascular anatomy, standardization and reproducibility of follow-up MRIs can be challenging . Bolus injection should not be repeated, due to the limited amount of contrast agent that should be given. 4) Imaging of intracardiac structures is limited, as CEMRA sequences are not triggered to cardiac motion.
ECG-triggered and respiratory-navigator-gated 3D MR imaging techniques such as T2-SSFP are another approach to image intrathoracic structures and have shown some advantages in the clinical environment. 1) They are easy to plan, as they do not need to be adjusted to specific cardiovascular structures and usually cover the complete thorax. 2) ECG-triggering and respiratory-navigator-gating is applied for compensation of respiratory and cardiac motion. This allows the evaluation of intracardiac structures. 3) Isotropic voxels can be acquired, allowing highly standardized and reproducible evaluation of data sets by post-processing. 4) Arterial and venous anomalies can be imaged simultaneously, without the need for precise bolus timing . A shortcoming of this technique is the reduced contrast between vascular structures and surrounding soft tissues and fluids. This can lead to limitations for imaging structures like the interventricular septum, the pulmonary vasculature and coronary arteries , as these structures may sometimes not be clearly distinguishable from surrounding tissue, fluids or blood.
A single injection of an intravascular contrast agent (GdT) in combination with a contrast agent specific sequence design (IR-SSFP) brings together the advantages of first-pass imaging with high signal form the vasculature with motion compensated high-resolution isotropic imaging using T2-SSFP. In this study, IR-SSFP with GdT allowed imaging with high signal from the blood pool, while signal from extravascular tissues and fluids was suppressed. This approach improved the image quality of all investigated intrathoracic structures significantly compared to T2-SSFP with GdD and GdT and respective CEMRAs. All cardiac segments including the course of the coronary artery system, the ventricular septum and the arterial and venous pulmonary vasculature were assessed with higher contrast from the blood pool and suppression of extravascular structures like pericardial fluid.
The arterial and venous system was imaged simultaneously with high intravascular contrast using the IR-SSFP with GdT without the need for precise contrast bolus timing. If sequence acquisition fails due to patient movement or technical issues, scans can be repeated without the need for an additional injection of a contrast agent, due to the long blood half live of the intravascular contrast agent (t1/2alpha~29mins; t1/2beta~16h18mins) .
Using IR-SSFP with GdT all information can be obtained during a single, high-resolution isotropic motion-compensated acquisition, independent from bolus timing. The isotropic property of data sets allows reformatting of any desired imaging plane, thus precise morphological evaluation may be performed with post-processing. This can have diagnostic implications. The exclusion of relevant pathologies (65%, n=15/23) and the diagnosis of new unsuspected pathologies (22%, n=5/23) indicate the potential clinical benefit of using IR-SSFP with GdT. The results of this study demonstrate that IR-SSFP with GdT could be superior in diagnosing intra- and extracardiac vascular and cardiac abnormalities in patients with CHD compared to the currently used combination of T2-SSFP with GdD and GdT and respective CEMRA. Other investigators confirmed in a retrospective study the superior image quality of the navigator-gated and ECG-triggered IR-SSFP compared to first-pass breath-hold CEMRA using GdT . T2-SSFP and extravascular contrast agents such as GdD were however not available for comparison in this study .
If IR-SSFP is applied in combination with a extravascular contrast agent with only a short blood half-life, like GdD , the compound will rapidly extravasate into the extravascular space during image acquisition. Even if the IR pulse is timed to achieve a high intravascular signal after contrast agent injection and imaging is started immediately after contrast agent injection (while the contrast agent is still in the distribution phase), a significant portion of the contrast agent will have diffused out of the intravascular space before the imaging sequence is complete . This will result in a mixed contrast between intra- and extravascular space. The compound may accumulate in e.g. the cardiac valves, scar tissue in the myocardium ("myocardial late enhancement") or in atherosclerotic vessel wall, either as a consequence of an increased distribution volume, endothelial permeability or neovascularization .
The improved contrast between intra- and extravascular structures, using IR-SSFP in combination with GdT, may have resulted from the higher relaxivity of GdT when bound to albumin in combination with its prolonged intravascular half life , compared to currently used extravascular contrast agents. Additionally, the IR-SSFP sequence applied was heavily T1 weighted and thus specifically useful to highlight regions with shortened T1 times. Therefore this sequence design was well suited for the detection of gadolinium based contrast agents. The T2-SSFP sequence used was mainly T2 weighted, because of the T2prep prepulse applied, and some additional T2/T1 weighting resulted from the SSFP readout. The T2-SSFP sequence we used was therefore not highly sensitive for the detection of T1 lowering contrast agents but rather for the detection of fluids including blood.
No conclusion from our data is applicable to patients with cardiac arrhythmia. Motion artifacts cannot be addressed with the improved blood pool signal and sequence design presented in this study. Even though significant improvement in imaging of cardiac valves was achieved using GdT, echocardiography is still considered to be the reference standard for anatomical imaging . The role of intravascular contrast agents in imaging of myocardial delayed-enhancement is still subject to investigations and was not addressed in this study. In case information on myocardial scar imaging is needed, GdD may be the preferred imaging agent at the moment. As the use of MRI methodology and contrast agent administration was not randomized due to technical reasons, it could not be determined to what degree the order of administration confounded the power of the method.
A single injection of an intravascular contrast agent (GdT) in combination with a contrast agent specific sequence design (IR-SSFP) combines the advantages of first-pass CEMRA imaging with high signal form the vasculature with respiratory and cardiac-motion compensated high-resolution isotropic T2-SSFP sequences. In this study IR-SSFP with GdT improved image quality and diagnostic performance for the assessment of cardiovascular morphology in patients with complex CHD compared to T2-SSFP with GdD and GdT and respective CEMRAs. This may broaden the use of cardiac MRI and improve follow-up in patients with CHD.
Figure 1: 43 year old patient with aortic coarctation and drainage of the left and right upper pulmonary veins to the innominate vein.
The volume rendered pulmonary phase of the first-pass CEMRA (A1) using GdT demonstrates the drainage of the stenotic left upper pulmonary vein curling around the left pulmonary artery (LPA) to the innominate vein (IV) (white box on volume rendered image (A1) and multiplanar reformatted plane through this area (A2). The same stenotic area is displayed on multiplanar reformatted images using T2-SSFP (A3). Highest contrast from the blood pool was achieved using IR-SSFP (A4, zoomed area with arrows in white box). With CEMRA the atretic distal aortic arch (B1, dotted arrow on x-ray angiography) was insufficiently imaged due to bolus timing and an insufficient breath hold of the patient (CEMRA, B2). The atretic segment of the aorta was imaged with T2-SSFP (B3). However, highest contrast from the blood pool was achieved using IR-SSFP (B4). Ao: aorta; daA: distal aortic arch; dA: descending aorta; IV: innominate vein; PA: pulmonary artery; LPA: left pulmonary artery.
Figure 2: Imaging of the left coronary artery and proximal pulmonary vessels. Multiplanar reformatted (MPR) using T2-SSFP (A1, B1, C1) and IR-SSFP with GdT (A2, B2, C2). Pericardial fluid and extravascular tissue (black arrowheads) using the T2-SSFP obviated the clear definition of the origin of the left coronary artery (LCA, A1, B1). Suppression of perivascular fluid and tissue using IR-SSFP allowed a clear delineation of the course of the LCA (A2, B2). C1 and C2 is showing an axial plane of the ascending aorta (AAo), superior vena cava (SVC), right pulmonary artery (RPA) and right upper pulmonary veins (*). The presence of perivascular fluid (C1, white arrowheads) in the T2-SSFP mimics connections between the SVC, RPA and right upper pulmonary vein (C1, arrows). Due to suppression of perivascular fluid (arrowheads) using IR-SSFP with GdT these connections could be excluded (C2).
Figure 3: T2-SSFP and IR-SSFP with the use of an extra (GdD)- and intravascular (GdT) contrast agent:
Transversal T2-SSFP (A1, B1) and IR-SSFP (A2, B2) images in the same patient after the injection of GdD (A1, A2) and GdT (B1, B2). Using T2-SSFP (A1, B1) no obvious difference in image quality using the two contrast agents can be found. Using T2-SSFP myocardial tissue is not fully suppressed and fluid around the spinal cord is bright (arrow-heads). Using IR-SSFP with GdD (A2) fibrous tissues enhance (like the aortic valve, arrows), while the blood pool yields no signal, reflecting the fast washout of the extravascular contrast agent (GdD) from the blood pool. Using IR-SSFP with GdT (B2), an intravascular contrast agent, a strong signal from the blood pool can be appreciated. Simultaneously, extravascular tissue (fibrous tissue, myocardium, spinal or pericardial fluid, arrow-heads) and fluids not containing GdT are suppressed.
Supplementary Figure 1: Comparison of thoracic 3D reconstruction, aortic coarctation and left pulmonary artery using different MRI imaging sequences.
First-pass CEMRA (A), T2-SSFP (B) and IR-SSFP of the thorax (C) with segmented aortic coarctation (A1, B1, C1) and LPA (A2, B2, C2). In all shown images GdT was used. With first-pass CEMRA (A) cardiovascular structures are imaged within one breath hold without ECG-triggering (blurry vascular borders; A1, A2). Using T2-SSFP blurring is reduced (B1, B2), extravascular fluid is however not suppressed (B: white arrow heads demonstrating spinal fluid). Using IR-SSFP with GdT (C) vascular structures give a high signal while the surrounding tissue is suppressed (C1, C2) and blurring is reduced compared to CEMRA (A1, A2).
Supplementary Figure 2: Imaging of the interventricular septum.
A: Multiplanar reformatted (MPR) aortic phase of the first-pass CEMRA (A1), T2-SSFP (A2) and IR-SSFP (A3) with GdT. Only IR-SSFP with GdT allowed a clear delineation of a small membranous subaortic VSD (A3, arrow), which was confirmed by echocardiography. B: MPR showing the ventricular septum (arrowheads) using T2-SSFP (B1) and IR-SSFP (B2) with GdT. Due to the suppression of tissue signal, IR-SSFP with GdT clearly delineated the interventricular septum from the blood pool with diagnostic image quality and therefore septal defects (arrows) could be excluded in this area. The diagnosis was confirmed by echocardiography.