The value of true-FISP sequence added to conventional gadolinium-enhanced MRA of abdominal aorta and its major branches

The value of true-FISP sequence added to conventional gadolinium-enhanced MRA of abdominal aorta and its major branches

European Journal of Radiology 72 (2009) 489–493 Contents lists available at ScienceDirect European Journal of Radiology journal homepage: www.elsevi...

538KB Sizes 0 Downloads 14 Views

European Journal of Radiology 72 (2009) 489–493

Contents lists available at ScienceDirect

European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad

The value of true-FISP sequence added to conventional gadolinium-enhanced MRA of abdominal aorta and its major branches Andrea Iozzelli 1 , Giovanni D’Orta 2 , Alberto Aliprandi 1 , Francesco Secchi 1 , Giovanni Di Leo 1 , Francesco Sardanelli ∗ University of Milan School of Medicine, Department of Medical and Surgical Sciences, Radiology Unit, IRCCS Policlinico San Donato, via Morandi 30, 20097 San Donato Milanese, Milan, Italy

a r t i c l e

i n f o

Article history: Received 6 February 2008 Received in revised form 30 July 2008 Accepted 2 September 2008 Keywords: Magnetic resonance (MR) Aorta Magnetic resonance angiography (MRA)

a b s t r a c t To test true-fast imaging with steady-state precession (true-FISP) added to gadolinium-based MR angiography (Gd-MRA) for imaging abdominal aorta and major abdominal vessels, 35 consecutive patients (age 67 ± 11 years) with known or suspected abdominal and/or peripheral vascular disease were studied with sagittal and axial 2D true-FISP during free breathing and coronal 3D fast low-angle shot (FLASH) Gd-MRA (breath-holding, 0.2 mmol/kg of Gd-DOTA at 2 ml/s). We evaluated: suprarenal aorta, celiac trunk, superior mesenteric artery, right renal artery, left renal artery, infrarenal aorta, inferior mesenteric artery, aortic bifurcation/common iliac arteries, lumbar arteries and aortic atheromasia. The possible presence of accessory renal arteries, collateral vasculature and vascular prosthesis/stent was evaluated. A quality four-point score was assigned to each item on both sequences, from 0 (not visible) to 3 (good-to-excellent image quality) and Wilcoxon test was used. Main diagnoses resulted: normal or atheromasic aorta (n = 25); aortic aneurysm (n = 2); patent aorto-iliac surgical prosthesis (n = 2); patent vascular iliac stent (n = 2); aneurysm of iliac artery (n = 1); patent aortic endovascular prosthesis (n = 1); patent aorto-femural bypass (n = 1) and aorto-iliac surgical prosthesis endoleak (n = 1). We also found three patients with accessory renal arteries, two with collateral circulation, and three with surgical aorto-iliac prosthesis. The score of true-FISP (25.9 ± 4.1, median 27) was significantly higher (p = 0.003) than that of Gd-MRA (23.9 ± 3.6, median 24). True-FISP was superior for visualizing inferior mesenteric artery (score 2.5 ± 1.1 vs. 1.0 ± 1.4; p < 0.001) and atheromasic plaques (2.5 ± 1.1 vs. 1.2 ± 1.1; p < 0.001). One collateral vasculature was demonstrated only with Gd-MRA. Summarizing, true-FISP is a power and fast non-breath-hold sequence to be added to Gd-MRA, obtaining an information increase. © 2008 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Incidence of atherosclerosis of aortic and abdominal arteries is growing up in developed countries [1,2]. During the last decade non-invasive diagnostic tools, such as computed tomography angiography and magnetic resonance angiography (MRA), were added to digital subtraction angiography. Earlier MR approaches, i.e. black and bright blood techniques, did not use contrast agents [3–5]. Nowadays, MRA with intravenous administration of param-

∗ Corresponding author. Tel.: +39 0252774468; fax: +39 025279695. E-mail addresses: [email protected] (A. Iozzelli), [email protected] (G. D’Orta), [email protected] (A. Aliprandi), [email protected] (F. Secchi), [email protected] (G. Di Leo), [email protected] (F. Sardanelli). 1 Tel.: +39 0252774468; fax: +39 025279695. 2 Present address: Clinica S. Anna, Department of Radiology, via del Franzone, 31, 25127 Brescia, Italy. Tel.: +39 03031971; fax: +39 0303197635. 0720-048X/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2008.09.004

agnetic contrast agent [6] is the commonly accepted MR technique to evaluate abdominal aorta and its main branches. Recently, a new unenhanced sequence has been introduced in clinical practice: the true-fast imaging with steady-state precession (true-FISP), a hybrid T2/T1-weighted acquisition which enable us to depict anatomy and morphology of organs and vessels with high details [7,8]. The purpose of our study was to assess the possibility to evaluate aorta and its main abdominal branches using true-FISP sequence and to investigate whether this sequence, added to the usual threedimensional (3D) Gd-enhanced MRA (Gd-MRA), is able to give an increase in information. 2. Materials and methods Institutional Review Board approval for this retrospective study was obtained and informed consent was not required. We retrospectively evaluated 35 consecutive patients referred to our

490

A. Iozzelli et al. / European Journal of Radiology 72 (2009) 489–493

Department for abdominal MRA from January to August 2004. Patients were 21 males and 14 females (mean age ± standard deviation 67 ± 11 years, median 69 years, range 25–80 years), with known or suspected abdominal and/or peripheral vascular disease. All patients underwent the examination using a 1.5-T MR unit (Magnetom Sonata Maestro Class, Siemens, Erlangen, Germany). No patient preparation was needed. Patients were examined in supine position with a system of phased array body coils composed by eight channels for abdomen, pelvis and spine. After a 3-plane localizer sequence, our MR protocol consisted of:

Table 1 Mean scores assigned to 3D Gd-MRA and true-FISP sequences in 35 patients, for each of the ten items. 3D Gd-MRA Suprarenal aorta Celiac trunk Superior mesenteric artery Right renal artery Left renal artery Infrarenal aorta Inferior mesenteric artery Iliac arteries—aortic bifurcation Lumbar arteries Wall atheroma Total score

1 Free-breathing axial and sagittal true-FISP with the following sequence parameters: TR/TE, 4.3/2.2 ms; flip angle, 74◦ ; two excitations; slice thickness, 3 mm (sagittal plane) or 5 mm (axial plane); 64 (axial) and 48 (sagittal) no-gap slices; pixel size, 1.4 mm × 1.4 mm; acquisition time, 50 s (sagittal) and 73 s (axial). 2 Coronal breath-hold fast low-angle shot (FLASH) 3D-MRA (TR/TE, 3.68/1.2 ms; flip angle, 30◦ ; one excitation; 56 partitions; slice thickness, 1.7 mm; pixel size, 1.7 mm × 1.3 mm; acquisition time, 15 s) acquired before and after the intravenous administration of 0.2 mmol/kg of Gd-DOTA (Dotarem, Guerbet, France) at 2 ml/s, followed by 20 ml of saline flush, administered using an automatic injector (Spectris Injection System, Medrad, Pittsburgh, USA). To define the time delay between contrast injection and image acquisition, we used a semi-automatic synchronization protocol based on the visual inspection of the contrast bolus arrival in the left ventricle (care-bolus).

The software automatically generated subtracted images (postcontrast minus precontrast) for 3D Gd-MRA. Gd-MRA postprocessing was performed with maximum intensity projection, multiplanar reconstruction and surface shaded display. True-FISP images did not require post-processing. Two radiologists not involved in the first clinical report of these examinations examined all the cases, by consensus. They had 5 years (first author) and 2 years (second author) of experience in reporting abdominal MRA. They performed two independent sessions with patient’s random order of presentation, separated by two months time interval, one session for true-FISP images, one session for Gd-enhanced MRA (native unenhanced and contrast-enhanced images, subtracted images and multiplanar reconstructions). Ten items were evaluated for both sessions: suprarenal aorta; celiac trunk; superior mesenteric artery; right renal artery; left renal artery; infrarenal aorta; inferior mesenteric artery; aortic bifurcation/common iliac arteries; lumbar arteries; aortic atheromasia. For each of these ten items and for both sequences, a four-point quality score (0 = not visible; 1 = poor quality representation; 2 = fair quality representation; 3 = good-to-excellent quality representation) was assigned. Wilcoxon signed rank test was used for statistical analysis. p-values less than 0.05 were considered as significant. Calculations were done using SPSS software (SPSS for Windows, release 6.0, SPSS, Chicago, IL, USA). The presence of accessory renal arteries, collateral arterial vessels and arterial prosthesis or stent was evaluated by consensus for visibility, distinguishing among non-visible; poor quality representation; fair quality representation; good-to-excellent quality representation.

3.0 2.5 2.8 2.6 2.4 3.0 1.0 2.9 1.9 1.2

± ± ± ± ± ± ± ± ± ±

0.0 (0) 0.9 (3) 0.6 (3) 0.7 (3) 0.9 (3) 0.0 (3) 1.4 (0) 0.4 (3) 1.0 (2) 1.1 (1)

23.9 ± 3.6 (24)

True-FISP 3.0 2.5 2.7 2.4 2.3 2.9 2.5 3.0 1.7 2.5

± ± ± ± ± ± ± ± ± ±

0.0 (3) 0.9 (3) 0.7 (3) 0.7 (3) 1.1 (3) 0.4 (3) 1.1 (3) 0.2 (3) 1.0 (2) 1.1 (3)

25.9 ± 4.1 (27)

p-Valuea n.s. n.s. n.s. n.s. n.s. n.s. <0.001 n.s. n.s. <0.001 0.003

Data are reported as mean ± standard deviation (median). a Wilcoxon test.

3. Results The overall examination time was approximately 15 min in all patients, including patient positioning and obtaining the venous access. The main final radiological diagnosis resulted as follows: normal or atheromasic aorta (n = 25); aortic aneurysm (n = 2); patent aorto-iliac surgical prosthesis (n = 2); patent vascular iliac stent (n = 2); aneurysm of iliac artery (n = 1); patent aortic endovascular prosthesis (n = 1); patent aorto-femural bypass (n = 1); aorto-iliac surgical prosthesis endoleak (n = 1). For the true-FISP the total score per patient was 25.9 ± 4.1 (mean ± standard deviation), median 27, significantly greater (p = 0.003) than that of MRA (23.9 ± 3.6, median 24). The performance of true-FISP was higher (p < 0.001) than Gd-MRA in particular in the depiction of the inferior mesenteric artery (Fig. 1), with a mean score of 2.5 ± 1.1 (median 3) vs. 1.0 ± 1.4 (median 0) and of intravascular atheromasic plaque (Fig. 2), with a mean score of 2.5 ± 1.1 (median 3) vs. 1.2 ± 1.1 (median 1). For all other examined items, the differences were not statistically significant. These results are summarized in Table 1. Three accessory renal arteries (two right and one left) were found in three patients, with a good-to-excellent representation with Gd-MRA and poor quality representation with true-FISP. Aorto-iliac surgical prosthesis was found in three patients. A common iliac artery endovascular stent was found in two patients. An endovascular prosthesis of an infrarenal aortic aneurism was found in one patient (Fig. 3). Collateral circulation was observed in two patients with monolateral femoro-iliac occlusion. The evaluation of these particular findings is summarized in Table 2, showing that while only Gd-MRA enabled us to detect collateral vessels, only true-FISP permitted a good-to-excellent evaluation of arterial prostheses.

4. Discussion The availability of MR units with stronger and more rapid gradients allowed the development of fast 3D Gd-MRA that replaced almost completely the use of flow-dependent MRA techniques without contrast agent administration. The pitfalls of these older sequences relied on their dependence on flow velocity (with consequent insufficient demonstration of vessels with extremely low flow) and flow dynamics (with consequent artefacts from turbulence and the typical signal-void artefact overestimating the real degree of stenosis using time-of-flight sequences) [9,10]. Moreover, the sequence time was too long for breath-hold acquisition while Gd-MRA can be acquired during a single breath-hold.

A. Iozzelli et al. / European Journal of Radiology 72 (2009) 489–493

491

Fig. 1. 72-year-old woman with abdominal atherosclerosis. The axial true-FISP (a and b) clearly shows the inferior mesenteric artery on two consecutive slices (arrow). This artery was not visible with Gd-enhanced MR angiography, as shown by maximum intensity projections: left oblique (c); coronal (d) and right oblique (e).

Fig. 2. 84-year-old woman with aortosclerosis previously detected by doppler ultrasound examination. Axial reconstructions from 3D gradient-echo data volumes for MR angiography at the level of infrarenal aortic aneurysm: unenhanced (a), contrast-enhanced (b) and subtracted (c) images. True-FISP axial image (d) obtained at the same level shown in a–c. True-FISP sagittal image (e) and sagittal reconstruction of subtracted Gd-enhanced MR angiography (f) showing the largest antero-posterior diameter of the aortic aneurysm image. The thrombotic deposition on the aortic wall cannot be detected in a and could be only suspected due to the polycyclic contour of the contrast-enhanced lumen (arrows and arrowheads) in b and c; in particular, in b, a low-signal thickened aortic wall is poorly visualized with an external indistinct contour. The thrombus is clearly depicted and differentiated from the lumen and from periaortic tissues in the axial (d) and sagittal (e) true-FISP images (d). The sagittal maximum intensity projection of subtracted contrast-enhanced MR angiography shows the aneurysm but the cranio-caudal thrombus extent is clearly visible only in e (arrows).

492

A. Iozzelli et al. / European Journal of Radiology 72 (2009) 489–493

Fig. 3. 65-year-old man with an infrarenal aortic aneurysm treated with endovascular prosthesis. Only the axial true-FISP image (a) demonstrates the wall of the endoprosthesis (arrows), not visible at all with Gd-enhanced MR angiography, as shown in the coronal maximum intensity projection (b). Table 2 Evaluation of particular findings in eight patients.

Accessory renal arteries Collateral circulationa Aortic surgical prosthesisb Aortic endovascular prosthesisc

Patients

Number

3D Gd-MRA

True-FISP

3 2 3 1

3 2 3 1

Good-to-excellent Good-to-excellent Poor quality Poor quality

Poor quality Not visible Good-to-excellent Good-to-excellent

Note: One patient had both collateral circulation and aorto-iliac prosthesis, so that nine particular findings were detected in eight patients. The patients with common iliac artery stent showed metallic artifacts, producing signal void, simulating occlusion in the Gd-MRA sequence. a There were collateral circulation from lumbar arteries because of monolateral common iliac artery occlusion. b There were three surgical aorto-iliac prostheses in gore-tex or dacron; no artifact from metallic components was present with both sequences, so that the evaluation here reported concerned the image quality of the prostheses themselves. c Not specified type.

True-FISP is a gradient-echo sequence, also called balanced fast field eco (FFE) or fast imaging employing steady-state acquisition (FIESTA). In this sequence the transverse magnetization is maintained in the steady state from one TR to the next one by rewinding the gradient waveforms on all axes. The gradients are perfectly balanced and the total gradient area is zero at each radio-frequency pulse [11–13]. This sequence scheme generates a spin-echo from a gradient-echo sequence (due to the refocusing effect of the 90◦ pulse on the residual transverse magnetization), responsible for important T2-weighting and low motion artefact susceptibility. Moreover, it shows high contrast-to-noise ratio thanks to the high contrast between different anatomic structures mainly due to its hybrid T2/T1 weighting. Thus, a typical feature of this sequence is to show intravascular bright signal both in arteries and veins, independently from the direction of the vessel axis respect to the sequence acquisition plane and from the flow velocity. As a consequence, there is no need of intravenous injection of paramagnetic contrast agent to show the vessel lumen. The outer vessel contour is also well depicted with good contrast resolution. Breath-hold acquisition of true-FISP sequence is generally preferred, but the short acquisition time and the retroperitoneal position of the aorta allow a fair quality also in free breathing, a possibility for non-cooperating patients. Vascular inner contours and morphology of atheromasic plaques are sometimes not clearly defined on images obtained with T2-weighted fast spin-echo sequences, [14,15] while using 3D GdMRA, blurred vascular contours are frequently generated by the passage of contrast agent because of non-uniform k-space weighting in 3D acquisition [16]. The usefulness of true-FISP in the aortic evaluation was already previously observed by Pereles et al. in their study on the morphologic evaluation of thoracic aorta with dissection or aneurysm [17]. We confirm this evaluation on abdominal aorta and its main branches. Our experience indicates that true-FISP added to 3D Gd-

MRA may provide an increase of information in clinical practice. The former sequence resulted in displaying important details especially in the detection of arterial wall thrombosis and in visualizing inferior mesenteric artery. True-FISP may depict the vessel wall and the outer parts of the vessels with high image quality, while only the lumen is demonstrated using Gd-MRA, as it is with all angiographic techniques. Moreover, true-FISP correctly detected the presence and morphology of an endovascular device, not demonstrated with Gd-MRA (see Fig. 3). We could speculate that true-FISP might demonstrate possible extrinsic vascular compressions and extravascular abdominal pathology, difficult to evaluate using only 3D Gd-MRA. The reasons for a significantly lower visualization of the inferior mesenteric artery at Gd-MRA than that possible with true-FISP could be related to a combination of different factors: (a) relatively low spatial resolution of the Gd-MRA (3.8-mm3 voxel) to be compared with a vessel like an inferior mesenteric artery which could have a very small diameter, especially in a subject with atherosclerosis (note, in our experience, the superior in-plane resolution of the true-FISP sequence and its independence of the arterial lumen visualization from the arrival of a contrast agent - see Fig. 1); (b) a possible presence of stenosis at the origin of the inferior mesenteric artery limiting the contrast-enhanced blood flow; and (c) the effect of post-processing based on maximum intensity projection algorithm which could delete a low-enhancing vessel from the reconstructed 3D images. Some details probably would have been better depicted using a Gd-MRA with a higher matrix, but that sequence would take more time than that limited to arterial phase. Probably better results using high resolution MRA will be achieved with forthcoming blood pool contrast agents [18–20]. True-FISP sequences could be useful for non-collaborating patients, especially in emergency situations, when it is not possible

A. Iozzelli et al. / European Journal of Radiology 72 (2009) 489–493

to obtain a good breath-hold MRA. In these conditions, trueFISP could obtain even better results than unenhanced spin-echo or fast spin-echo sequences, as used in the past [21]. In some cases true-FISP alone can be a good alternative to MRA or computed tomography. This sequence can be used in non-cooperating patients and in case of contraindication to paramagnetic contrast administration or difficult peripheral vein catheterization. 5. Conclusion In conclusion, true-FISP resulted in a powerful and fast freebreathing sequence that can be added to a standard Gd-MRA obtaining an increase of diagnostic information with good details in showing the anatomy of abdominal arteries. Conflict of interest None. References [1] Jayalath RW, Mangan SH, Golledge J. Aortic calcification. Eur J Vasc Endovasc Surg 2005;30:476–88. [2] Solberg LA, Strong JP. Risk factors and atherosclerotic lesions. A review of autopsy studies. Arteriosclerosis 1983;3:187–98. [3] Anderson CM, Lee RE. Time-of-flight techniques: pulse sequences and clinical protocols. Magn Reson Imaging Clin N Am 1993;1:217–27. [4] Jara H, Barish MA, Black-blood. MR angiography: techniques, and clinical applications. Magn Reson Imaging Clin N Am 1999;7:303–17. [5] O’Donnell MN. MR blood flow imaging using multiecho phase contrast sequences. Med Phys 1985;12:59–64. [6] Prince MR, Yucel EK, Kaufman JA, Harrison DC, Geller SC. Dynamic gadoliniumenhanced three-dimensional abdominal MR arteriography. J Magn Reson Imaging 1993;3:877–81. [7] Duerk JL, Lewin JS, Wendt M, Petersilge C. Remember true FISP? A high SNR, near 1-second imaging method for T2-like contrast in interventional MRI at.2 T. J Magn Reson Imaging 1998;8:203–8.

493

[8] Lewin JS, Connell CF, Duerk JL, et al. Interactive MRI-guided radiofrequency interstitial thermal ablation of abdominal tumors: clinical trial for evaluation of safety and feasibility. J Magn Reson Imaging 1998;8: 40–7. [9] Sebok NR, Sebok DA, Wilkerson D, Mezrich RS, Zatina M. In-vitro assessment of the behavior of magnetic resonance angiography in the presence of constrictions. Invest Radiol 1993;28:604–10. [10] Tsuruda J, Saloner D, Norman D. Artifacts associated with MR neuroangiography. Am J Neuroradiol 1992;13:1411–22. [11] Bundy J, Simonetti O, Laub G, Finn JP. Segmented trueFISP imaging of the heart (abstract). In: Proceedings of the Seventh Meeting of the International Society for Magnetic Resonance in Medicine. 1999. p. 1282. [12] Deimling M, Heid O. Magnetization prepared trueFISP imaging (abstract). In: Proceedings of the Second Meeting of the Society of Magnetic Resonance. 1994. p. 495. [13] Oppelt A, Graumann R, Barfuss H. FISP: a new fast MRI sequence. Electromedica 1986;54:15–8. [14] Kelekis NL, Semelka RC, Worawattanakul S, Molina PL, Mauro MA. Magnetic resonance imaging of the abdominal aorta and iliac vessels using combined 3-D gadolinium-enhanced MRA and gadolinium-enhanced fatsuppressed spoiled gradient echo sequences. Magn Reson Imaging 1999;17: 641–51. [15] Rodenwaldt J, Kopka L, Meyer HJ, Vosshenrich R, Funke M, Grabbe E. Determination of contour clarity in contrast-enhanced MR angiography: definition and clinical evaluation exemplified by ECG-triggered imaging of the thoracic aorta. Rofo 1998;169:616–21. [16] Fain SB, Riederer SJ, Bernstein MA, Huston III J. Theoretical limits of spatial resolution in elliptic-centric contrast-enhanced 3D-MRA. Magn Reson Med 1999;42:1106–16. [17] Pereles FS, McCarthy RM, Baskaran V, et al. Thoracic aortic dissection and aneurysm: evaluation with nonenhanced true-FISP MR angiography in less than 4 minutes. Radiology 2002;223:270–4. [18] de Haen C, Anelli PL, Lorusso V, et al. Gadocoletic acid trisodium salt (b22956/1): a new blood pool magnetic resonance contrast agent with application in coronary angiography. Invest Radiol 2006;41:279–91. [19] Prince MR, Meaney JF. Expanding role of MR angiography in clinical practice. Eur Radiol 2006;16(Suppl. 2):B3–8. [20] Tombach B, Reimer P, Bremer C, et al. First-pass and equilibrium-MRA of the aortoiliac region with a superparamagnetic iron oxide blood pool MR contrast agent (SH U 555 C): results of a human pilot study. NMR Biomed 2004;17: 500–6. [21] Nienaber CA, von Kodolitsch Y, Nicholas V, et al. The diagnosis of thoracic aortic dissection by noninvasive imaging procedures. N Engl J Med 1993;328: 1–9.