4d flow rm imaging: a new diagnostic tool for congenital heart

Revista Chilena de Radiología. Vol. 17 Nº 3, 2011; 134-140.

134

Dra. Natalia Pinochet R y cols.

4D flow RM imaging: a new diagnostic tool for congenital heart disease

IntroductionCardiac magnetic resonance (CMR) imaging is a

useful complementary tool for the diagnosis of patients with congenital heart disease (CHD)(1). This technique has replaced other imaging modalities, thus becoming the gold standard procedure for assessment and/or monitoring of some CHD, such as Tetralogy of Fallot or single ventricle.

CMR imaging is an ideal diagnostic procedure for pediatric patients with CHD, since many of them

present complex cardiovascular anatomies and re-quire numerous tests to monitor their condition and treatment. The technique offers important advantages: it is noninvasive, uses neither ionizing radiation nor iodinated contrast media. It also allows an accurate analysis of anatomy and quantification of cardiovas-cular function and flows.

In CMR the most widely used technique for quanti-fication of cardiovascular flows is the two-dimensional (2D) phase-contrast imaging. This method allows

Abstract: Purpose. To demonstrate the utility of 4D flow MR imaging for analyzing blood flow patterns and flow distribution in patients with congenital heart diseases. Methods: Six patients with congenital heart diseases were scanned using a standard cardiac MRI protocol, according to their condition. Additionally, 2D flow sequen-ces of the great vessels, and a 4D flow sequence covering the entire heart were acquired. Flow patterns were visualized by using vector fields, streamlines and particle traces. Results: 4D flow technique depicted vortices and helical flow in the pulmonary artery of most patients, as well as in the aorta and superior vena cava of one patient with corrected aortic coarctation and a levoatrial cardinal vein. Conclusion: 4D flow MR imaging enables the identification of flow patterns difficult to detect with other diagnostic modalities. Comprehensive evaluation of flow patterns might help to understand the hemodynamic consequences of congenital heart diseases and their surgical procedures.Keywords: 4D flow, Congenital heart defect, Flow patterns, Magnetic resonance imaging.

Resumen: Objetivo. Demostrar la utilidad de 4D flow para el análisis de patrones y distribución de flujos en pacientes con cardiopatías congénitas. Métodos: Seis pacientes con cardiopatías congénitas fueron escaneados con un protocolo de resonancia magnética cardíaca estándar. Además se incluyeron secuencias de flujo 2D en los principales vasos del tórax y una secuencia 4D flow que cubría todo el corazón. Para la visualización de los patrones de flujo se emplearon vectores de velocidad, líneas de flujo y trazadores de partículas. Resultados: 4D flow reveló vórtices y hélices en la arteria pulmonar de la mayoría de los pacientes, y en la aorta y vena cava superior de un paciente con coartación aórtica reparada y vena cardinal levoatrial. Conclusiones: 4D flow permite identificar patrones de flujo en pacientes con cardiopatías congénitas, difíciles de observar con otros métodos diagnósticos. La evaluación de patrones de flujo podría contribuir a comprender las consecuencias hemodinámicas de diferentes cardiopatías congénitas.Palabras claves: 4D flow, Cardiopatía congénita, Patrones de flujo, Resonancia magnética.

Pinochet N et al. 4D-Flow: una nueva herramienta de diagnóstico para cardiopatías congénitas. Rev Chil Radiol 2011; 17 (3): 134-140.Correspondence to: Dr. Sergio Uribe A. / [email protected] received August 5, 2011. Accepted for publication August 16, 2011.

Drs. Natalia Pinochet R(1), Pablo Bächler S(1), Cristián Tejos N(2,3,6), Gerard Crelier(4), Rodrigo Parra R(5), José Miguel Allendes C(5), Pablo Irarrázaval M(2,3,6), Sergio Uribe A(3,5,6).

1. School of Medicine, Pontificia Universidad Católica de Chile.2. Department of Electrical Engineering, Pontificia Universidad Católica de Chile.3. Biomedical Imaging Center, Pontificia Universidad Católica de Chile.4. Institute for Biomedical Engineering, University and ETH, Zurich, Switzerland.5. Department of Radiology, Pontificia Universidad Católica de Chile.6. Biomedical Engineering Group, Pontificia Universidad Católica de Chile.

Revista Chilena de Radiología. Vol. 17 Nº 3, año 2011; 134-140.

135

TÓRAX

velocity encoding of hydrogen protons (spins) present in the patient (particularly in the blood) through the application of bipolar magnetic field gradients(2). These bipolar gradients transform the velocity of the spins. The magnitude of the observed phase is proportional to the area of gradients, the time required to appli-cation of gradients, and particularly, to the velocity of the spins. These gradients may be applied to encode velocity in any spatial direction. For cardiac blood flow quantification, a sequence perpendicular to the target vessel is usually acquired, with bipolar gradients applied in the same direction as slice selection gradients. Cardiovascular flow curves are obtained by selecting a region of interest in the desired vessel. The average velocity multiplied by ROI yields the amount of blood flow at a certain phase of the cardiac cycle. The total volume of blood flowing through that vessel during the cardiac cycle is obtained by adding blood flows registered during all stages of the cardiac cycle. This technique is widely used and has been va-lidated in patients with CHD(3-4). However, data acquisition is difficult; it requires an experienced user to determine the location of sequences, and if the plane is not completely perpendicular to the analyzed vessel, flows may be underestimated(5-6). 4D flow technique is an extension of the 2D phase-contrast MRI that allows obtaining the velocity of the spins in all directions and throughout the complete cardiac cycle(7-9). Although this technique has proven very useful for understanding certain cardiovascular conditions, it has not gained wi-despread application in clinical practice. This is mainly due to long data acquisition times, and to the difficulty in processing and interpreting data. In recent years, faster magnetic resonators along with exceptionally user-friendly image processing softwares have given this technique a renewed value as a useful method for understanding the dynamics of blood flow. Most investigations on blood flow have focused on the understanding of aortic hemodynamics under different conditions(3,10-13). However, little research on the application of this technique in patients with CHD has been conducted.

The aim of this paper is to demonstrate the use-fulness of the 4D Flow imaging method in analyzing blood flow patterns and distribution in patients with CHD.

Material and methods4D flow Technique

4D flow technique involves applying consecutive bipolar gradients for velocity encoding in three ortho-gonal directions. Like in 2D phase-contrast MRI, it is necessary to define the maximum velocity expected in the volume to be encoded. This parameter, refe-rred to as VENC, can be defined independently for each direction. It is very important to get an accura-tely defined VENC, because if it is less than actual velocities, velocity measurements will be distorted. By contrast, if VENC is much greater than the actual velocity, measurements will be less accurate(11), sin-ce discrimination between slightly different velocity magnitudes will be hindered.

4D flow imaging method allows acquisition of two different types of images, i.e., an anatomical image and a velocity encoded image, which in turn reveals three flow directions in three separate images (Figure 1).

The 4D flow sequence may be applied either in a single slice or in one volume. Volumetric whole-heart sequence takes about 10 to 15 mi-nutes, even when using rapid acquisition tech-niques such as k-t SENSE and kt BLAST(14). A Despite the long acquisition time this method can provide hemodynamic information about both the whole heart and great vessels in a single scan(15), under the same physiological conditions.

Four-dimensional (4D) flow of the whole heart allows a posteriori data reorganizion in any direction; therefore, it is feasible to quantify blood flows of all large vessels present in the volume acquired. In addition to blood flow quantification, 4D-flow sequences enable semi-quantitative evaluation of flow patterns such as vorticity or helicity, determination of blood trajectory, calculation of cardiac shunts due to arterio-venous connections, and estimation of the wall shear stress in large blood vessels(16).

Figure 1. (A) Anatomical image, and 3 velocity-encoded images in 3 directions (b) Left to right, (c) botom to top (d) and anteroposterior view. Zero velocity appears medium gray, whilst positive and negative velocities (depending on the direction previously established) are proportionaly encoded in light and dark gray, respectively.

1a 1b 1c 1d


136


Data AcquisitionTo demonstrate the usefulness of this technique,

6 patients with CHD (4 men and 2 women, mean age 19.3 years, range: 10-46 years) were scanned with a 1.5T Philips resonator (Philips Healthcare, Best, The Netherlands). Each patient underwent the standard protocol for RMC according to type of pathology; in most cases this included 2D phase-contrast sequences of the great vessels of the chest: Aorta (Ao), pulmonary artery (PA) and its main branches (APD API), infe-rior vena cava (IVC) and superior vena cava (SVC). Additionally, in each patient a 4D-flow sequence of the whole heart and of mediastinal great vessels was obtained with the following acquisition parameters: 50 slices, excitation angle of 6°, 38 ms of temporal resolution, spatial resolution of 2.5 mm, and VENC = 200-350 cm / s.

Analysis and quantificationFor analysis of blood flow patterns various gra-

phic tools were used: velocity vectors, flowlines or streamlines and particle tracers (Figure 2). To this end, a GT Flow software (Gyrotools LLC, Zurich, Switzerland) was used.

Velocity vectorsThis method allows visualization of the velocity

component parallel to the image plane. This data re-presentation is useful for local analysis of vortices(17).

Flowlines or streamlinesThey represent imaginary lines oriented according

to the blood velocity field in a given cardiac phase(18). These imaginary lines provide a 3D perspective of the velocity vector, enabling creation of maps of blood connection; however, it is a mere static representation.

Particle tracersParticle tracers enable to visualize the path that

blood takes along the cardiac cycle(19). This tech-nique involves selecting a plane of interest and then drawing a region of interest (ROI) in a vessel from which fictitious particles will be emitted, which will be followed throughout the cardiac cycle.

ResultsDiagnoses of study patients were as follows: 1)

Partial anomalous pulmonary venous drainage (PA-PVD), and interauricular communication; 2) repair of aortic coarctation, and levoatrial cardinal vein (abnormal venous connection between the left atrium and left brachiocephalic vein); 3) and 4) pulmonary atresia, status post one and a half ventricle repair with Glenn anastomosis 5) Fontan-type repair; and 6) Transpo-sition of the great vessels, status post arterial switch operation.

4D flow technique revealed blood flow patterns

as vortices and helices in the pulmonary artery in most of the patients, and in the superior vena cava of one patient with levoatrial cardinal vein. In velocity vectors (Figure 3a) flow patterns are seen in patients undergoing one and a half ventricular repair and Glenn-type anastomosis during three cardiac phases. By using particle tracers it is possible to observe that vortices and retrograde flows occur in the pulmonary artery. Connectivity Maps (Figure 3b) also enabled visualization of these vortices as well as blood flow distribution from pulmonary artery into its branches. Surprisingly, in both patients with Glenn anastomosis most of the flow from SVC is diverted into the PA trunk as well as into the right ventricle during diastolic phases (Figure 4).

In Figure 5, a helical flow in the SVC of patient with repaired coarctation of the aorta and levoatrial cardinal vein secondary to communication between the left atrium and left brachiocephalic vein may be observed.

In the same patient, a turbulent aortic flow in the repaired coarctation area was also observed (Figure 5b and 5c). It has been suggested that blood flow patterns similar to those detected in the aorta of this patient could contribute to subsequent development of aortic aneurysms(20).

Figure 2. Examples of methods for visualizing blood flow patterns in the aorta and pulmonary artery using (a) velocity vectors, (b) flowlines or streamlines, and (c) particle tracers.

2a

2b

2c


137

TÓRAX

Figure 4. In patients with one and a half ventricle repair and Glenn anastomosis it was found that blood flow from the superior vena cava (SVC) mainly drains into the pulmonary artery (PA) and right ventricle (RV).

Figure 3. (A) Velocity vectors showing blood flow patterns in a patient who underwent Glenn anastomosis. Such vectors reveal vortices (arrows in columns 1, 2, and 3) in the pulmonary artery (PA) and retrograde blood flow during diastole (arrow in column 4). (B) Connectivity maps also show vortices (indicated by the arrow in all columns), and blood flow distribution of left pulmonary artery (LPA) and right pulmonary artery (RPA).

DiscussionWhole heart 4D Flow MRI not only allows the

quantification of cardiovascular flows(15,21), but also makes it possible to analyze the complex blood flow dynamics. Through velocity vectors, flowlines

3a

3b

or streamlines and particle tracers a qualitative and quantitative analysis of flow patterns and their distri-bution in the cardiac chambers and great vessels of patients with CHD is feasible.

In this group of patients, 4D flow imaging technique evidenced the presence of vortices in PA and SVC in one patient with levoatrial cardinal vein. Abnormal aortic vortices in patient with repair of aortic coarctation as well as in patient with arterial switch were also detected. An important drawback of 4D-flow technique is the compensation for respiratory motion, i.e., the pro-cess to prevent or correct generation of artifacts in MR imaging caused by patients breathing. Some authors have demonstrated that some techniques for minimizing distortions caused by respiration are compatible with 4D flow MRI(15, 21). Nevertheless, most of these methods use a respiration- triggered scan, which implies discarding data obtained from incorrect respiratory positions and performing a new acquisition of data. Consequently, this technique involves a signi-ficantly increased scanning time.In some patients with stenotic areas of the PA or the Ao, it is necessary to define a fairly high VENC to prevent errors of velocity quantification. Nevertheless, this is detrimental to the accuracy of measurements in vessels exhibiting low blood velocity.In children with high heart rates, standard 4D-flow temporal resolution (~ 38 ms) might be insufficient to detect short-duration blood flow patterns. This could affect assessment accuracy during peak systolic velocity(10).Pediatric cardiac structures are smaller, so it is necessary to improve


138


the spatial resolution (of 1.5 mm3), thus increasing the scanning time. This is especially necessary for visualizing pulmonary veins difficult to be identified due to their small size.

By adding the 4D-flow sequence to the clinical protocol, scanning time increases by10 to15 minutes, which added to the 30-minute data processing time might be incompatible with clinical practice. This appears to be entirely impractical with children in critical conditions, who require general anesthesia and mechanical ventilation.

Automation of definition of ROI of great vessels and cardiac chambers could reduce this time con-siderably, making 4D- flow and 2D phase contrast MRI comparable in terms of acquisition times. Thus, 4D-flow sequence could replace 2D sequences, and therefore total test time could be reduced(22).

Application of 4D-flow MRI in different CHD could help to understand the relationship between hemodynamics and the behavior of vascular walls, i.e., the possibility of understanding not only the re-lationship between patients with bicuspid aortic valve and development of aortic aneurysms(23), but also development of aneurysms in patients undergoing aortic coarctation repair(20,24,25).

It has been documented that some patients with single ventricle after Fontan surgery have had a poor outcome after surgery. Although this diagnostic pro-cedure has been applied over the last four decades, it remains controversial which is the most efficient pul-monary connection from a hemodynamic point of view. In this case, 4D- flow technique could reveal blood flow patterns and distribution in order to clarify this issue(26).

Future ProspectsNew research works and developments in 4D-

flow technique promise major advancements in the understanding of cardiovascular hemodynamics.

It is necessary to fully understand blood flow pat-terns in healthy volunteers in order to understand the blood flow dynamics in patients with CHD. Most studies in this area have focused on the Ao(10), describing a great variety of blood flow patterns.

Although not shown in this work, 4D-flow MRI also enables estimation of other parameters associated with the cardiovascular system, e.g., wall shear stress in blood vessels(27), creation of blood pressure maps(28), and new ways to quantify the helicity and vorticity of blood flows(16).

In conclusion, the 4D-flow technique represents a breakthrough for the understanding of cardiovascular blood flow dynamics in patients with CHD. New ad-vances in both data acquisition and processing could become an important contribution for implementation of this technique in day-to-day clinical practice.

AcknowledgementsThe authors wish to acknowledge the financial

support granted by Fondecyt Projects 1100864 and 11100427, Anillo ACT79, and Vicerrectoría Académica de Investigación y Doctorado de la Pontificia Univer-sidad Católica de Chile; and to Philips Healthcare for joint working schemes implemented so far.

Bibliography1. Didier D, Ratib O, Beghetti M, Oberhaensli I, Friedli B.

Morphologic and functional evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging 1999; 10: 639-655.

2. Underwood SR, Firmin DN, Klipstein RH, Rees RS, Longmore DB. Magnetic resonance velocity mapping: clinical application of a new technique. Br Heart J 1987; 57: 404-412.

3. Bogren HG, Buonocore MH. 4D magnetic resonance velocity mapping of blood flow patterns in the aorta in young vs. elderly normal subjects. J Magn Reson Imaging 1999; 10: 861-869.

5a 5b 5c

Figure 5. (a) In the patient with repaired coarctation of the aorta and levoatrial cardinal vein (CV) a helical flow in the superior vena cava (SVC)(indicated by arrow), secondary to communication between the left atrium (LA) and left brachiocephalic vein (b and c), was detected. In the same patient, abnormal flows at repaired coarctation site were observed (See arrow).


139

TÓRAX

4. Beerbaum P, Körperich H, Barth P, Esdorn H, Gieseke J, Meyer H. Noninvasive quantification of left-to-right shunt in pediatric patients: phase-contrast cine magnetic resonance imaging compared with invasive oximetry. Circulation 2001; 103: 2476-2482.

5. Varaprasathan GA, Araoz PA, Higgins CB, Reddy GP. Quantification of flow dynamics in congenital heart disease: applications of velocity-encoded cine MR imaging. Radiographics 2002; 22: 895-905.

6. Higgins CB, Sakuma H. Heart disease: functional evaluation with MR imaging. Radiology 1996; 199: 307-315.

7. Moran PR. A flow velocity zeugmatographic interlace for NMR imaging in humans. J Magn Reson Imaging 1982; 1: 197-203.

8. Markl M, Draney MT, Miller DC, Levin JM, Williamson EE, Pelc NJ, et al. Time-resolved three-dimensional magnetic resonance velocity mapping of aortic flow in healthy volunteers and patients after valve-sparing aortic root replacement. J Thorac Cardiovasc Surg 2005; 130: 456-463.

9. Markl M, Chan FP, Alley MT, Wedding KL, Draney MT, Elkins CJ, et al. Time-resolved three-dimensional phase-contrast MRI. J Magn Reson Imaging 2003; 17: 499-506.

10. Hope TA, Markl M, Wigström L, Alley MT, Miller DC, Herfkens RJ. Comparison of flow patterns in ascending aortic aneurysms and volunteers using four-dimensional magnetic resonance velocity mapping. J Magn Reson Imaging 2007; 26: 1471-1479.

11. Hope TA, Herfkens RJ. Imaging of the thoracic aorta with time-resolved three-dimensional phase-contrast MRI: a review. Semin Thorac Cardiovasc Surg 2008; 20: 358-364.

12. Frydrychowicz A, Markl M, Harloff A, Stalder AF, Bock J, Bley TA, et al. Flow-sensitive in-vivo 4D MR imaging at 3T for the analysis of aortic hemodynamics and derived vessel wall parameters. RoFo 2007; 179: 463-472.

13. Frydrychowicz A, Harloff A, Jung B, Zaitsev M, Wei-gang E, Bley TA, et al. Time-resolved, 3-dimensional magnetic resonance flow analysis at 3 T: visualization of normal and pathological aortic vascular hemodyna-mics. J Comput Assist Tomogr 2007; 31: 9-15.

14. Stadlbauer A, van der Riet W, Crelier G, Salomonowitz E. Accelerated time-resolved three-dimensional MR velocity mapping of blood flow patterns in the aorta using SENSE and k-t BLAST. Eur J Radiol 2010; 75: e15-21.

15. Uribe S, Beerbaum P, Sørensen T, Rasmusson A, Razavi R, Schaeffter T. Four-dimensional (4D) flow of the whole heart and great vessels using real-time respiratory self-gating. Magn Reson Med 2009; 62: 984-992.

16. Frydrychowicz A, Berger A, Russe MF, Stalder AF, Harloff A, Dittrich S, et al. Time-resolved magnetic re-sonance angiography and flow-sensitive 4-dimensional

magnetic resonance imaging at 3 Tesla for blood flow and wall shear stress analysis. J Thorac Cardiovasc Surg 2008; 136: 400-407.

17. Morbiducci U, Ponzini R, Rizzo G, Cadioli M, Esposito A, De Cobelli F, et al. In vivo quantification of helical blood flow in human aorta by time-resolved three-dimensional cine phase contrast magnetic resonance imaging. Ann Biomed Eng 2009; 37: 516-531.

18. Napel S, Lee DH, Frayne R, Rutt BK. Visualizing three-dimensional flow with simulated streamlines and three-dimensional phase-contrast MR imaging. J Magn Reson Imaging 1992; 2: 143-153.

19. Wigstrom L, Ebbers T, Fyrenius A, Karlsson M, Engvall J, Wranne B, et al. Particle trace visualization of intra-cardiac flow using time-resolved 3D phase contrast MRI. Magn Reson Med 1999; 41: 793-799.

20. Frydrychowicz A, Arnold R, Hirtler D, Schlensak C, Stalder AF, Hennig J, et al. Multidirectional flow analysis by cardiovascular magnetic resonance in aneurysm development following repair of aortic coarctation. J Cardiovasc Magn Reson 2008; 10: 30.

21. Markl M, Harloff A, Bley TA, Zaitsev M, Jung B, Weigang E, et al. Time-resolved 3D MR velocity mapping at 3T: improved navigator-gated assessment of vascular anatomy and blood flow. J Magn Reson Imaging 2007; 25: 824-831.

22. Nordmeyer S, Riesenkampff E, Crelier G, et al. Flow-sensitive four-dimensional cine magnetic resonance imaging for offline blood flow quantification in multiple vessels: A validation study. J Magn Reson Imaging 2010; 32: 677-683.

23. Hope M, Meadows A, Hope T, Ordovas K, Reddy G, Alley M, et al. Evaluation of Bicuspid Aortic Valve and Aortic Coarctation With 4D Flow Magnetic Resonance Imaging. Circulation 2008; 117: 2818-2819.

24. von Kodolitsch Y, Aydin MA, Koschyk DH, Loose R, Schalwat I, Karck M, et al. Predictors of aneurysmal formation after surgical correction of aortic coarctation. J Am Coll Cardiol 2002; 39: 617-624.

25. de Divitiis M, Pilla C, Kattenhorn M, Zadinello M, Donald A, Leeson P, et al. Vascular dysfunction after repair of coarctation of the aorta: impact of early surgery. Circulation 2001; 104: I-165-I-170.

26. Frydrychowicz A, Arnold R, Harloff A, Schlensak C, Hennig J, Langer M, et al. Images in cardiovascular medicine. In vivo 3-dimensional flow connectivity map-ping after extracardiac total cavopulmonary connection. Circulation 2008; 118: e16-17.

27. Cheng C, Tempel D, van Haperen R, van der Baan A, Grosveld F, Daemen MJAP, et al. Atherosclerotic lesion size and vulnerability are determined by patterns of fluid shear stress. Circulation 2006; 113: 2744-2753.

28. Tyszka JM, Laidlaw DH, Asa JW, Silverman JM. Three-dimensional, time-resolved (4D) relative pressure mapping using magnetic resonance imaging. J Magn Reson Imaging 2000; 12: 321-329.

4d flow rm imaging: a new diagnostic tool for congenital heart

Documents