- Email: [email protected]
Unenhanced steady state free precession versus traditional MR imaging for congenital heart disease
European Journal of Radiology 82 (2013) 1743–1748
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Unenhanced steady state free precession versus traditional MR imaging for congenital heart disease Dandan Chang a,∗ , Xiangquan Kong b,∗,1 , Xuhui Zhou a,2 , Shurong Li a,3 , Huanjun Wang a,4 a b
Department of Diagnostic Radiology, the First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, Guangdong, China Department of Radiology, the Affiliated Union Hospital, Huazhong University of Science and Technology, Wuhan, Hubei, China
a r t i c l e
i n f o
Article history: Received 23 July 2012 Received in revised form 1 February 2013 Accepted 25 March 2013 Keywords: Congenital heart disease Three dimensional steady state free precession Magnetic resonance angiography Image quality Diagnosis Diameter measurement
a b s t r a c t Purpose: To assess potential benefits of three dimensional (3D) steady state free precession (SSFP) magnetic resonance sequence for congenital heart disease (CHD). Materials and methods: Twenty consecutive patients with CHD (male:female ratio,14:6, mean age, 27.5 ± 8.5 years) underwent both 3D SSFP and traditional MR imaging (TMRI) [including two dimensional (2D) SSFP and contrast enhanced magnetic resonance angiography (CEMRA)]. Image quality and diagnosis were compared, and Bland–Altman analysis was used to evaluate consistency of 3D SSFP and CEMRA for diameter measurements. Results: A total of 35 intra and 81 extra cardiac anomalies were identified in all patients. The image quality of 3D SSFP and TMRI for either intra or extra cardiac anomalies of all patients scored ≥3, which allowed an establishment of diagnosis for all cases. The diagnostic sensitivity, specificity, and accuracy of 3D SSFP for the detection of intra cardiac anomalies were all 100%, whereas for extra cardiac anomalies they were 93.8%, 93.8%, 100%, respectively. Mean differences (3D SSFP minus CEMRA) for aorta and pulmonary arteries were 0.5 ± 1.2 mm and 0.0 ± 1.7 mm, respectively, showing good consistency of 3D SSFP and CEMRA for diameter measurements. Conclusion: 3D SSFP MRI can be an alternative image modality to TMRI for patients with congenital heart disease, especially for those who have renal insufficiency, breath-hold difficulty or who are allergic to contrast agent. It can also provide powerful complementary information for patients who undergo TMRI, especially at ventriculoarterial connection site. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In traditional magnetic resonance imaging procedure for congenital heart disease (CHD), two dimensional (2D) steady state free precession (SSFP) is mostly used to evaluate intra cardiac anomalies and contrast enhanced magnetic resonance angiography (CEMRA) to evaluate extra cardiac anomalies [1]. Other techniques like cine SSFP, phase contrast (PC), spin echo (SE) sequences are
∗ Corresponding authors. Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 15920386056; fax: +86 020 87615805. E-mail addresses: [email protected] (D. Chang), [email protected] (X. Kong), [email protected] (X. Zhou), [email protected] (S. Li), [email protected] (H. Wang). 1 Present address: No. 1277 Jiefang Road, Wuhan, China. Tel.: +86 13707158198; fax: +86 027 85726919. 2 Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 18620005019; fax: +86 020 87615805. 3 Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 13570577479; fax: +86 020 87615805. 4 Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 15013219685; fax: +86 020 87615805. 0720-048X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2013.03.014
used to evaluate cardiac function, blood flow and to overcome artifacts [2,3], so as to give additional information when the malformation is not certain or functional evaluation is needed. The disadvantages of TMRI are as follows: first, the multiple sequences with corresponding complex MRI techniques and image acquisition position are too complicated for most technicians to handle; second, several times of long breath-hold, which is needed during procedure, are difficult for certain patients, especially those with cardiopulmonary insufficiency, which is not uncommon in patients with cardiovascular diseases; furthermore, concerns about the safety of gadolinium-based contrast agents, especially those with reduced renal function [4,5], sometimes inhibit the usage of CEMRA for cardiovascular anomalies. Unenhanced three-dimensional (3D) SSFP (namely “true FISP” or “FIESTA”), a novel technique with respiratory-gated, isotropic, three dimensional data acquisition, provides a potential prospect to solve these problems. This technique is almost as simple as ordinary sequences for technicians to operate, and patients do not need to hold breath during the examination procedure. Besides, contrast agent will not be injected neither, which will exclude potential harm to the body. This technique, which has been validated for thoracic and abdominal
1744
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
large- and medium-sized vessels [6–9], shows great promise in assessment of congenital heart disease. In light of the fact that Fenchel et al. demonstrated its several limitations for infants [10], our study enrolled a group of juvenile and adult patients, for whom whole chest 3D SSFP and TMRI (mainly included 2D SSFP and CEMRA) were performed and compared, in order to assess the potential benefits of 3D SSFP sequence for congenital heart disease. 2. Materials and methods 2.1. Study population From August 2011 to April 2012, 20 consecutive patients (range, 19–36 years old, mean, 27.5 ± 8.5 years, male: female ratio, 14:6) with preoperative (n = 16) or postoperative (n = 4) complicated congenital heart defects referred to MRI for further evaluation about cardiovascular anomalies after transthoracic echocardiography. Exclusion criteria: arrhythmia, implantation of a pacemaker or a steel stent, claustrophobia, early pregnancy and other contraindications to MR imaging. Additional related diagnostic data including transthoracic echocardiogram, cardiac catheterization, or surgical confirmation were available for confirmation of anomalies. The study was in accordance with the ethical standards of the World Medical Association and ethics committee approval was obtained; informed consent was obtained from patients, parents, or caretakers. 2.2. Examination protocol A 32-channel 1.5-T whole-body MR scanner [Avanto; Siemens Medical Systems, maximum gradient 45 mT/m amplitude; slew rate, 200 (mT m−1 )/ms], with 17-element matrix coil (8 anterior and 9 posterior) activated for data collection, three-electrode electrocardiographic (ECG) leads applied for ECG gating, was used for image acquisition. Before the examination, patients were asked to take a 10–20 min rest to get breath and heartbeat in restful state, then lay supine on the imaging table, and were entered headfirst into the magnet. Free breathing 3D SSFP was performed first, then breath-hold sequences were followed, during which 2D SSFP was performed for all patients, and cine SSFP, PC, SE sequences were used for additional information when needed. Non-ECGtriggered contrast enhanced 3D fast low angle shot imaging (FLASH) sequence was performed at last. Retrospective trigger were induced for all ECG gated aforementioned sequences. Parallel acquisition technique was applied for all sequences above, and the acceleration factor was 2.
steadily, while deep breath or cough should be restrained. Parameters were as follows: TR 4ms, TE 1.6ms, flip angle 90◦ , matrix 173 × 256, slice thickness 0.9–1.25 mm, slices (112–192), segment 16–81, bandwidth 592 Hz/pixel. 3. TMRI For TMRI, 2D SSFP in cardiac short axis, vertical and horizontal long axis were performed to demonstrate intra cardiac morphology. Cine SSFP, PC, SE sequences of particular orientation was performed to get additional information when needed. CEMRA was performed for extra cardiac malformations with the same 3D volume localization as 3D SSFP. Contrast medium gadopentetate dimeglumine (Magnevist, Bayer Schering pharma AG, Berlin) 0.2–0.3 mmol/kg was injected into an antecubital vein using a MRcompatible High-voltage electric syringe (spectris solaris, medrad, Indianola, PA, U.S.A.) at a flow rate ranging from 1.5 ml/s to 3.0 ml/s, flushed by physiological saline 15–20 ml with the same flow rate. CARE Bolus technique was used to inspect peak concentration of contrast medium at the ascending aorta to switch to two or three sequential CEMRA acquisitions, which with 15-s interval, each lasted 16–20 s. Parameters of the sequences mentioned above were as follows: 2D SSFP: TR 2.9 ms,TE 1.22 ms, slice thickness 6–8 mm, gap 10–25%, slices 12–21, FOV 320–400 mm2 , matrix 136 × 256, bandwidth 889 Hz/pixel, flip angle 80◦ , average 1. Cine SSFP: TR 2.8ms, TE1.12ms, slice thickness 6∼8 mm, FOV 320–400 mm2 , matrix 156 × 192, bandwidth 930, flip angle 80◦ , average 1, temporal resolution 34.7–40.2 ms. 3D FLASH: TR 2.52 ms, TE 1.03 ms, slice thickness 0.8–1.1 mm, slices 112–144, FOV 320–400 mm2 , matrix 246 × 384, bandwidth 690 Hz/pixel, flip angle 15◦ , average 1. 3.1. Image analysis 3.1.1. Scoring and comparison of image qualities for 3D SSFP and TMRI A five-level scoring system was induced to evaluate the image quality of 3D SSFP, 2D SSFP and CEMRA. Scoring Principles were as follows [11]: (1) poor-quality information: nondiagnostic; (2) structures visible but markedly blurred: diagnosis suspected but not established; (3) anatomy visible with moderate blurring: able to establish diagnosis; (4) minimal blurring: good quality diagnostic information with definite diagnosis; and (5) sharply defined borders: excellent quality diagnostic information. For scoring of intra cardiac image quality, 3D SSFP was compared with 2D SSFP, which is mostly used for intra cardiac morphology evaluation in TMRI. For extra cardiac image quality, 3D SSFP was compared with CEMRA, which is ordinarily used for extra cardiac morphology illustration in TMRI.
2.3. Whole chest 3D SSFP Localizing images of coronal, axial and sagittal planes were obtained to prescribe image position. The volume of 3D-SSFP was localized from the top of subclavian artery to diaphragm, from cardiac apex to posterior internal chest wall, and sidewards to lateral internal chest wall. Multi phase images of a coronal plane crossing the root of aorta during the whole respiratory cycle were obtained to determine the top of the right hemi-diaphragm at the end of expiration. Multi phase images of cardiac four-chamber view transversing proximal right coronary artery were obtained to determine the stationary phase of cardiac motion. Image acquisition was triggered at the conjunction of stationary phase of right hemi-diaphragm at the end of expiration and cardiac motion during either end diastole or end systole phase. Based on individual anatomic status, field of view (FOV), slices, encoding orientation and oversampling were adjusted to avoid wrap-around artifacts. During the image acquisition, patients were instructed to breathe
3.1.2. Diagnosis of 3D SSFP Image data sets of 3D SSFP, 2D SSFP and CEMRA were processed and analyzed using volume rendering (VR), multi planar reformation (MPR), and maximum intensity projection (MIP) techniques. Additional information from other MRI sequences (cine SSFP, PC, SE), other image modalities (DSA, CTA, Echocardiography) and surgery was obtained to confirm anomalies. Sensitivity, specificity and accuracy of 3D SSFP diagnosis for intra cardiac and extra cardiac anomalies were calculated. 3.1.3. Consistency of 3D SSFP and CEMRA for diameter measurements of thoracic great arteries Aorta was divided into ascending aorta, aortic arch, descending aorta; pulmonary artery was divided into pulmonary trunk, left pulmonary artery and right pulmonary artery. The measurement sites are displayed in Fig. 1. Perpendicular dimension of each segment was evaluated at MIP reconstruction from the source
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
1745
Table 1 Confirmed intra cardiac anomalies (n = 35) diagnosed by 2D SSFP and 3D SSFP.
Fig. 1. Measurement sites of ascending aorta, aortic arch, descending aorta in a, pulmonary trunk, left pulmonary artery and right pulmonary artery in b: a1, the level of pulmonary bifurcation; a2, between the left common carotid and left subclavian artery; a3, the level of diaphragm; b1, halfway between the offspring and pulmonary bifurcation; b2 and b3, one centimeter before offspring of lobe branches.
images of 3D SSFP and CEMRA without the knowledge of the corresponding measurements through other techniques. Consistency of measurements obtained from the two techniques was assessed using Bland–Altman analysis [12]. 3.1.4. Statistic analysis Continuous variables are reported as mean ± standard deviation (SD). The agreement of image quality score between 2D SSFP and 3D SSFP for intra cardiac morphology and between 3D SSFP and CEMRA for extra cardiac morphology was assessed by two tailed Wilcoxon signed rank test (using software SPSS,version17.0; SPSS Inc, Chicago, Ill, USA, 2008), for which P < 0.05 was considered statistically significant. Perpendicular dimension measurements obtained from 3D SSFP and CEMRA in aortic and pulmonic segments were assessed by Bland–Altman analysis (using software MedCalc, version 8.0; MedCalc Software, Mariakerke, Belgium, 2005). 4. Results All patients including preoperative (n = 16) and postoperative (n = 4) complicated CHD (referred to the classification of CHD from AHA in 2008 [13]) underwent examination of 3D SSFP (acquisition time, 5 48 –27 2 , mean, 12 13 ) and 2D SSFP, and 16 of them underwent CEMRA further. The image acquisition was technically successful without any adverse events for all patients. Echocardiography (n = 20), cine SSFP MRI (n = 20), surgery (n = 16), DSA (n = 3), CTA (n = 1) were achieved to confirm uncertain diagnosis. 4.1. Image quality analysis Fig. 2 generally shows the image quality scoring distribution assessed from 3D SSFP for intra cardiac (4.30 ± 0.86), 3D SSFP for extra cardiac (3.90 ± 0.70), 2D SSFP for intra cardiac (3.85 ± 0.37), and CEMRA for extra cardiac (3.88 ± 0.34) morphology. All image
Fig. 2. The distribution of image quality scoring assessed from 3D SSFP and TMRI.
Anomalies
2D SSFP (n = 35)
3D SSFP (n = 35)
Single atrium Atrial isomerism right Common atrialventricular valve Atrial septal defect Right atrial dilation Ebstein’s anomaly Single ventricle Ventricular septal defect Right ventricular dilation
1 1 1 1 11 1 2 9 8
1 1 1 1 11 1 2 9 8
data sets of each patient for either intra or extra cardiac morphology scored ≥3, which allowed an establishment of diagnosis for all cases. The difference between 3D SSFP and 2D SSFP for intra cardiac morphology was statistically significant (P = 0.039), which suggested the former is better than the latter. However, the difference between 3D SSFP and CEMRA for extra cardiac morphology did not reach a statistical significance (P = 0.705). 4.2. The Diagnosis of 3D SSFP All of the confirmed malformations are showed in Table 1 (intra cardiac malformations) and Table 2 (extra cardiac malformations). The diagnostic sensitivity, specificity, and accuracy of SSFP for intra cardiac anomalies were all 100%, whereas for extra cardiac anomalies they were 93.8%, 93.8%, 100%, respectively. 4.3. Agreement of 3D SSFP and CEMRA for aortic and pulmonary diameters measurements Among the three patients who did not undergo CEMRA, one patient’s pulmonary trunk was not shown distinctly because of extreme stenosis and one patient showed an absence of pulmonary trunk and left pulmonary. There were finally a total of 51 aortic Table 2 Confirmed extra cardiac anomalies (n = 81) diagnosed by CEMRA and 3D SSFP. Anomalies
CEMRA (n = 81)
3D SSFP (n = 76)
Pulmonary artery stenosis Pulmonary artery dilation Anomalous origin of main pulmonary artery Overriding pulmonary trunk Anomalous origin of right pulmonary artery Left pulmonary artery absence Overriding aorta Ascending aorta dilation Right aortic arch Hypoplastic aorta Descending aorta coarctation Transposition of the great arteries Persistent truncus arteriosus Patent ductus arteriosus Aortopulmonary collateral artery Malposition of three major branch of aortic arch Anomalous origin of right common carotid artery First left intercostal artery dilation Inferior vena cava dilation Superior vena cava dilation Persistent left superior vena cava Double liver veins Left brachiocephalic vein aberration Twisted left subclavian artery Right coronary–right atrium fistula
16 14 1
16 14 1
1 1
1 1
1 9 6 5 2 1 3 1 3 2 1
1 9 6 5 2 1 3 1 2 0 1
1
1
1 4 1 1 1 2
1 4 1 1 1 2
2 1
0 1
1746
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
Fig. 3. Bland–Altman analysis shows good consistency of 3D SSFP and CEMRA for diameter measurements of aorta (a) and pulmonary artery (b).
segments and 48 pulmonic segments of individual measurements for the two techniques obtained. Referred to Bland–Altman analysis, the mean difference between 3D SSFP and CEMRA for aortic diameter measurements (3D SSFP minus CEMRA) was 0.5 ± 1.2 mm (Fig. 3a), and 95% confidence interval was (−1.9 mm, 2.9 mm). About 2.0% (1/51) of points lay outside the confidence scale. Within the confidence scale, the maximum measurement difference of the two image techniques was 1.9 mm, which is acceptable in clinical application. Hence, there was good agreement between 3D SSFP and CEMRA, and 3D SSFP can be an alternative for aortic diameter measurement. The mean difference between 3D SSFP and CEMRA for pulmonic diameter measurements (3D SSFP minus CEMRA) was 0.0 ± 1.7 mm (Fig. 3b), and 95% confidence interval was (−3.4 mm, 3.4 mm). About 4.2% (2/48) of points lay outside the confidence scale. Within the confidence scale, the maximum measurement difference of the two image techniques was 1.8 mm, which is acceptable in clinical application. Hence, similarly, there was good agreement between 3D SSFP and CEMRA, and 3D SSFP can be an alternative for pulmonary artery diameter measurements. 5. Discussion Unenhanced MRA techniques include Time of Flight (TOF), phase contrast (PC) and 3D SSFP. The former two techniques yield unsatisfactory image data sets for the thoracic complex cardiovascular geometry, because of the long acquisition time and heart motion artifact [14]. The value of 3D SSFP MRA used for aorta [15], pulmonary arteries [7,16], thoracic veins [17] and renal arteries [18] has been validated by previous studies. The results of our study suggest that, the potential benefits of 3D SSFP for CHD are as follows: The image procedure of 3D SSFP is more simplified than TMRI. The 3D volume localization of 3D SSFP only requires the full involvement of the targeted cardiac chamber or vessel, which is much easier than the localization procedure of TMRI. And to make sure that the breathing navigation and ECG gating function correctly, the navigator pulse should be put on top of the right hemi-diaphragm at end expiration, and ECG trigger be localized at the stationary phase of the cardiac motion. The whole image data sets will be obtained after one single image acquisition. Unlike TMRI, the diagnostic value of obtained images does not lie too much on the experience of the technician, and the source images can be processed and viewed at any arbitrary plane offline attributed to its 3D nature. Contrast agent is not needed, which is extremely well-suitable for patients with renal failure, and those who are allergic to Iodine or Gadolinium agent. Furthermore, patient only need to stay calm, and keep in a restful state to maintain a relatively stable breath and
heartbeat during the image acquisition, which is much easier for them to accomplish when they have cardiopulmonary insufficiency or difficulties holding breaths. The source images of 3D SSFP are of high value for diagnosis. Because navigator and ECG trigger techniques were used, the image acquisition of 3D SSFP minimizes both respiratory and cardiac motion artifacts, and thus enables superior image quality for both intra and extra cardiac morphology simultaneously, which cannot be accomplished by any traditional MRI sequence. Processing the source image of 3D SSFP with multiple viewing and reconstruction techniques as VR, MPR, MIP, the anatomic relationship between cardiac and thoracic vessels can be demonstrated at any arbitrary plane offline, which is essential for showing the cardiac morphology (Fig. 4a and b), the spatial relationship between cardiac and large thoracic vessels (Fig. 4c–h) and the offspring site, course, stenosis or dilation of the thoracic vessels. These enabled its high value for demonstration and diagnosis for complex congenital cardiovascular anomalies. There is good agreement between 3D SSFP and CEMRA for diameter measurements for thoracic great arteries. CEMRA is traditionally an essential technique for extra cardiac anomalies, and it stands for the significant advantage of TMRI for extra cardiac anomalies for CHD in competition to echocardiography. The good agreement of diameter measurements for thoracic great arteries further approved the value of 3D SSFP for quantitative measurements, which may sometimes be helpful for clinician. Actually, 3D SSFP images can demonstrate vascular wall, hence can evaluate lumen diameter with greater accuracy than CEMRA, and even to detect potential vascular wall abnormality (for example, lateral thrombus). Furthermore, malformations at ventriculoarterial connection site, such as right ventricular out flow tract stenosis (Fig. 4c and d), aortic overriding (Fig. 4e and f), transposition of great arteries, etc., in 3D SSFP was of higher confidence to find the anomalies than in CEMRA, which often contains cardiac motion artifacts (Fig. 4d and f) at and surrounding cardiac because of disuse of ECG gating technique. However, a suspected limitation for 3D SSFP to visualize small or twisted vessels needs further research to be confirmed. In our study, it missed small vessels as two aortopulmonary collateral arteries (one from right subclavian artery, one from aorta) and one patent ductus arteriosus, with diameters of 2.1–6.3 mm. And there were two subclavian arteries that were not shown distinctly while one with twisted proximal segment and the other with malformation at proximal and middle segments. Meanwhile, the 3D SSFP image acquisition is closely interrelated with patients’ vital signs; which limits its use for some patients with too fast or unstable heartbeats or breaths, for example in children or patients with cardiomyopathy. In our study, all patients got image quality sufficient for clinical diagnosis. This may be attributed to the stable vital signs
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
1747
Fig. 4. Four multiplanar reformatted planes of 3D SSFP (a, c, e, g) compared with similar planes of TMRI [one 2D SSFP image (b), three multiplanar reformatted planes of CEMRA (d, f, h)] showing intra and extra cardiac anomalies. (a–d) Patient with atrial isomerism, for intra cardiac malformations, both 3D SSFP (a) and 2D SSFP (b) visualized the bilateral right atrium structure (asterisk), single ventricle (pound) and atrial septal defect (arrow); for extra cardiac malformations, both 3D SSFP (c) and CEMRA (d) depicted double liver veins (arrow), with the left one draining into the inferior vena cava (arrow head). (e–h) Patient with F4, 3D SSFP indentified the stenosis of right ventricular outflow tract (Arrow in e), overriding aorta and ventricular septal defect (arrow in g), while there were cardiac motion artifacts at and surrounding cardiac at CEMRA (f, h) images.
with no significant breath, ECG abnormality in relatively elder CHD patients. 6. Conclusions In conclusion, our study suggested that, 3D SSFP MRI can be an alternative to TMRI for relatively older patients with CHD, especially those who have renal insufficiency, breath-hold difficulty or who are allergic to contrast agent. It can also provide powerful information that complements TMRI for patients with congenital heart disease, especially at ventriculoarterial connection site.
References [1] Prasad SK, Soukias N, Hornung T, et al. Role of magnetic resonance angiography in the diagnosis of major aortopulmonary collateral arteries and partial anomalous pulmonary venous drainage. Circulation 2004;109(2):207–14. [2] Puranik R, Muthurangu V, Celermajer DS, et al. Congenital heart disease and multi-modality imaging. Heart, Lung and Circulation 2010;19(3):133–44. [3] Hundley WG, Bluemke DA, Finn JP, et al. ACCF/ACR/AHA/NASCI/SCMR 2010 Expert Consensus Document on Cardiovascular Magnetic Resonance: a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents. Circulation 2010;121(22):2462–508. [4] Ledneva E, Karie S, Launay-Vacher V, et al. Renal safety of gadoliniumbased contrast media in patients with chronic renal insufficiency. Radiology 2009;250(3):618–28.
1748
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
[5] Shellock FG, Kanal E. Safety of magnetic resonance imaging contrast agents. Journal of Magnetic Resonance Imaging 1999;10(3):477–84. [6] Francois CJ, Tuite D, Deshpande V, et al. Pulmonary vein imaging with unenhanced three-dimensional balanced steady-state free precession MR angiography: initial clinical evaluation. Radiology 2009;250(3):932–9. [7] Krishnam MS, Tomasian A, Deshpande V, et al. Noncontrast 3D steady-state free-precession magnetic resonance angiography of the whole chest using nonselective radiofrequency excitation over a large field of view: comparison with single-phase 3D contrast-enhanced magnetic resonance angiography. Investigative Radiology 2008;43(6):411–20. [8] Potthast S, Mitsumori L, Stanescu LA, et al. Measuring aortic diameter with different MR techniques: comparison of three-dimensional (3D) navigated steady-state free-precession (SSFP), 3D contrast-enhanced magnetic resonance angiography (CE-MRA), 2D T2 black blood, and 2D cine SSFP. Journal of Magnetic Resonance Imaging 2010;31(1):177–84. [9] Krishnam MS, Tomasian A, Malik S, et al. Three-dimensional imaging of pulmonary veins by a novel steady-state free-precession magnetic resonance angiography technique without the use of intravenous contrast agent: initial experience. Investigative Radiology 2009;44(8):447–53. [10] Fenchel M, Greil GF, Martirosian P, et al. Three-dimensional morphological magnetic resonance imaging in infants and children with congenital heart disease. Pediatric Radiology 2006;36(12):1265–72. [11] Sorensen TS, Korperich H, Greil GF, et al. Operator-independent isotropic threedimensional magnetic resonance imaging for morphology in congenital heart disease: a validation study. Circulation 2004;110(2):163–9.
[12] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1(8476): 307–10. [13] Warnes CA, Williams RG, Bashore TM, et al. ACC/AHA 2008 Guidelines for the Management of Adults with Congenital Heart Disease: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to develop guidelines for the management of adults with congenital heart disease). Circulation 2008;118(23):2395–451. [14] Miyazaki M, Lee VS, Nonenhanced MR angiography. Radiology 2008;248(1):20–43. [15] Francois CJ, Tuite D, Deshpande V, et al. Unenhanced MR angiography of the thoracic aorta: initial clinical evaluation. American Journal of Roentgenology 2008;190(4):902–6. [16] Hui BK, Noga ML, Gan KD, et al. Navigator-gated three-dimensional MR angiography of the pulmonary arteries using steady-state free precession. Journal of Magnetic Resonance Imaging 2005;21(6):831–5. [17] Tomasian A, Lohan DG, Laub G, et al. Noncontrast 3D steady state free precession magnetic resonance angiography of the thoracic central veins using nonselective radiofrequency excitation over a large field of view: initial experience. Investigative Radiology 2008;43(5):306–13. [18] Herborn CU, Watkins DM, Runge VM, et al. Renal arteries: comparison of steady-state free precession MR angiography and contrast-enhanced MR angiography. Radiology 2006;239(1):263–8.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Unenhanced steady state free precession versus traditional MR imaging for congenital heart disease Dandan Chang a,∗ , Xiangquan Kong b,∗,1 , Xuhui Zhou a,2 , Shurong Li a,3 , Huanjun Wang a,4 a b
Department of Diagnostic Radiology, the First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, Guangdong, China Department of Radiology, the Affiliated Union Hospital, Huazhong University of Science and Technology, Wuhan, Hubei, China
a r t i c l e
i n f o
Article history: Received 23 July 2012 Received in revised form 1 February 2013 Accepted 25 March 2013 Keywords: Congenital heart disease Three dimensional steady state free precession Magnetic resonance angiography Image quality Diagnosis Diameter measurement
a b s t r a c t Purpose: To assess potential benefits of three dimensional (3D) steady state free precession (SSFP) magnetic resonance sequence for congenital heart disease (CHD). Materials and methods: Twenty consecutive patients with CHD (male:female ratio,14:6, mean age, 27.5 ± 8.5 years) underwent both 3D SSFP and traditional MR imaging (TMRI) [including two dimensional (2D) SSFP and contrast enhanced magnetic resonance angiography (CEMRA)]. Image quality and diagnosis were compared, and Bland–Altman analysis was used to evaluate consistency of 3D SSFP and CEMRA for diameter measurements. Results: A total of 35 intra and 81 extra cardiac anomalies were identified in all patients. The image quality of 3D SSFP and TMRI for either intra or extra cardiac anomalies of all patients scored ≥3, which allowed an establishment of diagnosis for all cases. The diagnostic sensitivity, specificity, and accuracy of 3D SSFP for the detection of intra cardiac anomalies were all 100%, whereas for extra cardiac anomalies they were 93.8%, 93.8%, 100%, respectively. Mean differences (3D SSFP minus CEMRA) for aorta and pulmonary arteries were 0.5 ± 1.2 mm and 0.0 ± 1.7 mm, respectively, showing good consistency of 3D SSFP and CEMRA for diameter measurements. Conclusion: 3D SSFP MRI can be an alternative image modality to TMRI for patients with congenital heart disease, especially for those who have renal insufficiency, breath-hold difficulty or who are allergic to contrast agent. It can also provide powerful complementary information for patients who undergo TMRI, especially at ventriculoarterial connection site. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In traditional magnetic resonance imaging procedure for congenital heart disease (CHD), two dimensional (2D) steady state free precession (SSFP) is mostly used to evaluate intra cardiac anomalies and contrast enhanced magnetic resonance angiography (CEMRA) to evaluate extra cardiac anomalies [1]. Other techniques like cine SSFP, phase contrast (PC), spin echo (SE) sequences are
∗ Corresponding authors. Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 15920386056; fax: +86 020 87615805. E-mail addresses: [email protected] (D. Chang), [email protected] (X. Kong), [email protected] (X. Zhou), [email protected] (S. Li), [email protected] (H. Wang). 1 Present address: No. 1277 Jiefang Road, Wuhan, China. Tel.: +86 13707158198; fax: +86 027 85726919. 2 Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 18620005019; fax: +86 020 87615805. 3 Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 13570577479; fax: +86 020 87615805. 4 Present address: No. 58 Zhongshan Road 2nd, Guangzhou, China. Tel.: +86 15013219685; fax: +86 020 87615805. 0720-048X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2013.03.014
used to evaluate cardiac function, blood flow and to overcome artifacts [2,3], so as to give additional information when the malformation is not certain or functional evaluation is needed. The disadvantages of TMRI are as follows: first, the multiple sequences with corresponding complex MRI techniques and image acquisition position are too complicated for most technicians to handle; second, several times of long breath-hold, which is needed during procedure, are difficult for certain patients, especially those with cardiopulmonary insufficiency, which is not uncommon in patients with cardiovascular diseases; furthermore, concerns about the safety of gadolinium-based contrast agents, especially those with reduced renal function [4,5], sometimes inhibit the usage of CEMRA for cardiovascular anomalies. Unenhanced three-dimensional (3D) SSFP (namely “true FISP” or “FIESTA”), a novel technique with respiratory-gated, isotropic, three dimensional data acquisition, provides a potential prospect to solve these problems. This technique is almost as simple as ordinary sequences for technicians to operate, and patients do not need to hold breath during the examination procedure. Besides, contrast agent will not be injected neither, which will exclude potential harm to the body. This technique, which has been validated for thoracic and abdominal
1744
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
large- and medium-sized vessels [6–9], shows great promise in assessment of congenital heart disease. In light of the fact that Fenchel et al. demonstrated its several limitations for infants [10], our study enrolled a group of juvenile and adult patients, for whom whole chest 3D SSFP and TMRI (mainly included 2D SSFP and CEMRA) were performed and compared, in order to assess the potential benefits of 3D SSFP sequence for congenital heart disease. 2. Materials and methods 2.1. Study population From August 2011 to April 2012, 20 consecutive patients (range, 19–36 years old, mean, 27.5 ± 8.5 years, male: female ratio, 14:6) with preoperative (n = 16) or postoperative (n = 4) complicated congenital heart defects referred to MRI for further evaluation about cardiovascular anomalies after transthoracic echocardiography. Exclusion criteria: arrhythmia, implantation of a pacemaker or a steel stent, claustrophobia, early pregnancy and other contraindications to MR imaging. Additional related diagnostic data including transthoracic echocardiogram, cardiac catheterization, or surgical confirmation were available for confirmation of anomalies. The study was in accordance with the ethical standards of the World Medical Association and ethics committee approval was obtained; informed consent was obtained from patients, parents, or caretakers. 2.2. Examination protocol A 32-channel 1.5-T whole-body MR scanner [Avanto; Siemens Medical Systems, maximum gradient 45 mT/m amplitude; slew rate, 200 (mT m−1 )/ms], with 17-element matrix coil (8 anterior and 9 posterior) activated for data collection, three-electrode electrocardiographic (ECG) leads applied for ECG gating, was used for image acquisition. Before the examination, patients were asked to take a 10–20 min rest to get breath and heartbeat in restful state, then lay supine on the imaging table, and were entered headfirst into the magnet. Free breathing 3D SSFP was performed first, then breath-hold sequences were followed, during which 2D SSFP was performed for all patients, and cine SSFP, PC, SE sequences were used for additional information when needed. Non-ECGtriggered contrast enhanced 3D fast low angle shot imaging (FLASH) sequence was performed at last. Retrospective trigger were induced for all ECG gated aforementioned sequences. Parallel acquisition technique was applied for all sequences above, and the acceleration factor was 2.
steadily, while deep breath or cough should be restrained. Parameters were as follows: TR 4ms, TE 1.6ms, flip angle 90◦ , matrix 173 × 256, slice thickness 0.9–1.25 mm, slices (112–192), segment 16–81, bandwidth 592 Hz/pixel. 3. TMRI For TMRI, 2D SSFP in cardiac short axis, vertical and horizontal long axis were performed to demonstrate intra cardiac morphology. Cine SSFP, PC, SE sequences of particular orientation was performed to get additional information when needed. CEMRA was performed for extra cardiac malformations with the same 3D volume localization as 3D SSFP. Contrast medium gadopentetate dimeglumine (Magnevist, Bayer Schering pharma AG, Berlin) 0.2–0.3 mmol/kg was injected into an antecubital vein using a MRcompatible High-voltage electric syringe (spectris solaris, medrad, Indianola, PA, U.S.A.) at a flow rate ranging from 1.5 ml/s to 3.0 ml/s, flushed by physiological saline 15–20 ml with the same flow rate. CARE Bolus technique was used to inspect peak concentration of contrast medium at the ascending aorta to switch to two or three sequential CEMRA acquisitions, which with 15-s interval, each lasted 16–20 s. Parameters of the sequences mentioned above were as follows: 2D SSFP: TR 2.9 ms,TE 1.22 ms, slice thickness 6–8 mm, gap 10–25%, slices 12–21, FOV 320–400 mm2 , matrix 136 × 256, bandwidth 889 Hz/pixel, flip angle 80◦ , average 1. Cine SSFP: TR 2.8ms, TE1.12ms, slice thickness 6∼8 mm, FOV 320–400 mm2 , matrix 156 × 192, bandwidth 930, flip angle 80◦ , average 1, temporal resolution 34.7–40.2 ms. 3D FLASH: TR 2.52 ms, TE 1.03 ms, slice thickness 0.8–1.1 mm, slices 112–144, FOV 320–400 mm2 , matrix 246 × 384, bandwidth 690 Hz/pixel, flip angle 15◦ , average 1. 3.1. Image analysis 3.1.1. Scoring and comparison of image qualities for 3D SSFP and TMRI A five-level scoring system was induced to evaluate the image quality of 3D SSFP, 2D SSFP and CEMRA. Scoring Principles were as follows [11]: (1) poor-quality information: nondiagnostic; (2) structures visible but markedly blurred: diagnosis suspected but not established; (3) anatomy visible with moderate blurring: able to establish diagnosis; (4) minimal blurring: good quality diagnostic information with definite diagnosis; and (5) sharply defined borders: excellent quality diagnostic information. For scoring of intra cardiac image quality, 3D SSFP was compared with 2D SSFP, which is mostly used for intra cardiac morphology evaluation in TMRI. For extra cardiac image quality, 3D SSFP was compared with CEMRA, which is ordinarily used for extra cardiac morphology illustration in TMRI.
2.3. Whole chest 3D SSFP Localizing images of coronal, axial and sagittal planes were obtained to prescribe image position. The volume of 3D-SSFP was localized from the top of subclavian artery to diaphragm, from cardiac apex to posterior internal chest wall, and sidewards to lateral internal chest wall. Multi phase images of a coronal plane crossing the root of aorta during the whole respiratory cycle were obtained to determine the top of the right hemi-diaphragm at the end of expiration. Multi phase images of cardiac four-chamber view transversing proximal right coronary artery were obtained to determine the stationary phase of cardiac motion. Image acquisition was triggered at the conjunction of stationary phase of right hemi-diaphragm at the end of expiration and cardiac motion during either end diastole or end systole phase. Based on individual anatomic status, field of view (FOV), slices, encoding orientation and oversampling were adjusted to avoid wrap-around artifacts. During the image acquisition, patients were instructed to breathe
3.1.2. Diagnosis of 3D SSFP Image data sets of 3D SSFP, 2D SSFP and CEMRA were processed and analyzed using volume rendering (VR), multi planar reformation (MPR), and maximum intensity projection (MIP) techniques. Additional information from other MRI sequences (cine SSFP, PC, SE), other image modalities (DSA, CTA, Echocardiography) and surgery was obtained to confirm anomalies. Sensitivity, specificity and accuracy of 3D SSFP diagnosis for intra cardiac and extra cardiac anomalies were calculated. 3.1.3. Consistency of 3D SSFP and CEMRA for diameter measurements of thoracic great arteries Aorta was divided into ascending aorta, aortic arch, descending aorta; pulmonary artery was divided into pulmonary trunk, left pulmonary artery and right pulmonary artery. The measurement sites are displayed in Fig. 1. Perpendicular dimension of each segment was evaluated at MIP reconstruction from the source
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
1745
Table 1 Confirmed intra cardiac anomalies (n = 35) diagnosed by 2D SSFP and 3D SSFP.
Fig. 1. Measurement sites of ascending aorta, aortic arch, descending aorta in a, pulmonary trunk, left pulmonary artery and right pulmonary artery in b: a1, the level of pulmonary bifurcation; a2, between the left common carotid and left subclavian artery; a3, the level of diaphragm; b1, halfway between the offspring and pulmonary bifurcation; b2 and b3, one centimeter before offspring of lobe branches.
images of 3D SSFP and CEMRA without the knowledge of the corresponding measurements through other techniques. Consistency of measurements obtained from the two techniques was assessed using Bland–Altman analysis [12]. 3.1.4. Statistic analysis Continuous variables are reported as mean ± standard deviation (SD). The agreement of image quality score between 2D SSFP and 3D SSFP for intra cardiac morphology and between 3D SSFP and CEMRA for extra cardiac morphology was assessed by two tailed Wilcoxon signed rank test (using software SPSS,version17.0; SPSS Inc, Chicago, Ill, USA, 2008), for which P < 0.05 was considered statistically significant. Perpendicular dimension measurements obtained from 3D SSFP and CEMRA in aortic and pulmonic segments were assessed by Bland–Altman analysis (using software MedCalc, version 8.0; MedCalc Software, Mariakerke, Belgium, 2005). 4. Results All patients including preoperative (n = 16) and postoperative (n = 4) complicated CHD (referred to the classification of CHD from AHA in 2008 [13]) underwent examination of 3D SSFP (acquisition time, 5 48 –27 2 , mean, 12 13 ) and 2D SSFP, and 16 of them underwent CEMRA further. The image acquisition was technically successful without any adverse events for all patients. Echocardiography (n = 20), cine SSFP MRI (n = 20), surgery (n = 16), DSA (n = 3), CTA (n = 1) were achieved to confirm uncertain diagnosis. 4.1. Image quality analysis Fig. 2 generally shows the image quality scoring distribution assessed from 3D SSFP for intra cardiac (4.30 ± 0.86), 3D SSFP for extra cardiac (3.90 ± 0.70), 2D SSFP for intra cardiac (3.85 ± 0.37), and CEMRA for extra cardiac (3.88 ± 0.34) morphology. All image
Fig. 2. The distribution of image quality scoring assessed from 3D SSFP and TMRI.
Anomalies
2D SSFP (n = 35)
3D SSFP (n = 35)
Single atrium Atrial isomerism right Common atrialventricular valve Atrial septal defect Right atrial dilation Ebstein’s anomaly Single ventricle Ventricular septal defect Right ventricular dilation
1 1 1 1 11 1 2 9 8
1 1 1 1 11 1 2 9 8
data sets of each patient for either intra or extra cardiac morphology scored ≥3, which allowed an establishment of diagnosis for all cases. The difference between 3D SSFP and 2D SSFP for intra cardiac morphology was statistically significant (P = 0.039), which suggested the former is better than the latter. However, the difference between 3D SSFP and CEMRA for extra cardiac morphology did not reach a statistical significance (P = 0.705). 4.2. The Diagnosis of 3D SSFP All of the confirmed malformations are showed in Table 1 (intra cardiac malformations) and Table 2 (extra cardiac malformations). The diagnostic sensitivity, specificity, and accuracy of SSFP for intra cardiac anomalies were all 100%, whereas for extra cardiac anomalies they were 93.8%, 93.8%, 100%, respectively. 4.3. Agreement of 3D SSFP and CEMRA for aortic and pulmonary diameters measurements Among the three patients who did not undergo CEMRA, one patient’s pulmonary trunk was not shown distinctly because of extreme stenosis and one patient showed an absence of pulmonary trunk and left pulmonary. There were finally a total of 51 aortic Table 2 Confirmed extra cardiac anomalies (n = 81) diagnosed by CEMRA and 3D SSFP. Anomalies
CEMRA (n = 81)
3D SSFP (n = 76)
Pulmonary artery stenosis Pulmonary artery dilation Anomalous origin of main pulmonary artery Overriding pulmonary trunk Anomalous origin of right pulmonary artery Left pulmonary artery absence Overriding aorta Ascending aorta dilation Right aortic arch Hypoplastic aorta Descending aorta coarctation Transposition of the great arteries Persistent truncus arteriosus Patent ductus arteriosus Aortopulmonary collateral artery Malposition of three major branch of aortic arch Anomalous origin of right common carotid artery First left intercostal artery dilation Inferior vena cava dilation Superior vena cava dilation Persistent left superior vena cava Double liver veins Left brachiocephalic vein aberration Twisted left subclavian artery Right coronary–right atrium fistula
16 14 1
16 14 1
1 1
1 1
1 9 6 5 2 1 3 1 3 2 1
1 9 6 5 2 1 3 1 2 0 1
1
1
1 4 1 1 1 2
1 4 1 1 1 2
2 1
0 1
1746
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
Fig. 3. Bland–Altman analysis shows good consistency of 3D SSFP and CEMRA for diameter measurements of aorta (a) and pulmonary artery (b).
segments and 48 pulmonic segments of individual measurements for the two techniques obtained. Referred to Bland–Altman analysis, the mean difference between 3D SSFP and CEMRA for aortic diameter measurements (3D SSFP minus CEMRA) was 0.5 ± 1.2 mm (Fig. 3a), and 95% confidence interval was (−1.9 mm, 2.9 mm). About 2.0% (1/51) of points lay outside the confidence scale. Within the confidence scale, the maximum measurement difference of the two image techniques was 1.9 mm, which is acceptable in clinical application. Hence, there was good agreement between 3D SSFP and CEMRA, and 3D SSFP can be an alternative for aortic diameter measurement. The mean difference between 3D SSFP and CEMRA for pulmonic diameter measurements (3D SSFP minus CEMRA) was 0.0 ± 1.7 mm (Fig. 3b), and 95% confidence interval was (−3.4 mm, 3.4 mm). About 4.2% (2/48) of points lay outside the confidence scale. Within the confidence scale, the maximum measurement difference of the two image techniques was 1.8 mm, which is acceptable in clinical application. Hence, similarly, there was good agreement between 3D SSFP and CEMRA, and 3D SSFP can be an alternative for pulmonary artery diameter measurements. 5. Discussion Unenhanced MRA techniques include Time of Flight (TOF), phase contrast (PC) and 3D SSFP. The former two techniques yield unsatisfactory image data sets for the thoracic complex cardiovascular geometry, because of the long acquisition time and heart motion artifact [14]. The value of 3D SSFP MRA used for aorta [15], pulmonary arteries [7,16], thoracic veins [17] and renal arteries [18] has been validated by previous studies. The results of our study suggest that, the potential benefits of 3D SSFP for CHD are as follows: The image procedure of 3D SSFP is more simplified than TMRI. The 3D volume localization of 3D SSFP only requires the full involvement of the targeted cardiac chamber or vessel, which is much easier than the localization procedure of TMRI. And to make sure that the breathing navigation and ECG gating function correctly, the navigator pulse should be put on top of the right hemi-diaphragm at end expiration, and ECG trigger be localized at the stationary phase of the cardiac motion. The whole image data sets will be obtained after one single image acquisition. Unlike TMRI, the diagnostic value of obtained images does not lie too much on the experience of the technician, and the source images can be processed and viewed at any arbitrary plane offline attributed to its 3D nature. Contrast agent is not needed, which is extremely well-suitable for patients with renal failure, and those who are allergic to Iodine or Gadolinium agent. Furthermore, patient only need to stay calm, and keep in a restful state to maintain a relatively stable breath and
heartbeat during the image acquisition, which is much easier for them to accomplish when they have cardiopulmonary insufficiency or difficulties holding breaths. The source images of 3D SSFP are of high value for diagnosis. Because navigator and ECG trigger techniques were used, the image acquisition of 3D SSFP minimizes both respiratory and cardiac motion artifacts, and thus enables superior image quality for both intra and extra cardiac morphology simultaneously, which cannot be accomplished by any traditional MRI sequence. Processing the source image of 3D SSFP with multiple viewing and reconstruction techniques as VR, MPR, MIP, the anatomic relationship between cardiac and thoracic vessels can be demonstrated at any arbitrary plane offline, which is essential for showing the cardiac morphology (Fig. 4a and b), the spatial relationship between cardiac and large thoracic vessels (Fig. 4c–h) and the offspring site, course, stenosis or dilation of the thoracic vessels. These enabled its high value for demonstration and diagnosis for complex congenital cardiovascular anomalies. There is good agreement between 3D SSFP and CEMRA for diameter measurements for thoracic great arteries. CEMRA is traditionally an essential technique for extra cardiac anomalies, and it stands for the significant advantage of TMRI for extra cardiac anomalies for CHD in competition to echocardiography. The good agreement of diameter measurements for thoracic great arteries further approved the value of 3D SSFP for quantitative measurements, which may sometimes be helpful for clinician. Actually, 3D SSFP images can demonstrate vascular wall, hence can evaluate lumen diameter with greater accuracy than CEMRA, and even to detect potential vascular wall abnormality (for example, lateral thrombus). Furthermore, malformations at ventriculoarterial connection site, such as right ventricular out flow tract stenosis (Fig. 4c and d), aortic overriding (Fig. 4e and f), transposition of great arteries, etc., in 3D SSFP was of higher confidence to find the anomalies than in CEMRA, which often contains cardiac motion artifacts (Fig. 4d and f) at and surrounding cardiac because of disuse of ECG gating technique. However, a suspected limitation for 3D SSFP to visualize small or twisted vessels needs further research to be confirmed. In our study, it missed small vessels as two aortopulmonary collateral arteries (one from right subclavian artery, one from aorta) and one patent ductus arteriosus, with diameters of 2.1–6.3 mm. And there were two subclavian arteries that were not shown distinctly while one with twisted proximal segment and the other with malformation at proximal and middle segments. Meanwhile, the 3D SSFP image acquisition is closely interrelated with patients’ vital signs; which limits its use for some patients with too fast or unstable heartbeats or breaths, for example in children or patients with cardiomyopathy. In our study, all patients got image quality sufficient for clinical diagnosis. This may be attributed to the stable vital signs
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
1747
Fig. 4. Four multiplanar reformatted planes of 3D SSFP (a, c, e, g) compared with similar planes of TMRI [one 2D SSFP image (b), three multiplanar reformatted planes of CEMRA (d, f, h)] showing intra and extra cardiac anomalies. (a–d) Patient with atrial isomerism, for intra cardiac malformations, both 3D SSFP (a) and 2D SSFP (b) visualized the bilateral right atrium structure (asterisk), single ventricle (pound) and atrial septal defect (arrow); for extra cardiac malformations, both 3D SSFP (c) and CEMRA (d) depicted double liver veins (arrow), with the left one draining into the inferior vena cava (arrow head). (e–h) Patient with F4, 3D SSFP indentified the stenosis of right ventricular outflow tract (Arrow in e), overriding aorta and ventricular septal defect (arrow in g), while there were cardiac motion artifacts at and surrounding cardiac at CEMRA (f, h) images.
with no significant breath, ECG abnormality in relatively elder CHD patients. 6. Conclusions In conclusion, our study suggested that, 3D SSFP MRI can be an alternative to TMRI for relatively older patients with CHD, especially those who have renal insufficiency, breath-hold difficulty or who are allergic to contrast agent. It can also provide powerful information that complements TMRI for patients with congenital heart disease, especially at ventriculoarterial connection site.
References [1] Prasad SK, Soukias N, Hornung T, et al. Role of magnetic resonance angiography in the diagnosis of major aortopulmonary collateral arteries and partial anomalous pulmonary venous drainage. Circulation 2004;109(2):207–14. [2] Puranik R, Muthurangu V, Celermajer DS, et al. Congenital heart disease and multi-modality imaging. Heart, Lung and Circulation 2010;19(3):133–44. [3] Hundley WG, Bluemke DA, Finn JP, et al. ACCF/ACR/AHA/NASCI/SCMR 2010 Expert Consensus Document on Cardiovascular Magnetic Resonance: a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents. Circulation 2010;121(22):2462–508. [4] Ledneva E, Karie S, Launay-Vacher V, et al. Renal safety of gadoliniumbased contrast media in patients with chronic renal insufficiency. Radiology 2009;250(3):618–28.
1748
D. Chang et al. / European Journal of Radiology 82 (2013) 1743–1748
[5] Shellock FG, Kanal E. Safety of magnetic resonance imaging contrast agents. Journal of Magnetic Resonance Imaging 1999;10(3):477–84. [6] Francois CJ, Tuite D, Deshpande V, et al. Pulmonary vein imaging with unenhanced three-dimensional balanced steady-state free precession MR angiography: initial clinical evaluation. Radiology 2009;250(3):932–9. [7] Krishnam MS, Tomasian A, Deshpande V, et al. Noncontrast 3D steady-state free-precession magnetic resonance angiography of the whole chest using nonselective radiofrequency excitation over a large field of view: comparison with single-phase 3D contrast-enhanced magnetic resonance angiography. Investigative Radiology 2008;43(6):411–20. [8] Potthast S, Mitsumori L, Stanescu LA, et al. Measuring aortic diameter with different MR techniques: comparison of three-dimensional (3D) navigated steady-state free-precession (SSFP), 3D contrast-enhanced magnetic resonance angiography (CE-MRA), 2D T2 black blood, and 2D cine SSFP. Journal of Magnetic Resonance Imaging 2010;31(1):177–84. [9] Krishnam MS, Tomasian A, Malik S, et al. Three-dimensional imaging of pulmonary veins by a novel steady-state free-precession magnetic resonance angiography technique without the use of intravenous contrast agent: initial experience. Investigative Radiology 2009;44(8):447–53. [10] Fenchel M, Greil GF, Martirosian P, et al. Three-dimensional morphological magnetic resonance imaging in infants and children with congenital heart disease. Pediatric Radiology 2006;36(12):1265–72. [11] Sorensen TS, Korperich H, Greil GF, et al. Operator-independent isotropic threedimensional magnetic resonance imaging for morphology in congenital heart disease: a validation study. Circulation 2004;110(2):163–9.
[12] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1(8476): 307–10. [13] Warnes CA, Williams RG, Bashore TM, et al. ACC/AHA 2008 Guidelines for the Management of Adults with Congenital Heart Disease: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to develop guidelines for the management of adults with congenital heart disease). Circulation 2008;118(23):2395–451. [14] Miyazaki M, Lee VS, Nonenhanced MR angiography. Radiology 2008;248(1):20–43. [15] Francois CJ, Tuite D, Deshpande V, et al. Unenhanced MR angiography of the thoracic aorta: initial clinical evaluation. American Journal of Roentgenology 2008;190(4):902–6. [16] Hui BK, Noga ML, Gan KD, et al. Navigator-gated three-dimensional MR angiography of the pulmonary arteries using steady-state free precession. Journal of Magnetic Resonance Imaging 2005;21(6):831–5. [17] Tomasian A, Lohan DG, Laub G, et al. Noncontrast 3D steady state free precession magnetic resonance angiography of the thoracic central veins using nonselective radiofrequency excitation over a large field of view: initial experience. Investigative Radiology 2008;43(5):306–13. [18] Herborn CU, Watkins DM, Runge VM, et al. Renal arteries: comparison of steady-state free precession MR angiography and contrast-enhanced MR angiography. Radiology 2006;239(1):263–8.