HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2231010858v1 223/2/509    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mallouhi, A. Articles by Jaschke, W. R. Search for Related Content PubMed PubMed Citation Articles by Mallouhi, A. Articles by Jaschke, W. R.

3D MR Angiography of Renal Arteries: Comparison of Volume Rendering and Maximum Intensity Projection Algorithms¹

¹ From the Department of Radiology, Innsbruck University Hospital, Anichstrasse 35, 6020 Innsbruck, Austria. From the 2000 RSNA scientific assembly. Received April 30, 2001; revision requested May 25; revision received August 9; accepted September 28. Address correspondence to A.M. (e-mail: ammar.mallouhi@uibk.ac.at).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To compare volume rendering (VR) and maximum intensity projection (MIP) as postprocessing techniques of magnetic resonance (MR) angiography for detection and quantification of renal artery stenosis.

MATERIALS AND METHODS: Twenty-seven patients underwent three-dimensional contrast material–enhanced MR angiography of the renal arteries with a 1.5-T imager. For each renal artery, targeted MIP and VR images were reconstructed in oblique coronal and transverse orientations. For each modality, image generation and evaluation were performed interactively by two independent radiologists blinded to angiographic results. In comparison with digital subtraction angiography (DSA) findings, stenosis quantification and detection by using MIP and VR were evaluated with the use of 50% and 70% cutoff points by using linear regression analysis and 2 x 2 tables. Overall image quality and vascular delineation on MIP and VR images were also compared.

RESULTS: All main and accessory renal arteries depicted at DSA were also demonstrated on MIP and VR images. VR performed slightly better than MIP for quantification of stenoses greater than 50% (VR: r² = 0.84, P < .001; MIP: r² = 0.38, P = .001) and significantly better for severe stenoses (VR: r² = 0.83, P < .001; MIP: r² = 0.21, P = .1). For detection of stenosis, VR yielded a substantial improvement in positive predictive value (VR: 95% and 90%; MIP: 86% and 68% for stenoses greater than 50% and 70%, respectively). Image quality obtained with VR was not significantly better than that with MIP; however, vascular delineation on VR images was significantly better.

CONCLUSION: The VR technique of renal MR angiography enabled more accurate detection and quantification of renal artery stenosis than did MIP, with significantly improved vascular delineation.

© RSNA, 2002

Index terms: Magnetic resonance (MR), vascular studies, 961.12942 • Renal arteries, MR, 961.12942, 961.12949 • Renal arteries, stenosis or obstruction, 961.721

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Renovascular disease has been recognized as an important cause of progressive renal insufficiency (1) and is believed to account for refractory hypertension in about 1% of patients with hypertension (2). Because renal artery stenosis is potentially curable with endovascular procedures (3–5), detection and precise quantification of renal artery stenosis are important for the purpose of treatment planning. Digital subtraction angiography (DSA) is currently the standard method used for the evaluation of renal arterial disease.

Since its introduction by Prince et al (6,7), three-dimensional (3D) contrast material–enhanced magnetic resonance (MR) angiography has been established as a safe, nonionizing, and noninvasive method for evaluating renal artery stenosis. Advances in gradient performance and improvements in MR angiographic sequence design have enabled high-speed imaging with sufficient spatial resolution and substantially improved image quality (8–14). Volumetric data acquired by using MR angiography permitted the arbitrary visualization of the oblique course of the renal arteries and thus allowed evaluation of stenosis that was unattainable with projection techniques such as conventional angiography (15).

However, since source images are subject to partial volume effects (16), 3D reconstructions were performed to enhance diagnostic confidence. The most widely used postprocessing technique for MR angiography is maximum intensity projection (MIP) (16–21). The diagnostic accuracy of contrast-enhanced MR angiography complemented with MIP images has been evaluated in studies (21–24) in which sensitivities and specificities of more than 90% for detection of renal artery stenosis greater than 50% were reported.

The volume rendering (VR) algorithm has been implemented successfully as an interactive 3D reconstruction technique for renal CT angiography, which enabled a higher specificity and more comprehensive evaluation of renal artery stenosis than did MIP (25). The application of VR in the evaluation of MR angiographic examinations remains limited, and clinical validation is lacking for the accuracy of this algorithm in the grading of vascular stenoses, although a number of study findings (26–28) have described the technique.

The purpose of this study was to compare MR angiography with VR and MIP algorithms with DSA, the reference standard, in the detection and quantification of renal artery stenosis and to compare both rendering algorithms for the estimation of image quality and vascular delineation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
This study was based on retrospective analysis of contrast-enhanced 3D MR angiographic examinations of the renal arteries. To determine the role of MIP and VR algorithms in the detection and quantification of renal artery stenosis, we retrieved digital source images of 27 consecutive patients examined between June 1999 and December 2000. Examinations that met the following criteria were included in the study: (a) a technically acceptable renal MR angiographic examination that was (i) acquired during strict expiratory apnea with (ii) a degree of contrast enhancement coded as "excellent" (high degree of vascular opacification, 21 patients) or "good" (degree of contrast enhancement was not high but was sufficient for the analysis of the renal arteries, six patients); and (b) MR angiography performed within 3 months before DSA that included selective renal arteriograms (22 patients), or an abdominal overview with sufficient visualization of the entire artery, including the ostia, without image degradation by overlapping of the superior mesenteric artery or bowel-gas artifacts (five patients). All examinations were performed to rule out renal artery stenosis in hypertensive patients, including one patient with a renal transplant. All patients had provided written informed consent for MR angiography and DSA. Approval from our institutional review board was not required.

DSA Technique and Image Analysis
Flush abdominal aortography was performed with an intraarterial injection of 36 mL of a nonionic contrast material at 12 mL/sec by using a 4-F pigtail catheter (Cordis, Johnson and Johnson, Roden, the Netherlands), followed by selective renal arteriography in 22 patients by using a 4-F cobra or 4-F Sidewinder I catheter (Cordis, Johnson and Johnson) with injection volumes (8–10 mL) and rates (3–5 mL/sec) varying with vessel size. Images were obtained in 10°–15° left and right anterior oblique projections. Pressures were also measured in the renal artery and the abdominal aorta to judge the significance of stenosis.

The DSA images were evaluated independently by two senior interventional radiologists (A.D., P.W.) who were blinded to MR angiographic results and graded each vessel, by using digital calipers, on the basis of percentage diameter reduction (0%–100%) at the point of maximum narrowing, as compared with the nearest distal normal segment. When the latter was absent, the comparison was made to the nearest proximal normal segment. Since DSA images are two dimensional, and discordance between transverse MIP and VR images might not be resolved by using a coronal DSA image, DSA reviewers subjectively used blood pressure gradients in their assessment that were available in 22 patients, to enhance their diagnostic confidence. A pressure gradient of less than 20 mm Hg was use to rule out a significant stenosis.

MR Imaging Technique
All MR examinations were performed by using a 1.5-T system (Magnetom Vision; Siemens Medical systems, Erlangen, Germany) by using a phased-array surface coil. The transit time of contrast media was determined by using a test bolus of 2 mL of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany). Contrast-enhanced MR angiography was performed after intravenous injection of a dose of 0.3 mL per kilogram of body weight of gadopentetate dimeglumine with a flow rate of 3 mL/sec. With the breath-hold technique, MR angiography was performed by using a coronal booster, T1-weighted fast low-angle shot (or FLASH) sequence (3.2/1.2 [repetition time msec/echo time msec]; flip angle, 30°; matrix, 256 x 168; section thickness, 1.6 mm; field of view, 340 mm; acquisition time, 12 seconds) in 17 patients. In 10 patients, a coronal high-resolution T1-weighted FLASH sequence (4.6/1.8; flip angle, 30°; matrix, 512 x 256; section thickness, 1.6 mm; field of view, 380 mm; acquisition time, 22 seconds) was performed with breath-hold technique. The MR angiographic sequence was repeated during venous and equilibrium phases, with separate breath holds. To reduce the signal of background tissue, a 3D data set was obtained prior to contrast material administration, to be subtracted from the arterial phase.

Reconstruction Parameters
For quantitative and semiquantitative image analysis, nonenhanced and arterial phase data sets were transferred to an independent workstation (Sun Ultra 60; Sun Microsystems, Mountain View, Calif), and software (Advantage Windows version 4.0; GE Medical Systems, Milwaukee, Wis) was used. Reconstructions were performed according to a standardized protocol. At the first step of reconstruction, the nonenhanced data set was subtracted from that of the arterial phase. Targeted subvolume MIP images were obtained interactively in coronal and transverse projections selected to encompass the renal artery from the ostium to the hilum. To standardize the volume of the targeted MIP images, a slab with a similar width (2 cm) was positioned on the multiplanar volume reformation images in all cases. The angle of the subvolume slab was adjusted to encompass the aorta, one renal artery, and the central portion of the kidney. The right and left renal arteries were analyzed separately. In cases of severe kinking of the renal artery, overlay of the enhanced renal vein or renal pelvis, or increased paravascular noise, a thinner slab of 5 mm was applied for the analysis of the stenotic segment. The MIP images were acquired interactively by two independent radiologists (M.S., C.W.) experienced in MR angiography.

VR images were created in transverse and coronal orientations to demonstrate each renal artery from the ostium to the kidney hilum. Optimal display parameters (window width and level, opacity, and brightness) for generating VR images of the renal arteries were determined in a preclinical study in which models of VR images from MR angiographic data sets were tested. The parameters were defined by means of consensus agreement between two of the authors (A.M., B.V.C.) familiar with the VR software. The opacity transfer function was determined to maximize the visualization of the renal arteries and most of all, the perception of the stenotic segment by using a low-to-high opacity curve type, the slope of which ranged from 40 to 180 units. The lower threshold value of the opacity curve was subjectively adjusted to represent the intensity of the renal arteries. Opacity and brightness values of 100% were assigned to the selected material. In cases with increased enhancement of the overlapping renal vein or renal pelvis or increased paravascular noise, editing procedures were performed by filtering out or cutting off the overlying structures. The time required to display a VR image of the renal arteries was approximately 1 minute. Each modification of parameter or any change in volume orientation took another 1–3 seconds. VR imaging was performed interactively by two independent radiologists (A.M., B.V.C.) experienced in vascular imaging and 3D reconstruction.

Image Analysis
Quantitative analysis.—By using digital calipers, stenosis was measured on coronal and transverse images at the point of maximum narrowing in relation to the diameter of the nearest distal, or, if not possible, to the nearest proximal renal artery segment considered to be free of disease. The overall stenosis percentage was assessed and compared with that obtained by using DSA. No measurement was performed when the renal artery appeared to be normal or occluded. The mean stenosis severity was separately determined for each renal artery from the values measured by the two reviewers of each modality.

Semiquantitative analysis.—For further assessment of MIP and VR reconstructions, coronal and transverse views were analyzed according to an ordinal four-point scale. General criteria included the impression of overall image quality assessed according to the following: 1, very good vascular visibility enabling detailed and reliable evaluation; 2, good vascular visibility enabling adequate evaluation; 3, average visibility with compromised evaluation; and 4, unsatisfactory, barely visible renal artery with inadequate evaluation. In addition, the delineation of vascular lumen was scored as follows: 1, well defined; 2, moderately defined but with definite demonstration and quantification of stenosis; 3, vaguely defined with demonstration of stenosis, but uncertainty of quantification of stenosis; or 4, lumen of stenotic segment not identified.

MIP images were evaluated in conjunction with source images, whereas the latter were not involved in the interpretation of VR views. Coronal and transverse MIP and VR reconstructions were independently and blindly interpreted by the reviewing radiologist (for MIP images, M.S., C.W.; and for VR images, A.M., B.V.C.).

Statistical Analysis
Data entry procedures and statistical analysis were performed with a statistical software system (SPSS version 10.0.0 for Windows; SPSS, Chicago, Ill). In the first step, the sensitivity, specificity, and accuracy of both MR angiographic techniques in detection of renal artery stenosis were calculated on a per-artery (ie, the ability to correctly identify all stenotic arteries) basis, by using a cutoff of 50% and 70% stenosis to evaluate the detection ability for various grades of stenosis. In the second step of analysis, linear regression analysis was performed to investigate the correlation between the percentage of stenosis obtained by using MIP and VR and that measured by using DSA. The coefficient of determination (r²) and P value of the correlation were determined with the analysis. Finally, we evaluated semiquantitative data from image quality and delineation ability mode scores to assess the performance of MIP and VR on coronal and transverse images by using the Wilcoxon signed rank test. P values lower than .05 were considered to indicate a significant difference.

For comparison of observer performance for stenosis severity measurements, without consensus, the limits of agreement method (29) was used. With this method, the arithmetic difference between the two reviewers was plotted against their mean. The statistic was used for the assessment of image quality and vascular delineation. Interobserver agreement was considered as slight ( 0.20), fair ( = 0.21–0.40), moderate ( = 0.41–0.60), substantial ( = 0.61–0.80), or almost perfect, ( = 0.81–1.00) (30).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Fifty-three renal arteries in 27 patients were identified at DSA, including 46 main and seven accessory renal arteries. They were categorized as shown in Table 1. Six accessory renal arteries were normal, and one had a mild stenosis (<50%). All renal arteries found at DSA were identified by both MIP and VR reviewers. No additional renal arteries were detected by using both MR angiography algorithms that were not detected at DSA.

Quantitative Image Analysis
Comparisons of MIP and VR algorithms with DSA are summarized in Figure 1 (stenosis quantification) and Table 2 (stenosis detection). Linear regression analysis for all renal arteries revealed a statistically significant correlation between mean stenosis severity measured at DSA and that assessed by using MIP and VR techniques (r² = 0.83, P < .001 and r² = 0.96, P < .001, respectively).

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. Graph shows linear regression analysis of mean percentages of renal artery stenosis, as measured by using MIP (x, dashed line) and VR (, solid line) plotted against DSA. Dotted line indicates perfect agreement of MR angiography and DSA. (a) Graph shows a positive and high correlation with DSA for MIP (r2 = 0.83, P < .001) and VR (r2 = 0.96, P < .001) when all arteries were considered. (b) By using a cutoff of 50% stenosis, correlation between MIP and DSA decreases substantially (r2 = 0.38, P = .001), with no considerable change in the relationship between VR and DSA (r2 = 0.84, P < .001). (c) By using a cutoff of 70% stenosis, graph shows there is no statistically significant correlation between MIP and DSA (r2 = 0.21, P = .1), whereas a significant relationship (r2 = 0.83, P < .001) between VR and DSA was maintained.

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. Graph shows linear regression analysis of mean percentages of renal artery stenosis, as measured by using MIP (x, dashed line) and VR (, solid line) plotted against DSA. Dotted line indicates perfect agreement of MR angiography and DSA. (a) Graph shows a positive and high correlation with DSA for MIP (r2 = 0.83, P < .001) and VR (r2 = 0.96, P < .001) when all arteries were considered. (b) By using a cutoff of 50% stenosis, correlation between MIP and DSA decreases substantially (r2 = 0.38, P = .001), with no considerable change in the relationship between VR and DSA (r2 = 0.84, P < .001). (c) By using a cutoff of 70% stenosis, graph shows there is no statistically significant correlation between MIP and DSA (r2 = 0.21, P = .1), whereas a significant relationship (r2 = 0.83, P < .001) between VR and DSA was maintained.

View larger version (24K): [in this window] [in a new window] [Download PPT slide]   Figure 1c. Graph shows linear regression analysis of mean percentages of renal artery stenosis, as measured by using MIP (x, dashed line) and VR (, solid line) plotted against DSA. Dotted line indicates perfect agreement of MR angiography and DSA. (a) Graph shows a positive and high correlation with DSA for MIP (r2 = 0.83, P < .001) and VR (r2 = 0.96, P < .001) when all arteries were considered. (b) By using a cutoff of 50% stenosis, correlation between MIP and DSA decreases substantially (r2 = 0.38, P = .001), with no considerable change in the relationship between VR and DSA (r2 = 0.84, P < .001). (c) By using a cutoff of 70% stenosis, graph shows there is no statistically significant correlation between MIP and DSA (r2 = 0.21, P = .1), whereas a significant relationship (r2 = 0.83, P < .001) between VR and DSA was maintained.

View larger version (142K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 57-year-old woman with hypertension. (a) Coronal oblique subvolume MIP image shows a truncal mild (40%) stenosis (arrow) of the left main renal artery. (b) Transverse oblique subvolume MIP image shows a truncal moderate (60%) stenosis (arrow) of the left main renal artery. Combination of findings on both MIP images indicated a moderate (50%) stenosis. (c) Coronal oblique anteroposterior view on VR image shows a normal left main renal artery (arrow) and a normal lower pole left accessory artery (arrowhead). (d) transverse oblique VR image depicts a mild (10%) stenosis (arrow) of the renal artery. Combination of findings on both VR images indicated a normal renal artery. (e) Left oblique intraarterial DSA image shows no stenosis of the left main and accessory renal arteries. Intraarterial blood pressure measurements revealed no gradient between the aorta and renal artery and thus ruled out the moderate stenosis depicted in b.

View larger version (111K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 57-year-old woman with hypertension. (a) Coronal oblique subvolume MIP image shows a truncal mild (40%) stenosis (arrow) of the left main renal artery. (b) Transverse oblique subvolume MIP image shows a truncal moderate (60%) stenosis (arrow) of the left main renal artery. Combination of findings on both MIP images indicated a moderate (50%) stenosis. (c) Coronal oblique anteroposterior view on VR image shows a normal left main renal artery (arrow) and a normal lower pole left accessory artery (arrowhead). (d) transverse oblique VR image depicts a mild (10%) stenosis (arrow) of the renal artery. Combination of findings on both VR images indicated a normal renal artery. (e) Left oblique intraarterial DSA image shows no stenosis of the left main and accessory renal arteries. Intraarterial blood pressure measurements revealed no gradient between the aorta and renal artery and thus ruled out the moderate stenosis depicted in b.

View larger version (156K): [in this window] [in a new window] [Download PPT slide]   Figure 2c. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 57-year-old woman with hypertension. (a) Coronal oblique subvolume MIP image shows a truncal mild (40%) stenosis (arrow) of the left main renal artery. (b) Transverse oblique subvolume MIP image shows a truncal moderate (60%) stenosis (arrow) of the left main renal artery. Combination of findings on both MIP images indicated a moderate (50%) stenosis. (c) Coronal oblique anteroposterior view on VR image shows a normal left main renal artery (arrow) and a normal lower pole left accessory artery (arrowhead). (d) transverse oblique VR image depicts a mild (10%) stenosis (arrow) of the renal artery. Combination of findings on both VR images indicated a normal renal artery. (e) Left oblique intraarterial DSA image shows no stenosis of the left main and accessory renal arteries. Intraarterial blood pressure measurements revealed no gradient between the aorta and renal artery and thus ruled out the moderate stenosis depicted in b.

View larger version (90K): [in this window] [in a new window] [Download PPT slide]   Figure 2d. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 57-year-old woman with hypertension. (a) Coronal oblique subvolume MIP image shows a truncal mild (40%) stenosis (arrow) of the left main renal artery. (b) Transverse oblique subvolume MIP image shows a truncal moderate (60%) stenosis (arrow) of the left main renal artery. Combination of findings on both MIP images indicated a moderate (50%) stenosis. (c) Coronal oblique anteroposterior view on VR image shows a normal left main renal artery (arrow) and a normal lower pole left accessory artery (arrowhead). (d) transverse oblique VR image depicts a mild (10%) stenosis (arrow) of the renal artery. Combination of findings on both VR images indicated a normal renal artery. (e) Left oblique intraarterial DSA image shows no stenosis of the left main and accessory renal arteries. Intraarterial blood pressure measurements revealed no gradient between the aorta and renal artery and thus ruled out the moderate stenosis depicted in b.

View larger version (164K): [in this window] [in a new window] [Download PPT slide]   Figure 2e. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 57-year-old woman with hypertension. (a) Coronal oblique subvolume MIP image shows a truncal mild (40%) stenosis (arrow) of the left main renal artery. (b) Transverse oblique subvolume MIP image shows a truncal moderate (60%) stenosis (arrow) of the left main renal artery. Combination of findings on both MIP images indicated a moderate (50%) stenosis. (c) Coronal oblique anteroposterior view on VR image shows a normal left main renal artery (arrow) and a normal lower pole left accessory artery (arrowhead). (d) transverse oblique VR image depicts a mild (10%) stenosis (arrow) of the renal artery. Combination of findings on both VR images indicated a normal renal artery. (e) Left oblique intraarterial DSA image shows no stenosis of the left main and accessory renal arteries. Intraarterial blood pressure measurements revealed no gradient between the aorta and renal artery and thus ruled out the moderate stenosis depicted in b.

View larger version (174K): [in this window] [in a new window] [Download PPT slide]   Figure 3a. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (4.6/1.8, 30° flip angle) in a 79-year-old woman with bilateral renal artery stenosis. (a) Coronal oblique anteroposterior subvolume MIP image shows a severe (80%) truncal stenosis (arrow) of the left main renal artery. Combining findings on transverse MIP image (not shown), which revealed an 80% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 80%. (b) Coronal oblique posteroanterior VR image depicts a moderate (50%) truncal stenosis (arrow) of the left renal artery. Combining findings on transverse VR image (not shown), which revealed a 70% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 60%. The delineation of the stenotic segment on the VR view is superior to that on the MIP image. (c) Anteroposterior intraarterial DSA image confirms the stenosis severity measured on VR view and shows a small lower pole left accessory artery (arrowheads). (d) Coronal oblique subvolume MIP and (e) anteroposterior VR images depict the accessory left renal artery (arrowheads). Because of interactive reconstructions, evaluation of the main, accessory, and segmental left renal arteries on VR views was not substantially hampered by the considerable enhancement of the left renal vein.

View larger version (137K): [in this window] [in a new window] [Download PPT slide]   Figure 3b. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (4.6/1.8, 30° flip angle) in a 79-year-old woman with bilateral renal artery stenosis. (a) Coronal oblique anteroposterior subvolume MIP image shows a severe (80%) truncal stenosis (arrow) of the left main renal artery. Combining findings on transverse MIP image (not shown), which revealed an 80% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 80%. (b) Coronal oblique posteroanterior VR image depicts a moderate (50%) truncal stenosis (arrow) of the left renal artery. Combining findings on transverse VR image (not shown), which revealed a 70% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 60%. The delineation of the stenotic segment on the VR view is superior to that on the MIP image. (c) Anteroposterior intraarterial DSA image confirms the stenosis severity measured on VR view and shows a small lower pole left accessory artery (arrowheads). (d) Coronal oblique subvolume MIP and (e) anteroposterior VR images depict the accessory left renal artery (arrowheads). Because of interactive reconstructions, evaluation of the main, accessory, and segmental left renal arteries on VR views was not substantially hampered by the considerable enhancement of the left renal vein.

View larger version (162K): [in this window] [in a new window] [Download PPT slide]   Figure 3c. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (4.6/1.8, 30° flip angle) in a 79-year-old woman with bilateral renal artery stenosis. (a) Coronal oblique anteroposterior subvolume MIP image shows a severe (80%) truncal stenosis (arrow) of the left main renal artery. Combining findings on transverse MIP image (not shown), which revealed an 80% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 80%. (b) Coronal oblique posteroanterior VR image depicts a moderate (50%) truncal stenosis (arrow) of the left renal artery. Combining findings on transverse VR image (not shown), which revealed a 70% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 60%. The delineation of the stenotic segment on the VR view is superior to that on the MIP image. (c) Anteroposterior intraarterial DSA image confirms the stenosis severity measured on VR view and shows a small lower pole left accessory artery (arrowheads). (d) Coronal oblique subvolume MIP and (e) anteroposterior VR images depict the accessory left renal artery (arrowheads). Because of interactive reconstructions, evaluation of the main, accessory, and segmental left renal arteries on VR views was not substantially hampered by the considerable enhancement of the left renal vein.

View larger version (164K): [in this window] [in a new window] [Download PPT slide]   Figure 3d. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (4.6/1.8, 30° flip angle) in a 79-year-old woman with bilateral renal artery stenosis. (a) Coronal oblique anteroposterior subvolume MIP image shows a severe (80%) truncal stenosis (arrow) of the left main renal artery. Combining findings on transverse MIP image (not shown), which revealed an 80% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 80%. (b) Coronal oblique posteroanterior VR image depicts a moderate (50%) truncal stenosis (arrow) of the left renal artery. Combining findings on transverse VR image (not shown), which revealed a 70% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 60%. The delineation of the stenotic segment on the VR view is superior to that on the MIP image. (c) Anteroposterior intraarterial DSA image confirms the stenosis severity measured on VR view and shows a small lower pole left accessory artery (arrowheads). (d) Coronal oblique subvolume MIP and (e) anteroposterior VR images depict the accessory left renal artery (arrowheads). Because of interactive reconstructions, evaluation of the main, accessory, and segmental left renal arteries on VR views was not substantially hampered by the considerable enhancement of the left renal vein.

View larger version (140K): [in this window] [in a new window] [Download PPT slide]   Figure 3e. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (4.6/1.8, 30° flip angle) in a 79-year-old woman with bilateral renal artery stenosis. (a) Coronal oblique anteroposterior subvolume MIP image shows a severe (80%) truncal stenosis (arrow) of the left main renal artery. Combining findings on transverse MIP image (not shown), which revealed an 80% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 80%. (b) Coronal oblique posteroanterior VR image depicts a moderate (50%) truncal stenosis (arrow) of the left renal artery. Combining findings on transverse VR image (not shown), which revealed a 70% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 60%. The delineation of the stenotic segment on the VR view is superior to that on the MIP image. (c) Anteroposterior intraarterial DSA image confirms the stenosis severity measured on VR view and shows a small lower pole left accessory artery (arrowheads). (d) Coronal oblique subvolume MIP and (e) anteroposterior VR images depict the accessory left renal artery (arrowheads). Because of interactive reconstructions, evaluation of the main, accessory, and segmental left renal arteries on VR views was not substantially hampered by the considerable enhancement of the left renal vein.

View larger version (138K): [in this window] [in a new window] [Download PPT slide]   Figure 4a. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 50-year-old woman with bilateral renal artery stenosis and atrophied right kidney. (a) Coronal oblique subvolume MIP image shows a severe (95%) ostial stenosis (solid arrow) of the right renal artery with a poststenotic dilatation. The stenotic segment simulates occlusion. Combining findings on transverse MIP image (not shown), which revealed a 95% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 95%. (b) Coronal oblique VR view depicts, in contrast to the MIP image, the fine residual lumen of the right renal artery severe (90%) stenosis (solid arrow). Combining findings on transverse VR image (not shown), which revealed a 90% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 90%. MIP and VR images depict a mild proximal stenosis of the left renal artery (open arrow) confirmed at DSA (image not shown). (c) Right oblique selective DSA image confirms the stenosis severity.

View larger version (110K): [in this window] [in a new window] [Download PPT slide]   Figure 4b. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 50-year-old woman with bilateral renal artery stenosis and atrophied right kidney. (a) Coronal oblique subvolume MIP image shows a severe (95%) ostial stenosis (solid arrow) of the right renal artery with a poststenotic dilatation. The stenotic segment simulates occlusion. Combining findings on transverse MIP image (not shown), which revealed a 95% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 95%. (b) Coronal oblique VR view depicts, in contrast to the MIP image, the fine residual lumen of the right renal artery severe (90%) stenosis (solid arrow). Combining findings on transverse VR image (not shown), which revealed a 90% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 90%. MIP and VR images depict a mild proximal stenosis of the left renal artery (open arrow) confirmed at DSA (image not shown). (c) Right oblique selective DSA image confirms the stenosis severity.

View larger version (159K): [in this window] [in a new window] [Download PPT slide]   Figure 4c. MIP and VR images from 3D MR angiography after administration of gadopentetate dimeglumine (3.2/1.2, 30° flip angle) in a 50-year-old woman with bilateral renal artery stenosis and atrophied right kidney. (a) Coronal oblique subvolume MIP image shows a severe (95%) ostial stenosis (solid arrow) of the right renal artery with a poststenotic dilatation. The stenotic segment simulates occlusion. Combining findings on transverse MIP image (not shown), which revealed a 95% stenosis, with findings on coronal MIP image resulted in an overall stenosis severity of 95%. (b) Coronal oblique VR view depicts, in contrast to the MIP image, the fine residual lumen of the right renal artery severe (90%) stenosis (solid arrow). Combining findings on transverse VR image (not shown), which revealed a 90% stenosis, with findings on coronal VR image resulted in an overall stenosis severity of 90%. MIP and VR images depict a mild proximal stenosis of the left renal artery (open arrow) confirmed at DSA (image not shown). (c) Right oblique selective DSA image confirms the stenosis severity.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this study, two postprocessing algorithms, the routinely used MIP and VR, of renal 3D contrast-enhanced MR angiography were compared. Both have comparable ability in interactive generation of 3D and reproducible reconstructions of the vasculature but have substantially different abilities in conveying the voxel components of the source images and in the number of voxels used in the final image.

Implementation of the VR algorithm on MR angiographic data sets contributed to the improved perceptibility and, consequently, to the precision in the evaluation of renal arterial lumina. The principle, as reported in previous articles (25,31,32), is that VR is based on the percentage classification technique, which is used to estimate the probability of a material being homogeneously present in a voxel. This method provides accurate determination of the amounts of materials when the voxel consists of two or more materials, which are volume averaged. The VR algorithm enables the volume-averaged voxels to be included in the final image because it calculates a weighted sum of data from all voxels along a ray projected through the data set. Accordingly, VR tends to maximize the potential of accurate elucidation, with better delineation of small-caliber vessels or stenotic segments of vascular structures.

As opposed to VR, the MIP algorithm selects only the voxel with the highest attenuation along a ray projected through the data set. Thus, volume-averaged voxels may be erroneously excluded from the final image, resulting in overestimation of stenosis. However, including the marginal volume-averaged voxels on the final MIP image results in vessel blurring (decreased delineation ability). Because of the limited spatial resolution of MR angiography, the minimal misclassification of the marginal voxels between vascular and perivascular signal intensities in an MR angiography data set, particularly in severely stenotic arteries, may lead to morphologic miscategorization of the degree of stenosis (16) and degradation of vascular conspicuity.

Our results showed that MR angiography with a VR technique enabled a more accurate quantification of renal artery stenosis inferred from the high correlation of findings with DSA findings. Although the majority of moderate and severely stenotic arteries were correctly classified on MIP and VR images, VR yielded a significant advantage in quantifying severe stenoses. Precise quantification of significant renal artery stenosis is important especially in patients with hypertension, since the application of interventional procedures has been found to be effective in restoring blood flow, improving blood pressure, and preventing progression of end-stage renal function (3–5). The degree of stenosis that necessitates intervention was indicated in the findings of several studies to be greater than 50% (5,33), 60% (4,34), or greater than 70% (1,35–37). At our institution, we agree with the later definition and consider that stenoses greater than 70% represent the critical stenosis level that reduces the blood flow and poststenotic perfusion pressure during reductions in systemic arterial pressure and hence, pose an increased risk of developing into an occlusion.

With regard to detection of moderate and severe stenosis, VR yielded a similar sensitivity to that of MIP. However, the specificity and accuracy of VR were found to be considerably higher, owing to a substantial increase in the positive predictive value. The improvement of positive predictive value indicates that VR enables a more reliable detection of clinically significant renal arterial stenoses than does MIP technique and therefore helps in selection of candidates for vascular interventions.

As another advantage over MIP, we observed a trend toward improved delineation of the renal artery and particularly of the stenotic segment with VR. The well-defined vasculature on VR images contributed to the accuracy of the assessment of the residual lumen in all cases.

Successful implementation of the VR algorithm, however, is predicated on the correct choice of display parameters (window width and level, opacity and brightness), which remains a crucial subjective decision. Since renal MR angiography data sets comprise only two signal intensities (ie, hyperintensive vascular structures and hypointensive perivascular structures), VR of MR angiography data involve substantially less processing time and easier application of transfer function. Generation of a reliable VR angiogram can be achieved, in our experience, by interactive modification of the window level so that it represents the mean intensity of the renal artery. Because it is difficult to reliably evaluate the intensity of severe stenotic arterial segments, we used a decreased window level to display the fine low-intensity residual lumen. In fact, the residual lumen was depicted in all cases of severe stenosis on VR images as compared with five of 10 severe stenoses on MIP images. These clinical findings confirm that the potential of correctly presenting an enhanced renal arterial lumen is higher with VR than with the MIP algorithm.

At the present time, MIP algorithm is widely used as a postprocessing technique of MR angiography to evaluate the renal arteries. Investigators in studies (21–24) found sensitivities between 93% and 100% and specificities between 71% and 90% in the detection of at least 50% renal artery stenosis with MR angiography complemented with MIP. The results of our study were in accordance with those in previous reports and confirmed the high sensitivity (95%) and specificity (91%) of MIP algorithm for identifying at least 50% stenoses. By virtue of the interactive generation of subvolume MIP views separately for each renal artery in two parallel orientations, we also found a high sensitivity (95%) and specificity (90%) for detecting stenoses of at least 70% diameter reduction.

In the assessment of factors that help to determine the hemodynamic significance of a renal artery stenosis by using MR angiography, such as poststenotic dilatation, delayed renal enhancement, and reduced renal parenchymal mass, we found no substantial differences between VR and MIP. Although the difference between MIP and VR regarding overall image quality was not significant, it is interesting to note that the maintenance of 3D appearance combined with the well-defined vasculature on VR images enabled a comprehensive illustration of the vascular interrelationships.

In summary, we conclude that VR as a postprocessing technique of renal MR angiographic data sets has three advantages over MIP: a higher positive predictive value, despite the application of two stenosis cutoff points; a better correlation with DSA, irrespective of stenosis severity, and an improved delineation of the renal artery. This could have a considerable clinical effect because VR faithfully conveys the original data acquired by using renal MR angiography. Hence, it supports the noninvasive role of MR angiography in evaluating patients who have suspected renovascular disease.

	FOOTNOTES

 
Abbreviations: DSA = digital subtraction angiography, MIP = maximum intensity projection, 3D = three dimensional, VR = volume rendering

Author contributions: Guarantors of integrity of entire study, A.M., W.R.J.; study concepts and design, A.M., W.J., W.R.J.; literature research, A.M.; experimental studies, A.M., B.V.C.; data acquisition, A.M., M.S., C.W., B.V.C.; data analysis/interpretation, A.M., M.S., C.W., A.D., B.V.C., P.W.; statistical analysis, A.M.; manuscript preparation, A.M.; manuscript definition of intellectual content, A.M., W.J., W.R.J.; manuscript editing and revision/review, A.M., M.S., W.J., W.R.J.; manuscript final version approval, all authors.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Textor SC. Atherosclerotic renovascular disease as a cause of end-stage renal disease: cost considerations. Blood Purif 1996; 14:305-314.
2. Hillman BJ. Imaging advances in the diagnosis of renovascular hypertension. AJR Am J Roentgenol 1989; 153:5-14.
3. Leertouwer TC, Gussenhoven EJ, Bosch JL, et al. Stent placement for renal arterial stenosis: where do we stand? A meta-analysis. Radiology 2000; 216:78-85.
4. Baumgartner I, von Aesch K, Do DD, Triller J, Birrer M, Mahler F. Stent placement in ostial and nonostial atherosclerotic renal arterial stenoses: a prospective follow-up study. Radiology 2000; 216:498-505.
5. Blum U, Krumme B, Flugel P, et al. Treatment of ostial renal-artery stenoses with vascular endoprostheses after unsuccessful balloon angioplasty. N Engl J Med 1997; 336:459-465.
6. Prince MR, Yucel EK, Kaufman JA, Harrison DC, Geller SC. Dynamic gadolinium-enhanced three-dimensional abdominal MR arteriography. J Magn Reson Imaging 1993; 3:877-881.
7. Prince MR. Gadolinium-enhanced MR aortography. Radiology 1994; 191:155-164.
8. Prince MR. Contrast-enhanced MR angiography: theory and optimization. Magn Reson Imaging Clin N Am 1998; 6:257-267.
9. Schoenberg SO, Prince MR, Knopp MV, Allenberg JR. Renal MR angiography. Magn Reson Imaging Clin N Am 1998; 6:351-370.
10. Fain SB, King BF, Breen JF, Kruger DG, Riederer SJ. High-spatial-resolution contrast-enhanced MR angiography of the renal arteries: a prospective comparison with digital subtraction angiography. Radiology 2001; 218:481-490.
11. Lee VS, Rofsky NM, Krinsky GA, Stemerman DH, Weinreb JC. Single-dose breath-hold gadolinium-enhanced three-dimensional MR angiography of the renal arteries. Radiology 1999; 211:69-78.
12. Shetty AN, Bis KG, Kirsch M, Weintraub J, Laub G. Contrast-enhanced breath-hold three-dimensional magnetic resonance angiography in the evaluation of renal arteries: optimization of technique and pitfalls. J Magn Reson Imaging 2000; 12:912-923.
13. Schoenberg SO, Essig M, Bock M, Hawighorst H, Sharafuddin M, Knopp MV. Comprehensive MR evaluation of renovascular disease in five breath holds. J Magn Reson Imaging 1999; 10:347-356.
14. Knopp MV, Floemer F, Schoenberg SO, von Tengg-Kobligk H, Bock M, van Kaick G. Non-invasive assessment of renal artery stenosis: current concepts and future directions in magnetic resonance angiography. J Comput Assist Tomogr 1999; 23:S111-S117.
15. Dong Q, Schoenberg SO, Carlos RC, et al. Diagnosis of renal vascular disease with MR angiography. RadioGraphics 1999; 19:1535-1554.
16. Lee VS, Martin DJ, Krinsky GA, Rofsky NM. Gadolinium-enhanced MR angiography: artifacts and pitfalls. AJR Am J Roentgenol 2000; 175:197-205.
17. Schoenberg SO, Bock M, Knopp MV, et al. Renal arteries: optimization of three-dimensional gadolinium-enhanced MR angiography with bolus-timing-independent fast multiphase acquisition in a single breath hold. Radiology 1999; 211:667-679.
18. De Cobelli F, Vanzulli A, Sironi S, et al. Renal artery stenosis: evaluation with breath-hold, three-dimensional, dynamic, gadolinium-enhanced versus three-dimensional, phase-contrast MR angiography. Radiology 1997; 205:689-695.
19. Thornton MJ, Thornton F, O’Callaghan , et al. Evaluation of dynamic gadolinium-enhanced breath-hold MR angiography in the diagnosis of renal artery stenosis. AJR Am J Roentgenol 1999; 173:1279-1283.
20. Davis CP, Hany TF, Wildermuth S, Schmidt M, Debatin JF. Postprocessing techniques for gadolinium-enhanced three-dimensional MR angiography. RadioGraphics 1997; 17:1061-1077.
21. Hany TF, Schmidt M, Davis CP, Goehde SC, Debatin JF. Diagnostic impact of four postprocessing techniques in evaluating contrast-enhanced three-dimensional MR angiography. AJR Am J Roentgenol 1998; 170:907-912.
22. Rieumont MJ, Kaufman JA, Geller SC, et al. Evaluation of renal artery stenosis with dynamic gadolinium-enhanced MR angiography. AJR Am J Roentgenol 1997; 169:39-44.
23. De Cobelli F, Venturini M, Vanzulli A, et al. Renal arterial stenosis: prospective comparison of color Doppler US and breath-hold, three-dimensional, dynamic, gadolinium-enhanced MR angiography. Radiology 2000; 214:373-380.
24. Hany TF, Leung DA, Pfammatter T, Debatin JF. Contrast-enhanced magnetic resonance angiography of the renal arteries. Invest Radiol 1998; 33:653-665.
25. Johnson PT, Halpern EJ, Kuszyk BS, et al. Renal artery stenosis: CT angiography—comparison of real-time volume-rendering and maximum intensity projection algorithms. Radiology 1999; 211:337- 343.
26. Sagami A. Evaluation of time of flight MR angiography for stenotic arterial lesions: including comparison of maximum intensity projection and volume rendering technique. Nippon Igaku Hoshasen Gakkai Zasshi 1994; 54:975-987.
27. Sakai G, Nishida N, Yamada R, Matsuoka T. Contrast-enhanced MR angiography of the epigastric and hepatic regions: visibility in three-dimensional reconstructions. Osaka City Med J 2000; 46:1-15.
28. Floemer F, Globitza G, Knopp MV, Schoenberg SO, Brockmeier K, Meinzer HP. Use of virtual reality for MRI data of complex vascular structures. Radiologe 2000; 40:246-255[German].
29. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307-310.
30. Svanholm H, Starklint H, Gundersen HJ, Fabricius J, Barlebo H, Olsen S. Reproducibility of histomorphologic diagnoses with special reference to the kappa statistic. APMIS 1989; 97:689-698.
31. Johnson PT, Heath DG, Bliss DF, Cabral B, Fishman EK. Three-dimensional CT: real-time interactive volume rendering. AJR Am J Roentgenol 1996; 167:581-583.
32. Ney DR, Fishman EK, Magig D. Volumetric rendering of computed tomography data: principles and techniques. Comput Graph Appl 1990; 10:24-32.
33. van de Ven PJ, Kaatee R, Beutler JJ, et al. Arterial stenting and balloon angioplasty in ostial atherosclerotic renovascular disease: a randomized trial. Lancet 1999; 353:282-286.
34. Rundback JH, Gray RJ, Rozenblit G, et al. Renal artery stent placement for the management of ischemic nephropathy. J Vasc Interv Radiol 1998; 9:413-420.
35. Tuttle KR, Chouinard RF, Webber JT, et al. Treatment of atherosclerotic ostial renal artery stenosis with the intravascular stent. Am J Kidney Dis 1998; 32:611-622.
36. Hoffman O, Carreres T, Sapoval MR, et al. Ostial renal artery stenosis angioplasty: immediate and mid-term angiographic and clinical results. J Vasc Interv Radiol 1998; 9:65-73.
37. Hennequin LM, Joffre FG, Rousseau HP, et al. Renal artery stent placement: long-term results with the Wallstent endoprosthesis. Radiology 1994; 191:713-719.

This article has been cited by other articles:

				M. Anzidei, A. Napoli, B. C. Marincola, I. Nofroni, D. Geiger, F. Zaccagna, C. Catalano, and R. Passariello Gadofosveset-enhanced MR Angiography of Carotid Arteries: Does Steady-State Imaging Improve Accuracy of First-Pass Imaging? Comparison with Selective Digital Subtraction Angiography Radiology, May 1, 2009; 251(2): 457 - 466. [Abstract] [Full Text] [PDF]

				G. Soulez, M. Pasowicz, G. Benea, L. Grazioli, J. P. Niedmann, M. Konopka, P. C. Douek, G. Morana, F. K. W. Schaefer, A. Vanzulli, et al. Renal Artery Stenosis Evaluation: Diagnostic Performance of Gadobenate Dimeglumine-enhanced MR Angiography--Comparison with DSA Radiology, April 1, 2008; 247(1): 273 - 285. [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]

				K. Nikolaou, H. Kramer, C. Grosse, D. Clevert, O. Dietrich, M. Hartmann, P. Chamberlin, S. Assmann, M. F. Reiser, and S. O. Schoenberg High-Spatial-Resolution Multistation MR Angiography with Parallel Imaging and Blood Pool Contrast Agent: Initial Experience Radiology, December 1, 2006; 241(3): 861 - 872. [Abstract] [Full Text] [PDF]

				A Persson, N Dahlstrom, O Smedby, and T B Brismar Volume rendering of three-dimensional drip infusion CT cholangiography in patients with suspected obstructive biliary disease: a retrospective study Br. J. Radiol., December 1, 2005; 78(936): 1078 - 1085. [Abstract] [Full Text] [PDF]

				H. Ota, K. Takase, H. Rikimaru, M. Tsuboi, T. Yamada, A. Sato, S. Higano, T. Ishibashi, and S. Takahashi Quantitative Vascular Measurements in Arterial Occlusive Disease RadioGraphics, September 1, 2005; 25(5): 1141 - 1158. [Abstract] [Full Text] [PDF]

				S. O. Schoenberg, J. Rieger, C. H. Weber, H. J. Michaely, T. Waggershauser, C. Ittrich, O. Dietrich, and M. F. Reiser High-Spatial-Resolution MR Angiography of Renal Arteries with Integrated Parallel Acquisitions: Comparison with Digital Subtraction Angiography and US Radiology, May 1, 2005; 235(2): 687 - 698. [Abstract] [Full Text] [PDF]

				M. Prokop, G. Schneider, A. Vanzulli, M. Goyen, S. G. Ruehm, P. Douek, M. Dapra, G. Pirovano, M. A. Kirchin, and A. Spinazzi Contrast-enhanced MR Angiography of the Renal Arteries: Blinded Multicenter Crossover Comparison of Gadobenate Dimeglumine and Gadopentetate Dimeglumine Radiology, February 1, 2005; 234(2): 399 - 408. [Abstract] [Full Text] [PDF]

				A. Mallouhi, S. Felber, A. Chemelli, A. Dessl, A. Auer, M. Schocke, W. R. Jaschke, and P. Waldenberger Detection and Characterization of Intracranial Aneurysms with MR Angiography: Comparison of Volume-Rendering and Maximum-Intensity-Projection Algorithms Am. J. Roentgenol., January 1, 2003; 180(1): 55 - 64. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2231010858v1 223/2/509    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mallouhi, A. Articles by Jaschke, W. R. Search for Related Content PubMed PubMed Citation Articles by Mallouhi, A. Articles by Jaschke, W. R.