| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology, New York University Medical Center, 530 First Ave, HCC Basement-MRI, New York, NY 10016. Received April 30, 1998; revision requested July 6; revision received July 24; accepted October 19. Address reprint requests to V.S.L.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Three-dimensional spoiled gradient-echo imaging (3.8–4.2/1.3–1.7 [repetition time msec/echo time msec], 25°–40° flip angle) was performed in 60 patients after administration of gadopentetate dimeglumine (average dose, 0.11 mmol/kg). Synchronization of contrast material administration with data acquisition was achieved with a 1-mL test dose of contrast material to estimate patient circulation parameters. Image quality was assessed by using contrast-to-noise (CNR), relative vascular enhancement, and venous-to-arterial enhancement ratios and subjective scoring of arterial and venous enhancement. The effect of the contrast material injection rate and the influence of breath holding during the timing examination also were examined.
RESULTS: Overall, of 60 studies, 58 were diagnostic and 56 demonstrated excellent arterial enhancement. Venous enhancement was seen in eight studies. The average aortic relative vascular enhancement (± SD) was 14.6 ± 5.9, with an aorta-to-inferior vena cava (IVC) CNR of 69.7 ± 43.9. The IVC-to-aorta venous-to-arterial enhancement ratio averaged 0.08 ± 0.16. There was no significant difference in image quality based on injection rates or the performance of breath holding during the timing examination (P > .1).
CONCLUSION: Breath-hold gadolinium-enhanced renal MR angiography free of venous enhancement can be performed consistently and reliably with 20 mL of contrast material when studies are synchronized to patient circulation time by using a timing examination.
Index terms: Gadolinium • Kidney, diseases, 81.1421, 81.1452, 81.3121, 81.72, 81.84, 81.893, 81.897, 961.721, 961.723 • Magnetic resonance (MR), angiography, 961.129412, 961.12942 • Magnetic resonance, contrast agents, 961.12943 • Magnetic resonance (MR), pulse sequences, 961.129412, 961.129417
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The success of gadolinium-enhanced MR angiography depends on maximal contrast enhancement of the arteries of interest during data acquisition. Early MR imaging sequences required relatively long imaging times (at least 2–4 minutes), and, therefore, slow continuous infusions of high doses of contrast material (30–60 mL) were used to ensure adequate enhancement of the arteries during imaging (8–11). However, the disadvantages of this approach include artifacts associated with respiratory and bulk motion and frequent inadvertent venous enhancement (12). Recently, faster three-dimensional (3D) gradient-echo MR imaging pulse sequences that can be used within the time frame of a single breath hold without loss of spatial resolution have become available (12–20). Faster imaging times help to minimize motion artifacts, improve patient tolerance, and increase patient throughput. Moreover, the decreased echo times of the newer sequences can help to reduce the signal intensity loss associated with spin dephasing and therefore may help to minimize the overestimation of stenoses that, to varying degrees, plague all bright-blood MR angiographic techniques (21).
Shorter data acquisition times also offer the possibility for reduced doses of contrast material and thus have the potential advantage of decreased venous enhancement. However, shorter acquisition times and reduced doses (resulting in shorter periods of arterial enhancement) place greater demands on the imaging technique and require more precise coordination of imaging with the contrast material administration. This can be difficult because the transit time of the contrast material bolus from the arm vein to the systemic artery, also referred to as the patient circulation time, may vary from 10–60 seconds between patients (22–24) and is not predictable a priori. Hence, the synchronization of peak arterial enhancement that lasts approximately 20 seconds and occurs anywhere from 10–60 seconds after the injection with a 20-second acquisition time can pose a difficult challenge.
Several techniques to ensure data acquisition during arterial transit of the contrast material bolus have been reported. These include the use of real-time detection of the bolus arrival to initiate the data acquisition either automatically (25,26) or with an operator (27), partial k-space updating to increase temporal resolution (28), and a timing examination with a test dose of contrast material to determine the patient circulation time (22,29). While each method has advantages, the use of a timing examination does not require special pulse sequences and can be performed with almost any imaging unit. This approach has proved to be reliable and efficacious for single-dose gadolinium-enhanced MR angiography of the aorta (22) and carotid arteries (30) and for arterial phase contrast-enhanced examinations of the liver (31).
We have used a strategy in which an initial 1-mL test dose of gadolinium-based contrast material was used to optimize renal MR angiography in 60 patients and assessed the resultant image quality, including the extent of venous enhancement. We considered the effects of two variables on patient circulation time and subsequent image quality—the effect of breath holding during the timing examination and the effect of the contrast material injection rate (ie, 2 mL/sec vs 3 mL/sec). Our purpose was to assess the quality of breath-hold gadolinium-enhanced 3D MR angiography of the renal arteries by using a single 19-mL dose of contrast material optimized with a timing examination with 1 mL of contrast material.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Imaging Technique
Imaging was performed in all patients by using a 1.5-T superconducting magnet (Magnetom Vision; Siemens, Erlangen, Germany), with commercially available high-performance gradients capable of a 600-µsec rise time and 25 mT/m maximum gradient strength. A torso phased-array coil was used in all cases. To minimize wraparound artifact, the arms were placed on support pads to elevate them outside the coronal plane of the kidneys. A 22-gauge venous catheter was placed in an antecubital or forearm vein before the start of the study and attached to an MR-compatible power injector (Spectris; Medrad, Pittsburgh, Pa).
For MR angiography, a 3D spoiled gradient-echo sequence with fat saturation and interpolated in the section-select direction was used both before and at two time points (ie, the arterial and venous phases) after contrast material injection. The parameters for all three acquisitions were identical, and all were performed with the patient instructed to stop breathing at end expiration. The imaging slab, which was localized by using axial T1-weighted gradient-echo images (183/4.1 [repetition time msec/echo time msec], 90° flip angle), was positioned in the oblique coronal plane to encompass the abdominal aorta and both kidneys in the thinnest possible slab.
In the first 40 patients, the sequence parameters for the 3D acquisition were 4.2/1.7 and a 25°–40° flip angle, whereas in the remaining 20 patients, a modified version of the same sequence with a shorter repetition time and echo time (ie, 3.8/1.3) and 25° flip angle became available. For all sequences, the interpolation in the section-select direction was performed by using zero filling. The remainder of the imaging parameters were as follows: bandwidth, 488 Hz per pixel; field of view, 300–450 cm (with the rectangular field of view determined on the basis of patient size); slab thickness, 72–130 mm; 24–36 partitions; actual section thickness, 2–5 mm; interpolated section thickness, 1.0–2.5 mm; and matrix, 256 x 128–230. Overall, the average acquisition time was 22 seconds ± 4 (range, 13–37 seconds). For these sequences, k space is acquired sequentially—that is, central lines of k space are acquired during the middle of the acquisition.
Arterial phase postcontrast 3D gradient-echo images were timed to contrast material injection on the basis of the timing examination described below. A delayed image was acquired approximately 45 seconds after the end of the first image acquisition to achieve a venous phase acquisition; the delay was selected on the basis of prior experience. The venous phase was used to determine the precise location of the veins and inferior vena cava (IVC) for arterial phase region of interest analysis. All patients received a total of 20 mL of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) (average dose, 0.11 mmol per kilogram of body weight ± 0.03; dose range, 0.07–0.2 mmol/kg). One milliliter of contrast material was used in the timing examination, and 19 mL was used in the diagnostic study. The injection rates used in the MR angiographic study matched those that were used in the timing examination (2 or 3 mL/sec).
Timing Examination
The technique used in the timing examination is based on that reported in the literature (22). A power injector was used to administer a 1-mL test dose of gadopentetate dimeglumine, which was immediately followed by 20 mL of saline, both of which were administered at a rate of either 2 mL/sec or 3 mL/sec, which were the same rates used in the diagnostic study. Axial magnetization-prepared fast spoiled gradient-echo (turbo fast low-angle shot [turboFLASH; Siemens]) imaging (7.7–11.0/4.2, 300-msec inversion time, 15° flip angle, 1-cm-thick sections) was performed to acquire one image through the aorta at the level of the renal arteries every 2 seconds for 60 seconds. The turbo fast low-angle shot acquisition was initiated simultaneously at the onset of the test dose injection. Viewed at the console, the image with maximal aortic enhancement was identified visually and used to determine the time to peak enhancement, which was considered to equal the patient circulation time (Tcirc) (Fig 1).
|
|
The equation used to calculate the delay between the onset of the intravenous contrast material injection and the start of the image acquisition was Tdelay = Tcirc + Tgad/2 - Tacq/2, where Tgad is the duration of the gadopentetate dimeglumine administration, and Tacq is the image acquisition time (23). This equation was derived to allow synchronization of the midpoint of arrival of the contrast material bolus, which occurs roughly at Tcirc + Tgad/2, with the acquisition of the central portion of k space (Tdelay + Tacq/2). For example, if the time to peak aortic enhancement was determined to be 18 seconds on the basis of the timing examination (Fig 1), then by using an imaging time of 24 seconds and an injection of 19 mL of contrast material at 2 mL/sec, the Tdelay would equal 18 sec + [(19 ÷ 2 mL/sec)/2] - (24 sec/2) = 18 sec + 4.5 sec - 12 sec = approximately 11 sec. The operator would then initiate imaging 11 seconds after the start of the contrast material administration. Because all MR angiographic imaging was performed at end expiration, breath-holding instructions for the patient were typically given during this image delay period.
Image Analysis
Source data were analyzed by a radiologist, who drew regions of interest over the abdominal aorta at the level of the renal arteries, right and left renal arteries, right and left renal veins, suprarenal IVC, both kidneys, retroperitoneal fat, paraspinal muscle, and noise outside the body to measure the mean and SD of signal intensity. Images acquired during the venous phase were used to define the locations of the venous structures on the precontrast and arterial phase images. The signal-to-noise ratios in the aorta and renal arteries during arterial phase imaging were defined as the mean signal intensity divided by the SD of the background noise. Aorta-to-IVC, right renal artery-to-right renal vein, and left renal artery-to-left renal vein contrast-to-noise ratios were also obtained to assess the relative enhancement of the venous structures. In cases in which the coronal field of view was defined to exclude all areas outside the patient's body, the SDs of the signal intensity measurements in the regions of interest in the peripheral lung were used to estimate the background noise. The values used with this method did not differ significantly (P > .1) from the background noise outside the body.
Relative vascular enhancement, which is equal to the arterial phase signal intensity divided by the precontrast signal intensity, was recorded in the six major vessels. We also calculated the venous-to-arterial enhancement ratios for the aorta-to-IVC and right and left renal vein-to-renal artery as follows: venous-to-arterial enhancement ratio = (arterial phase V - precontrast V)/(arterial phase A - precontrast A), where V represents venous signal intensity and A, arterial signal intensity (27). The ideal venous-to-arterial enhancement ratio, which would equal 0, would indicate a venous enhancement–free study with preferential arterial enhancement.
The maximum intensity projection (MIP) algorithm was applied to all contrast-enhanced studies by using commercially available software on the MR imaging system (Magnetom Vision, software Numaris 3, version VB31C; Siemens). MIP images and selective multiplanar reconstructions were then scored subjectively at the workstation for degree of arterial and venous enhancement on a scale of 0 to 2 (0 = no enhancement, 1 = mild enhancement, 2 = substantial enhancement) by a single reader (V.S.L.) who was involved in many of the clinical studies, but subsequently, during the retrospective review, was blinded to the method used to perform the study. The studies were also subjectively assessed for diagnostic quality and placed in one of two categories—diagnostic or nondiagnostic—on the basis of source images and MIP and multiplanar reconstruction images generated by the same investigator.
Comparison of Injection Rates and Breath-holding Effects on Circulation Times
The first 40 patients in this series were randomly assigned to one of two equal-sized groups. Patients in each group were injected with contrast material at rates of 2 or 3 mL/sec for both the timing examination and the diagnostic study. The acquisition times for the two groups (23.4 seconds ± 4.9 and 23.7 seconds ± 4.8, respectively) were not statistically different (P = .6). In all 40 patients, the timing examinations were performed during free breathing. All of these patients were examined by using 4.2/1.7 and a 25°–40° flip angle.
Subsequently, in the last 20 patients, an injection rate of 2 mL/sec was used; however, the timing examinations were performed with breath holding at end expiration in a manner intended to mimic that which was used in the diagnostic study. The patients were given breathing instructions (ie, inspiration and expiration repeated twice) during the first 10 seconds of the contrast material injection and instructed to perform breath holding at end expiration for the next 20 seconds of the timing examination. All of the patients in the latter group of 20 patients were examined by using a newer and faster sequence (3.8/1.3 and 25° flip angle) with otherwise similar imaging parameters. The postprocessing and analysis of these images were performed as previously described. Compared with the acquisition times in the 40 studies performed with the original sequence, the acquisition times (21.4 seconds ± 4.5 vs 23.7 seconds ± 4.8) in the second group were shorter (P < .05).
Statistical Analyses
Statistical analyses were performed by using Excel software (Microsoft, Redmond, Wash). Correlation coefficients were calculated to determine the relationship between patient circulation time and patient age and weight. Comparisons of image quality measurements between the studies that had and those that did not have venous enhancement were made by using the two-tailed Student t test. Comparisons between the groups injected at the two contrast material injection rates and between the groups whose timing examinations were performed with and without breath holding also were made by using the two-tailed Student t test.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
|
|
There was no statistically significant difference in the quantitative imaging parameters of the MR angiographic images between the two groups of patients injected at 2 mL/sec or 3 mL/sec who underwent free breathing during the timing examination (P > .2 for all parameters) (Table 4). However, the images obtained in all 20 of the patients in the group injected at 2 mL/sec demonstrated an aorta-to-IVC contrast-to-noise ratio greater than 10 compared with 18 (90%) of 20 patients in the group injected at 3 mL/sec. Venous enhancement–free images were obtained in 18 (90%) of the 20 patients in the first group (injected at 2 mL/sec) compared with in 16 (80%) of the 20 patients in the second group (injected at 3 mL/sec). In two (10%) of 20 patients in the group injected at 3 mL/sec, the studies were considered to be nondiagnostic compared with in 0 of 20 patients in the group injected at 2 mL/sec.
|
Breath-hold Timing Examination
When compared by age and weight, the groups of patients who underwent timing examinations with and without breath holding (both groups injected at 2 mL/sec) were not significantly different (P > .2). The circulation times obtained from the timing examinations were not significantly different between the patients who performed breath holding at end expiration during the timing examination and those who breathed freely (22.1 seconds vs 21.0 seconds, respectively; P = .4). Other parameters related to the timing bolus also were not significantly different between the two groups.
The quality of images also was not different between the two groups (Fig 7). In both groups, 19 (95%) of 20 studies were considered to demonstrate excellent arterial enhancement, and in both groups, two (10%) of 20 studies demonstrated mild venous enhancement.
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Originally used in the 1930s as a bedside procedure to evaluate circulatory hemodynamics in patients with heart disease (23,24), measurements of patient circulation time experienced a renaissance in the 1980s with the development of dynamic contrast-enhanced computed tomography (CT). To establish injection protocols for maximum arterial enhancement, a test dose of 12–15 mL of iodinated contrast material has been found to be a reliable means of optimizing the timing of CT acquisitions (33–35). The CT literature is a source of extensive research on the dynamics of contrast material bolus geometry (36–40) and can serve as a useful reference for assessing MR angiography timing examinations.
Critical to the success of the timing examination is the degree to which the dynamics of the test dose mimic those of the contrast material dose in the diagnostic study. One important potential factor is that the rate of injection used in the timing examination should match that used in the MR angiographic study. Although some have argued that manual injection is a reproducible method for administering contrast material, others have demonstrated the superiority of automatic power injection over manual contrast material administration in studies that rely on timing examinations (22,32).
We also sought to investigate the effect of breath holding on circulatory hemodynamics. The 60-second duration of a timing examination is not practical for breath holding during the entire acquisition. Yet, in a diagnostic MR angiographic study, a breath-holding approach is preferred (12). We considered whether the differences in circulation times between the studies with breath holding and those with free breathing could account for the suboptimal MR angiograms. As a rough indicator of cardiac output, the circulation time, as measured by using CT, has been shown to vary with the pulse rate in humans (36) and with pharmacologically induced changes in cardiac output in an animal model (38). The effects of breath holding on cardiac output or pulse rate are complex and variable; furthermore, they depend on the degree to which patients perform Valsalva or Müller maneuvers while breath holding (41). By using echocardiography, Brenner and Waugh (42) observed variations in the left ventricular dimensions during the peak expiratory and peak inspiratory phases and showed a corresponding slight decrease (2.6%) in cardiac output from peak inspiration to peak expiration. These subtle changes may be below the limits of detectability with our timing examination approach, particularly given our sample sizes. Thus, we cannot exclude the possibility that, at least for a subpopulation of patients, the effects of breath holding may be substantial enough to alter the circulation time adversely in MR angiographic studies. The infrequency with which suboptimal or nondiagnostic studies are obtained in our practice makes this a difficult issue to study, although our investigation is continuing.
Several groups have developed alternative approaches to ensuring MR data acquisition during peak arterial enhancement. Korosec and colleagues (28) have developed an elegant time-resolved approach that can be used to acquire MR angiographic images every 2–8 seconds. By using this method, at least one acquisition is likely to be performed during peak arterial enhancement, thereby obviating a timing examination. However, because the time at which the contrast material bolus arrives in the aorta is not known a priori, patients are required to initiate breath holding with the first in the series of acquisitions. Therefore, in patients with long circulation times, this approach could require more prolonged breath holding.
Two other groups have approached the problem of timing MR image acquisitions by developing methods for detecting the arrival of the bolus in the aorta and triggering the data acquisition by using a threshold approach with automated (25,26) or manual (27) techniques. By using an automated approach, Prince et al (26) have reported that double doses of gadolinium-based contrast material provide degrees of arterial enhancement that are comparable to the degree of enhancement obtained with triple doses. Moreover, they showed that compared with nontimed examinations, studies with automated triggering resulted in less venous enhancement. The high number of trigger failures reported (12/62) reflected technical limitations that were not worked out at the time of publication. However, with use of an average dose of 0.22 mmol/kg, their results, with the exclusion of the technical failures, are comparable to those achieved in this study, in which an average dose of 0.11 mmol/kg was used.
Wilman and colleagues (27) described a fluoroscopically triggered approach that allows the operator to view in almost real time the passage of the contrast material bolus into the aorta, at which time the sequence acquisition can be manually triggered. By using this method, 24 (96%) of 25 studies were diagnostic, whereas 22 (88%) were considered to be of good quality. The centric view ordered approach used by the authors enables acquisition of the contrast-determining central lines of k space during periods of high vascular contrast. However, the fine detail of small vascular structures may be obscured or lost if the concentration of contrast material has diminished during acquisition of the peripheral lines of k space. By using doses of 0.18 mmol/kg of contrast material, the venous-to-arterial enhancement ratios reported with the fluoroscopically triggered method (IVC-to-aorta, 0.06 ± 0.05; left renal vein-to-left renal artery, 0.19 ± 0.18) are remarkably similar to those found in our study (Table 1). In the same study (27), the authors were able to demonstrate markedly improved arterial enhancement with less venous enhancement contamination by using the triggered sequence approach versus the nontimed approach.
Compared with the methods described above, the timing examination method can be easily implemented on most commercially available systems (22,29). The additional 4 minutes of total examination time required to perform the timing examination is comparable to that previously reported (22). In our experience, the approach is robust; it enabled the acquisition of diagnostic-quality images in 58 (97%) of 60 cases, even when a single dose of contrast material was used. In the rare instances (in two [3%] of 60 cases in the present study) in which nondiagnostic studies are obtained, the single-dose study can simply be repeated without concern about exceeding the dose allowances.
In this study, we confined our evaluation to single-dose gadolinium-based contrast studies and did not investigate whether higher or lower doses might yield improved image quality. Lentschig et al (43) recently demonstrated, by using similar MR angiographic sequences (with a 1.0-T system) in patients who had also undergone conventional angiography, that a dose of 0.1 mmol/kg is sufficient for assessment of the aorta and great vessels compared to doses of 0.2 and 0.3 mmol/kg. However, as they emphasized, similar conclusions with regard to the abdominal aorta and its branch vessels remain to be established.
To our knowledge, the literature on single-dose renal MR angiography is limited. By using a two-dimensional fast spoiled gradient-echo sequence to obtain 5-mm-thick sections, Tello et al (44) reported accurate classification of renal arterial stenosis in 50 (98%) of 51 arteries, although they noted that 10%–20% of the studies had venous enhancement. No timing scheme was used in their study, and although their accuracy rates were high compared with those of conventional angiography, the spatial resolution achieved with their two-dimensional approach was less than optimal.
Steffens et al (45) reported on a test dose approach in which 3 mL of gadolinium-based contrast material, followed by 17 mL in the diagnostic study, was used with an imaging sequence with sequential acquisition and parameters similar to those used in the present study. The authors, however, used a different formula to determine the imaging delay (Tdelay = Tcirc - Tacq/2), which typically would have resulted in a difference in imaging delay of 5 seconds compared with the delay in our study. This may explain at least in part the higher prevalence of suboptimal studies in their report. Of 50 studies, they observed optimal enhancement in 15 (30%) and good results in 16 (32%), but they observed partial venous enhancement in 16 (32%) and venous enhancement severe enough to obscure the arterial anatomy in two (4%) of the 50 studies. Despite these limitations, the authors reported a sensitivity of 96% and a specificity of 95% for the classification of renal arterial stenosis compared with the sensitivity and specificity reported with conventional angiography.
De Cobelli and colleagues (46) also recently compared single-dose gadolinium-enhanced 3D MR angiography (with 2 mL/sec injection) by using a timing examination with conventional angiography. Although their method for determining the imaging delay differed from ours (Tdelay = Tcirc - 1/3 Tacq), the resultant difference in delay values would have been only 1–2 seconds at most. The authors did not describe the occurrence of nondiagnostic studies, and the degree of arterial or venous enhancement was not detailed. As a result, a direct comparison with our results is not possible. However, by using a single dose of gadolinium-based contrast material and optimizing the study with a timing examination, the authors did find high agreement between gadolinium-enhanced MR angiography of the renal arteries and conventional angiography in the detection of hemodynamically significant stenoses (ie, >50% narrowing), with a sensitivity and specificity of 100% and 97%, respectively.
One limitation of this research is the lack of correlation with conventional angiography in most studies. Compared with that in recent studies (17,20,45,46), our protocol uses slightly faster sequences that can provide comparable spatial resolution in shorter imaging times. Thus, we hypothesize that the degree of accuracy of the MR angiograms obtained in the present study is at least comparable to that in previously described studies. Nonetheless, comparative studies are currently being evaluated.
In conclusion, by using a test bolus of 1 mL of gadolinium-based contrast material to achieve accurate timing, a single dose of gadolinium-based contrast material can be used to achieve preferential arterial enhancement relatively free of venous enhancement. This method helps to achieve results that are comparable to those reported by using other timing approaches (all performed at higher doses) without additional software or hardware demands. This robust technique facilitates rapid 3D MR angiographic evaluation of the renal arteries that can be performed in approximately 15 minutes, with the result of time savings and improved tolerability to patients.
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, V.S.L.; study concepts and design, V.S.L., N.M.R.; definition of intellectual content, V.S.L., N.M.R., G.A.K., J.C.W.; literature research, V.S.L., G.A.K., N.M.R.; clinical and experimental studies, V.S.L., D.H.S.; data acquisition, V.S.L., D.H.S.; data and statistical analyses, V.S.L.; manuscript preparation, V.S.L.; manuscript editing, V.S.L., N.M.R., G.A.K., J.C.W.; manuscript review, V.S.L., N.M.R., G.A.K., J.C.W., D.H.S.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Schmidt and R. Morgan Renal Contrast-enhanced MR Angiography: Timing Errors and Accurate Depiction of Renal Artery Origins Radiology, October 1, 2008; 249(1): 178 - 186. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Kramer, J. Wiskirchen, M. C. Fenchel, A. Seeger, G. Laub, G. Tepe, J. P. Finn, C. D. Claussen, and S. Miller Isotropic High-Spatial-Resolution Contrast-enhanced 3.0-T MR Angiography in Patients Suspected of Having Renal Artery Stenosis Radiology, April 1, 2008; 247(1): 228 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Michaely, H. Kramer, O. Dietrich, K. Nael, K.-P. Lodemann, M. F. Reiser, and S. O. Schoenberg Intraindividual Comparison of High-Spatial-Resolution Abdominal MR Angiography at 1.5 T and 3.0 T: Initial Experience Radiology, September 1, 2007; 244(3): 907 - 913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Chen, C. V. Quijano, V. M. Mai, S. K. Krishnamoorthy, W. Li, P. Storey, and R. R. Edelman On Improving Temporal and Spatial Resolution of 3D Contrast-enhanced Body MR Angiography with Parallel Imaging Radiology, June 1, 2004; 231(3): 893 - 899. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. T. Bae Peak Contrast Enhancement in CT and MR Angiography: When Does It Occur and Why? Pharmacokinetic Study in a Porcine Model Radiology, June 1, 2003; 227(3): 809 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Spuentrup, W. J. Manning, P. Bornert, K. V. Kissinger, R. M. Botnar, and M. Stuber Renal Arteries: Navigator-gated Balanced Fast Field-Echo Projection MR Angiography with Aortic Spin Labeling: Initial Experience Radiology, November 1, 2002; 225(2): 589 - 596. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Loewe, M. Schoder, T. Rand, U. Hoffmann, J. Sailer, T. Kos, J. Lammer, and S. Thurnher Peripheral Vascular Occlusive Disease: Evaluation with Contrast-Enhanced Moving-Bed MR Angiography Versus Digital Subtraction Angiography in 106 Patients Am. J. Roentgenol., October 1, 2002; 179(4): 1013 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T K Mittal, C Evans, T Perkins, and A M Wood Renal arteriography using gadolinium enhanced 3D MR angiography--clinical experience with the technique, its limitations and pitfalls Br. J. Radiol., June 1, 2001; 74(882): 495 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
