| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology (K.L.C., J.A.V., H.T.B., D.J.B., G.M., P.H., R.H.O., G.J.M.) and Laboratory of Medical Imaging Research (F.M.), University Hospitals Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium; and Philips Medical Systems, Best, the Netherlands (R.M.H.). Received January 11, 2003; revision requested March 24; final revision received August 4; accepted September 29. Address correspondence to K.L.C. (e-mail: kcoenegrachts74@yahoo.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Twenty-five randomly selected patients (eight women and 17 men; age range, 27–88 years; mean age, 72 years) suspected of having renal artery stenosis underwent both DSA and balanced turbo field-echo MR angiography. A consensus result was obtained among three radiologists in evaluation of main renal arteries on balanced turbo field-echo images and DSA images. Sensitivity, specificity, and negative and positive predictive values of the balanced turbo field-echo technique were calculated, and receiver operating characteristic analysis was performed for depiction of hemodynamically significant stenosis. Cohen 
analysis was used to assess agreement between the two imaging methods in grading of stenoses and depiction of significant stenosis. Accessory renal arteries also were evaluated.
RESULTS: Fifty main renal arteries and 11 accessory arteries were fully depicted with DSA. DSA depicted 11 stenotic lesions in the main renal arteries. In comparison, balanced turbo field-echo MR angiography enabled visualization of 46 of 50 main renal arteries to their first branching points and depicted 10 of 11 accessory arteries. Sensitivity, specificity, negative predictive value, and positive predictive value of this technique for depiction of significant stenosis were 100% (four of four), 98% (41 of 42), 100% (41 of 41), and 80% (four of five), respectively. The area under the receiver operating characteristic curve was 0.988. was 0.782 for grading of stenoses and 0.877 for depiction of significant stenosis.
CONCLUSION: Multislab balanced turbo field-echo imaging has potential as an MR angiography technique for depiction of normal and diseased renal arteries.
© RSNA, 2004
Index terms: Magnetic resonance (MR), three-dimensional, 961.129419 • Magnetic resonance (MR), vascular studies, 961.12942 • Renal arteries, stenosis or obstruction, 961.721, 961.722
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Compared with contrast-enhanced MR techniques, flow-based MR techniques such as time-of-flight or phase-contrast imaging have never been used successfully for this purpose. One of the main problems has been insufficient vessel–to–soft-tissue signal intensity contrast relative to the saturation of the signal of inflowing blood. In addition, relatively long acquisition times made image acquisition during a single breath hold impossible with flow-based techniques. Image blurring was the result. In practice, only the proximal centimeters of the main renal arteries could be successfully visualized (9,10).
There are several clinical situations in which the use of a contrast agent for MR angiography compromises the quality of other investigations performed during the same MR imaging session. A typical example is the work-up of a patient prior to renal surgery: A single imaging session that includes the acquisition both of a renal perfusion study and of contrast-enhanced MR angiographic images is not optimal for either one or the other of these measurements. The relative invasiveness and cost of the intravenous contrast agent used at MR angiography limit the potential of this technique for large-scale screening.
The balanced turbo field-echo MR angiographic technique is based on a gradient-echo sequence in which the gradients are fully balanced without spoiling of the transverse magnetization. The result is a steady-state signal that is nearly independent of the actual repetition time. The signal-to-noise ratio is relatively high even without the use of a contrast agent (11). The short repetition time allows imaging during a single breath hold. High spatial resolution is achieved with the small voxel sizes and dedicated k-space schemes. Because of the specific characteristics of the balanced turbo field-echo technique on the one hand and the relatively high T2/T1 ratio of the blood on the other hand (11,12), blood vessels maintain a hyperintense signal over a longer distance than with conventional time-of-flight techniques.
The purpose of our study was to compare a multislab balanced turbo field-echo MR angiographic technique, without the use of a contrast agent, with DSA for imaging of the renal arteries.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MR Imaging Technique
All MR examinations were performed with a 1.5-T MR imager (Intera; Philips Medical Systems, Best, the Netherlands). The maximum gradient strength was 30 mT/m. The minimum time taken to reach this gradient amplitude was 200 µsec. All examinations were performed with a phased-array body coil.
A balanced fast field-echo sequence was applied in the coronal plane for localization of the origin of the main renal arteries and in the sagittal plane for localization of the origin of the superior mesenteric artery. A balanced turbo field-echo sequence then was applied with the following parameters: repetition time msec/echo time msec, 6.9/3.45; flip angle, 80°. The water-fat shift was 0.4 pixel (bandwidth, 542.8 Hz/pixel). The number of shots was 39. The 240 x 240 matrix was reconstructed to a 512 x 512 matrix. A field of view of 300 mm and a rectangular field of view of 35% (reduction to 35% in the number of phase encodings) were used. The imaging percentage was 100%. Reconstructed voxel dimensions were 0.59 x 0.59 x 1.5 mm. Water-selective excitation was performed for fat suppression. Phase wrapping was eliminated in the anteroposterior direction with activation of fold-over suppression (number of signals acquired, two) and caudocephalad with 5% section oversampling. Three saturation pulses were applied prior to each balanced turbo field-echo shot to provide venous signal suppression (Fig 1). Two saturation slabs covered the hilum of each kidney. A third saturation slab was positioned parallel and caudal to the three-dimensional (3D) imaging volume at a distance of 10 mm to provide suppression of signal from the inferior vena cava. The acquisition time per 3D slab was 21 seconds. Breath holding was performed at end expiration. The acquisition protocol consisted of two measurements: a single-slab acquisition at the level of the main renal arteries, and a multislab acquisition (multiple overlapping thin-slab acquisitions, or MOTSA) that covered the arterial anatomy from the origin of the superior mesenteric artery to the aortic bifurcation. Total imaging time was less than 10 minutes per patient.
|
Conventional angiography was preceded by sterile preparation of the groin with a 70% alcohol solution containing 0.5% chlorhexidine. Angiography was performed with local anesthesia by using a 1% xylocaine solution (Astra Zeneca, Brussels, Belgium). Prior to puncture of the femoral artery, a small cutaneous incision was made with a no. 11 surgical blade (Swann-Morton, Sheffield, England) held flat. Either a single arterial wall puncture or the Seldinger technique was used with a needle-styletted 18-gauge intravenous catheter (SurFlash; Terumo, Louvain, Belgium). After introduction of a 0.035-inch, 150-cm, angled-tip hydrophilic guide wire (Kayak; Boston Scientific Medi-Tech, Natick, Mass), a 5-F, 150-cm-long sheath (Radiofocus Introducer II; Terumo) was inserted, along which a calibrated pigtail catheter (Royal Flush; Cook, Winston-Salem, NC) was advanced to the estimated level of the ostium of the renal arteries. An anteroposterior view and two oblique projections at an angle of approximately 45° were obtained during the injection of 30 mL of iobitridol (Xenetix 300; Codali Guerbet, Brussels, Belgium), a nonionic contrast agent containing 300 mg of iodine per milliliter, at a rate of 15 mL per second by using an injector (Mark V Plus; Medrad, Indianola, Pa). Angiography was performed with a C-arm system (Integris 3000; Philips Medical Systems).
MR angiography and DSA were performed within 1 week of each other, with MR angiography preceding DSA in all cases.
Image Evaluation
All images were evaluated independently by three vascular radiologists (J.A.V., D.J.B., G.M., with 6, 16, and 11 years of experience, respectively, in the interpretion of MR angiography and DSA studies). All MR images were stored on optical disk and reviewed at a picture archiving and communication system workstation (Agfa, Antwerp, Belgium).
In a first session, the balanced turbo field-echo images were evaluated. To prevent bias, the images from the different patients were presented in random order. The DSA images were evaluated in a second session 2 weeks after the first session. In cases of disagreement among the three evaluating radiologists, a final evaluation was performed by consensus in a third evaluation session. A third evaluation session was performed for images from one patient.
Images were evaluated for depiction of the entire main renal artery to the point of its branching into the segmental arteries at the level of the renal hilum. Detection of stenosis at the level of the main renal arteries was recorded. Ostial location of stenosis was defined as location within 5 mm of the origin of the main renal artery; nonostial location was defined as location more than 5 mm from the origin of the main renal artery. Detected stenoses were classified in four groups according to grade (ie, percentage of reduction in luminal diameter): less than 50%, 50%–80%, more than 80%, and occlusion. All detected stenoses were measured with calipers on the images. Images also were evaluated for depiction of accessory renal arteries. In addition, the source images and MIP images obtained with balanced turbo field-echo MR angiography were assessed for artifacts. Motion artifacts were evaluated as absent, present without degradation of diagnostic quality, or present with degradation of diagnostic quality. Saturation of the venous signal and overlapping hyperintensities in the surrounding tissues were assessed as either absent or present.
Statistical Analysis
For statistical analysis, each stenosis grade was assigned an ordinal score. Absence of stenosis was assigned a score of 0, stenosis of less than 50% was assigned a score of 1, stenosis of 50%–80% was assigned a score of 2, stenosis of more than 80% was assigned a score of 3, and occlusion was assigned a score of 4.
Sensitivity, specificity, negative predictive value, and positive predictive value of balanced turbo field-echo MR angiography as a diagnostic test for renal artery stenosis were calculated for all main renal arteries evaluated for stenosis. Receiver operating characteristic analysis was performed for evaluation of all main renal arteries for hemodynamically significant stenosis. DSA was the method of reference. Because stenoses of more than 80% were considered hemodynamically significant, grades 0–2 were regarded as negative test results for renal artery stenosis and grades 3 and 4 were regarded as positive test results for renal artery stenosis. The Cohen analysis was used to test for agreement between balanced turbo field-echo MR angiography and DSA with respect to grade of stenosis and with respect to the presence or absence of hemodynamically significant stenosis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
value was 0.877 (95% CI: 0.639, 1.115). With respect to grade of stenosis, the value was 0.782 (95% CI: 0.578, 0.986). Balanced turbo field-echo MR angiography also depicted 10 of 11 accessory renal arteries (Fig 5).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Acquisition of volumes in the transverse plane is preferred to maximize the inflow effect, as in time-of-flight studies in the brain, in which multiple overlapping thin slabs are routinely acquired if the whole brain must be imaged (13). The shortest possible repetition time and echo time are used: A short echo time is used to minimize flow-induced dephasing, and a short repetition time guarantees an acceptable acquisition time.
Theoretically, the extent of the segment of the renal arteries visualized with the balanced turbo field-echo technique still partly depends on the inflow saturation effect. The amount of saturation is much smaller with the balanced turbo field-echo technique than with time-of-flight methods because of intrinsic preservation of the transverse magnetization. Saturation is not problematic for evaluation of the main renal arteries in either young or old patients on source images acquired with the balanced turbo field-echo technique, compared with the previously used flow-based techniques (time-of-flight and phase-contrast imaging). This finding is remarkable because only the first few proximal centimeters of the main renal arteries could be visualized with these earlier techniques on a routine basis (9,10).
The correct grading of stenoses has always been a challenge with MR angiography. Time-of-flight techniques tend to lead to overestimation of stenosis because of signal dispersion in turbulent flow. This situation is aggravated in phase-contrast imaging. Contrast-enhanced MR angiography is less sensitive to this effect because the echo time is extremely short. In a recent study, Saloner et al (14) found that even with contrast-enhanced MR angiography, grading of stenoses might be incorrect. A true reference standard may be intraoperative measurement. Ex vivo validation can be based on casts made from specimens, as proposed by Saloner et al (14). Specimens were not available to us, however, because most of the patients in our study did not undergo renal artery surgery. DSA was chosen as the reference standard.
This study shows that it is feasible to correctly grade stenotic lesions of different severity with the balanced turbo field-echo technique, although there is a tendency to overestimate stenoses with this technique, compared with DSA. This can be explained by a dephasing effect at the level of stenosis that is similar to known effects in time-of-flight and phase-contrast techniques. In the balanced turbo field-echo sequence, echo time is as short as possible to minimize its sensitivity to dephasing effects. There is no clear explanation for the underestimation of one stenosis with balanced turbo field-echo MR angiography.
In four of 50 main renal arteries in this study, imaging with the balanced turbo field-echo technique was suboptimal. In two main renal arteries, depiction was poor because the patient was uncooperative. In two main renal arteries, the part of the main renal artery distal to a high-grade stenosis could not be visualized until the point of branching into the segmental arteries. A dephasing effect at the level of these stenoses and a loss of sufficient inflow could explain the loss of signal in the main renal artery distal to the stenosis. In such cases, use of the balanced turbo field-echo technique for initial evaluation may result in a need for additional angiographic examinations that would have been obviated by use of contrast-enhanced MR angiography.
Stenosis evaluation with the balanced turbo field-echo sequence resulted in neither false-positive nor false-negative findings. This fact is important for the comparison of this technique with previously used flow-based techniques (time-of-flight and phase-contrast imaging), because many false-positive results in the past were caused by flow artifacts (9,15,16). The combination of a good inflow effect and a good T2/T1 ratio of the blood can explain these findings.
A benefit of the balanced turbo field-echo sequence is the acquisition speed and the ability to perform imaging during breath holding. The advantages of imaging during breath holding are obvious and include the reduction or avoidance of motion artifacts. The spatial resolution of images acquired with this technique is acceptable overall but may have to be further improved for more accurate assessment of stenoses.
The one patient with fibromuscular dysplasia is not a sufficiently large sample from which to draw conclusions concerning the ability of balanced turbo field-echo MR angiography to depict fibromuscular dysplasia. More patients with proved fibromuscular dysplasia at DSA will have to be evaluated with balanced turbo field-echo MR angiography for comparison in future studies.
The quality of the MIP images was mostly insufficient because of overlapping hyperintensities from fluid in the intestines, bile in the distal extrahepatic ducts, or fluid in a prominent ureter. In this regard, the source images from balanced turbo field-echo MR angiography are far more important for diagnosis and stenosis grading than are MIP images. More advanced visualization techniques should be used in future MR angiographic examinations with balanced turbo field-echo techniques.
The high sensitivity and negative predictive value of balanced turbo field-echo MR angiography for the evaluation of hemodynamically significant main renal artery stenoses suggest that this method may have prospects for use in the screening and selection of patients for interventional DSA of the renal arteries. With respect to grading of stenosis, the value reveals substantial agreement. With respect to depicting the presence of significant stenosis, the value reveals almost perfect agreement.
The main limitation of this study is the restricted number of cases. Further clinical studies are needed to confirm these preliminary results. Technical optimizations are needed to improve the quality of the MIP images and the grading of stenoses. A reduction of echo time may further reduce the dephasing effect at the level of stenoses. A further reduction of the acquisition time—with, for instance, parallel imaging or more powerful slew rates—will probably allow improvement of the results in uncooperative patients. The use of a navigator-gated technique might also help to improve image quality in uncooperative patients. An improvement of spatial resolution will probably be of benefit in the evaluation of fibromuscular dysplasia and of stenotic lesions.
In conclusion, the balanced turbo field-echo technique in most cases enabled accurate visualization of the main renal arteries to the point of their branching into the segmental arteries without the use of an intravenous contrast agent. Correlation with DSA was good in the grading of main renal artery stenoses and in the depiction of accessory renal arteries. The balanced turbo field-echo technique presented holds promise for accurate differentiation between normal arteries and arteries with atherosclerotic disease.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, K.L.C., H.T.B., R.H.O., G.J.M.; study concepts and design, K.L.C., H.T.B., R.H.O., G.J.M.; literature research, K.L.C., H.T.B.; clinical studies, K.L.C., J.A.V., G.M., P.H.; data acquisition, K.L.C., P.H.; data analysis/interpretation, K.L.C., R.M.H., H.T.B., D.J.B., G.M.; statistical analysis, K.L.C., F.M.; manuscript preparation, K.L.C., H.T.B.; manuscript definition of intellectual content, K.L.C., H.T.B., R.H.O., G.J.M.; manuscript editing, K.L.C., H.T.B.; manuscript revision/review, J.A.V., R.H.O., G.J.M.; manuscript final version approval, H.T.B., G.J.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  X. Liu, N. Berg, J. Sheehan, X. Bi, P. Weale, R. Jerecic, and J. Carr Renal Transplant: Nonenhanced Renal MR Angiography with Magnetization-prepared Steady-State Free Precession Radiology, May 1, 2009; 251(2): 535 - 542. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Miyazaki and V. S. Lee Nonenhanced MR Angiography Radiology, July 1, 2008; 248(1): 20 - 43. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Wyttenbach, A. Braghetti, M. Wyss, M. Alerci, L. Briner, P. Santini, L. Cozzi, M. Di Valentino, M. Katoh, C. Marone, et al. Renal Artery Assessment with Nonenhanced Steady-State Free Precession versus Contrast-enhanced MR Angiography Radiology, October 1, 2007; 245(1): 186 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Maki, G. J. Wilson, W. B. Eubank, D. J. Glickerman, J. A. Millan, and R. M. Hoogeveen Navigator-Gated MR Angiography of the Renal Arteries: A Potential Screening Tool for Renal Artery Stenosis Am. J. Roentgenol., June 1, 2007; 188(6): W540 - W546. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
