Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print August 23, 2007, 10.1148/radiol.2451062081
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Supplemental Tables
Right arrow All Versions of this Article:
2451062081v1
2451062081v2
245/1/276    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sutter, R.
Right arrow Articles by Willmann, J. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sutter, R.
Right arrow Articles by Willmann, J. K.
(Radiology 2007;245:276-284.)
© RSNA, 2007

Vascular and Interventional Radiology

Assessment of Aortoiliac and Renal Arteries: MR Angiography with Parallel Acquisition versus Conventional MR Angiography and Digital Subtraction Angiography1

Reto Sutter, MD, Daniel Nanz, PhD, Amelie M. Lutz, MD, Thomas Pfammatter, MD, Burkhardt Seifert, PhD, Anja Struwe, RT, Christina Heilmaier, MD, Dominik Weishaupt, MD, Borut Marincek, MD, and Jürgen K. Willmann, MD

1 From the Institute of Diagnostic Radiology, University Hospital Zurich, Zurich, Switzerland (R.S., D.N., A.M.L., T.P., A.S., C.H., D.W., B.M., J.K.W.); Molecular Imaging Program at Stanford, Department of Radiology and Bio-X Program, Stanford University School of Medicine, E 150 Clark Center, 318 Campus Dr, Palo Alto, CA, 94305-5427 (A.M.L., J.K.W.); and Department of Biostatistics, University of Zurich, Zurich, Switzerland (B.S.). From the 2006 RSNA Annual Meeting. Received December 7, 2006; revision requested February 16, 2007; revision received March 1; accepted March 20; final version accepted April 16. J.K.W. supported in part by the Swiss Foundation of Medical-Biological Grants, Novartis Research Foundation, and Swiss Society of Radiology. Address correspondence to J.K.W. (e-mail: willmann{at}stanford.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
 References
 
Purpose: To prospectively compare the image quality, sensitivity, and specificity of three-dimensional gadolinium-enhanced magnetic resonance (MR) angiography accelerated by parallel acquisition (ie, fast MR angiography) with MR angiography not accelerated by parallel acquisition (ie, conventional MR angiography) for assessment of aortoiliac and renal arteries, with digital subtraction angiography (DSA) as the reference standard.

Materials and Methods: The study was approved by the institutional review board; informed consent was obtained from all patients. Forty consecutive patients (33 men, seven women; mean age, 63 years) suspected of having aortoiliac and renal arterial stenoses and thus examined with DSA underwent both fast (mean imaging time, 17 seconds) and conventional (mean imaging time, 29 seconds) MR angiography. The arterial tree was divided into segments for image analysis. Two readers independently evaluated all MR angiograms for image quality, presence of arterial stenosis, and renal arterial variants. Image quality, sensitivity, and specificity were analyzed on per-patient and per-segment bases for multiple comparisons (with Bonferroni correction) and for dependencies between segments (with patient as the primary sample unit). Interobserver agreement was evaluated by using {kappa} statistics.

Results: Overall, the image quality with fast MR angiography was significantly better (P = .001) than that with conventional MR angiography. At per-segment analysis, the image quality of fast MR angiograms of the distal renal artery tended to be better than that of conventional MR angiograms of these vessels. Differences in sensitivity for the detection of arterial stenosis between the two MR angiography techniques were not significant for either reader. Interobserver agreement in the detection of variant renal artery anatomy was excellent with both conventional and fast MR angiography ({kappa} = 1.00).

Conclusion: Fast MR angiography and conventional MR angiography do not differ significantly in terms of arterial stenosis grading or renal arterial variant detection.

Supplemental material: http://radiology.rsnajnls.org/cgi/content/full/2451062081/DC1

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
 References
 
Contrast material–enhanced three-dimensional (3D) magnetic resonance (MR) angiography has been established as an accurate technique for imaging essentially the entire vascular system, including the aortoiliac and renal arteries (1–5). However, investigators in a number of studies have reported contrast-enhanced 3D MR angiography, as compared with digital subtraction angiography (DSA), to have low accuracy for grading arterial stenosis (6–9).

Several characteristics of the current 3D MR angiography technology may contribute to the reported range of sensitivities and specificities in these studies. The spatial resolution with 3D MR angiography is about three to five times lower than the spatial resolution with DSA, and this hampers evaluation—especially that of small arteries. In addition, 3D MR angiography is limited by a relatively long data acquisition time. To image the aortoiliac and renal arteries with reasonable spatial resolution, mean breath-hold times of more than 25–30 seconds are needed (4,5,10). This limits the use of this technique in patients with compromised respiratory function and in patients who are unable to remain still during 3D MR data acquisition. Furthermore, prolonged data acquisition times may prevent preferential arterial enhancement during the first pass of the extracellular contrast medium and thus result in interfering venous opacification, particularly when small renal arteries are being analyzed (11).

Attempts to shorten the duration of individual phase-encoding steps by using pulse sequence modifications combined with improved gradient hardware have considerably reduced imaging time. However, further increases in gradient performance remain restricted owing to physiologic limits (12). In recent years, major steps toward decreasing MR examination times have been made owing to the introduction of parallel imaging techniques that include simultaneous acquisition of spatial harmonics (13), generalized autocalibrating partially parallel acquisition (14), and sensitivity encoding (15). A major limitation of parallel imaging, however, is the decreased signal-to-noise ratio (SNR), which is inversely proportional to the square root of the acceleration factor times a geometric factor determined mainly according to the coil design.

Although parallel imaging results in reduced MR signal intensity, contrast-enhanced 3D MR angiography of the abdominal aorta with an acceleration factor ranging from two to four has been proved to be feasible (11,13,16,17). To our knowledge, there had been no reported prospective study in which the image quality and accuracy of contrast-enhanced 3D MR angiography accelerated by parallel imaging (hereafter referred to as fast MR angiography) were compared with those of contrast-enhanced 3D MR angiography without parallel imaging (hereafter referred to as conventional MR angiography) in the same patient for assessment of the aortoiliac and renal arteries, with DSA as the reference standard. Thus, the purpose of our study was to prospectively compare the image quality, sensitivity, and specificity of fast MR angiography with those of conventional MR angiography in the assessment of aortoiliac and renal arteries, with DSA as the reference standard.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
 References
 
Patients
For 6 months, from June 2005 through November 2005, 57 consecutive patients who underwent diagnostic DSA of the aortoiliac and renal arteries for clinical indications at University Hospital Zurich were asked to participate in this prospective study (Fig 1). The study was approved by the institutional review board of University Hospital Zurich, and oral and written informed consent was obtained from all patients. Clinical indications for DSA were symptomatic aortoiliac occlusive disease in 52 (91%) patients and assessment of renal arterial stenosis in five (9%). Study exclusion criteria were history of an adverse reaction to paramagnetic contrast media, age younger than 18 years or older than 80 years, childbearing age without negative pregnancy test results, breast feeding, clinical instability, general contraindications to MR imaging, and unwillingness to provide written informed consent in accordance with guidelines set forth by the institutional review board. Seventeen patients were excluded from the study (Fig 1): 11 patients did not want to provide written informed consent, five were not clinically stable, and one had a history of adverse reaction to paramagnetic contrast media.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Flowchart of patient progress through the study. Patients who underwent DSA were randomly assigned to undergo fast or conventional MR angiography (MRA).

 
Hence, the final study group consisted of 40 consecutive patients (33 men aged 27–78 years [mean, 64 years], seven women aged 26–78 years [mean, 59 years]). Clinical indications for DSA in this study group were symptomatic aortoiliac occlusive disease in 36 (90%) patients and suspicion of renal arterial stenosis in four (10%). There was no significant difference in age between the men and women (P = .59). All patients underwent DSA and MR angiography within a 3-week period. DSA was performed initially and was followed by MR angiography in all patients. The mean time between DSA and MR imaging was 18 days (range, 10–21 days). For the purposes of this study, if a transluminal therapeutic procedure was considered feasible on the basis of DSA findings, the patient was scheduled for a second DSA examination after the completion of MR imaging. Thus, no transluminal therapeutic procedure was performed between the DSA and MR examinations.

MR Imaging
In all patients, two MR examinations—fast MR angiography and conventional MR angiography—were performed on separate days. Patients were randomly assigned to undergo one of the two examinations first: Twenty patients underwent fast MR angiography before conventional MR angiography, and 20 underwent conventional MR angiography first. The mean delay between the two examinations was 1.1 days (range, 1–3 days). In all 40 patients, MR angiography was performed by using a 1.5-T MR system (Signa Excite HD; GE Healthcare, Milwaukee, Wis) with a maximum gradient amplitude of 33 mT/m and a slew rate of 120 mT/(m·msec). All patients were positioned supine and feet first on the imaging table. For signal reception, an anteroposterior eight-element phased-array surface coil that encompassed the entire abdominal aorta and its renal branches and the iliac arteries was placed around the patient. Four coil elements were placed above and four elements were placed beneath the patient. During the first MR examination, the exact craniocaudal position of the coil was marked on the patient's skin with a water-proof marker so that the coil could be positioned at the exact same location for the second examination.

For all MR examinations, the transit time of a 1-mL test bolus of gadobutrol (Gadovist; Schering, Berlin, Germany) from the injection site (antecubital fossa) to the abdominal aorta was determined by using a sagittal multiphase single-section gradient-echo sequence (5/1 [repetition time msec/echo time msec], 60° flip angle). The test bolus and subsequent 25-mL normal saline flush were administered through a 20-gauge needle at a flow rate of 2 mL/sec by using an automated injector (MR Spectris; Medrad, Pittsburgh, Pa). The mean transit time was 24 seconds (range, 20–27 seconds) for both conventional and fast MR angiography. During the test bolus administration, the delay between the injection and the first major peak of enhancement was measured in the abdominal aorta. This same delay was used between the injection and the initiation of both MR angiographic examinations. The transit times determined in the two MR examinations were identical to within a second in all patients. The transit time was determined during end-inspiration breath holding. Because both the contrast agent dose (in millimoles per kilogram of bodyweight) and the injection rate were kept constant for both examinations, the ratio of contrast agent injection time to data acquisition time was larger with fast MR angiography than with conventional MR angiography.

The fast MR angiography protocol was designed with the same voxel size as the conventional MR angiography protocol but with a substantially reduced imaging time. Conventional MR angiography is routinely used to assess aortoiliac and renal arteries at our institution. For both protocols, a 3D fast spoiled gradient-echo sequence was used to acquire 48 coronal-oblique sections with a section thickness of 1.6 mm in all patients; this section thickness was similar to that used in other studies (18–20). For conventional MR angiography, 3.5/0.9, a 25° flip angle, a receiver bandwidth of ±62.5 kHz, a 256 x 192 matrix, and a 44.0 x 35.2-cm field of view were used. For fast MR angiography, an increased field of view of 48.0 x 38.4 cm was used with a matrix of 280 x 236 to consistently avoid parallel imaging artifacts (21). The associated increase in matrix size resulted in increased repetition and echo times of 4.2 and 1.3 msec, respectively, despite an increased receiver bandwidth of ±83.3 kHz. Both MR angiography data sets were interpolated to a larger matrix size and an interpolated voxel size of 0.9 x 0.9 x 0.8 mm during reconstruction. Gadobutrol was administered at a dose of 0.1 mmol/kg at a flow rate of 2 mL/sec and was followed by a 25-mL normal saline flush at the same flow rate. The total volume of gadobutrol administered for both MR angiography examinations ranged from 13 to 24 mL, depending on the patient's body weight. MR data were acquired during end-inspiration breath holding. Both MR angiography acquisitions involved centric elliptical phase-encode ordering.

For fast MR angiography, a commercial implementation of the sensitivity-encoding technique (ASSET [array spatial sensitivity encoding technique]; GE Healthcare) was used (15). With this technique, one reconstructs the full field of view by evaluating the intercoil variation in the superimposed true and aliased signal intensities in the undersampled data set. This reconstruction algorithm cannot be used to consistently correct artifacts caused by signal from outside the phase-encoding field of view, which is aliased more than once in the undersampled data set. In contrast to the aliased signal intensity that appears along the image edges in conventional acquisitions, the corresponding artifacts in parallel acquisitions often appear in the center of the image (21). Thus, the minimum reconstructed field of view in parallel acquisitions is moderately larger than the minimum field of view in conventional acquisitions. An acceleration factor of two was set for fast MR angiography. The mean data acquisition time for fast MR angiography was 17 seconds compared with 29 seconds for conventional MR angiography. No adverse events particularly related to gadobutrol administration occurred with the MR angiography examinations.

DSA Examinations
A vascular radiologist (T.P.) with 14 years experience in DSA performed intraarterial DSA of the aortoiliac and renal arteries with a retrograde technique in all 40 patients by using one of two angiography units (Integris V3000 or Integris V5000; Philips Medical Systems, Best, the Netherlands). For assessment of the abdominal aorta and the renal arteries, the side holes of a transfemoral 5-F pigtail flush catheter (AngiOptic; Angiodynamics, Queensbury, NY) were placed at the expected origin of the renal arteries: the L1–L2 intervertebral space. In all patients, 30 mL (300 mg of iodine per milliliter) of a nonionic iodinated contrast medium (iopromidum, Ultravist 300; Schering) was injected at a flow rate of 15 mL/sec by using an automatic injector (Angiomat 6000; Liebel-Flarsheim, Cincinnati, Ohio), and frontal DSA images were acquired at a matrix size of 1024 x 1024, field of view of 38 cm, and film rate of two to three images per second. If the vascular radiologist determined that 38-cm field-of-view abdominal aortography was not adequate for assessment of the renal arteries, additional 15° left or right anterior-oblique projections were obtained. The catheter was subsequently repositioned in the infrarenal aorta, and DSA of the iliac arteries in 30° left and right oblique projections was performed with 20 mL of iopromidum injected at a flow rate of 15 mL/sec. Electronic calipers were used to measure stenosis, and a built-in calibration system was applied. The DSA examinations caused no adverse events.

Quantitative MR Angiogram Analysis
The MR angiograms were quantitatively and qualitatively analyzed at a dedicated interactive workstation (Advantage Windows Workstation 4.2; GE Healthcare, Buc, France). One of the authors (R.S.), who had 3 years experience in vascular MR imaging and was blinded to all patient data, measured the SNRs and contrast-to-noise ratios (CNRs) for both fast and conventional MR angiography. Quantitative angiogram analysis was performed randomly in terms of patient order and type of MR examination. Measurements were performed in the suprarenal and infrarenal regions of the abdominal aorta; in the renal arteries, divided into three segments (proximal [ostium, <1 cm from origin of renal artery], middle [>1 cm to first branching of renal artery], and distal [first-generation branches] segments); in the common iliac arteries; in the external iliac arteries, divided into proximal and distal portions; in the internal iliac arteries; and in the common femoral arteries. One patient had undergone nephrectomy of the right kidney 22 months previously and thus had only one main renal artery. The accessory renal arteries were also divided into three segments. Measurements were performed in 768 (99%) of the 774 evaluatable arterial segments; this constituted all arterial segments with which both readers had moderate or better visibility, including 57 segments of accessory renal arteries.

Reader-defined regions of interest were placed in the middle of the given artery, in the adjacent retroperitoneal or extraperitoneal fat, and in an image region in the air adjacent to the body within the coil. Region-of-interest sizes aimed at encompassing as much of the different arteries being evaluated as possible (mean size, 20 mm2; range, 5–180 mm2) were chosen. The SNR was calculated as follows: SIa/SDb, where SIa is the mean signal intensity of the artery and SDb is the standard deviation of the magnitude background signal intensity outside the body within the coil (air). The CNR was calculated as follows: (SIa – SIf-p)/SDb, where SIf-p is the mean signal intensity in the adjacent retroperitoneal fat or adjacent liver parenchyma. The SNR and CNR with parallel imaging generally are lower than those with conventional MR angiography, and the signal intensities over a given image may vary (11). Nevertheless, SNR and CNR are considered valid imaging parameters when conventional and parallel MR angiography techniques are compared (11,22).

Qualitative MR Image Analysis
For qualitative analysis, vessels in the arterial vascular system were divided into the same segments evaluated for quantitative analysis, including the accessory renal arteries. Two readers (J.K.W., A.M.L., with 8 and 7 years experience in vascular MR imaging, respectively) independently assessed the subjective image quality of the angiograms of all arterial segments depicted with both MR angiography techniques. The readers were blinded to the patients' names, clinical data, and type of MR examination. All MR angiograms were analyzed in random order. Both readers were allowed to individually adjust the window centers and level settings of the MR data sets for image analysis at the workstation, and a cine mode was available for rapid interactive interpretation. Both readers were also allowed to use maximum intensity projections of the MR data sets in different planes when these were considered useful.

For each vessel, image quality was graded on a five-point Likert grading scale: Grade 1 meant the vessel was not visible and no diagnostic information could be obtained from the images. Grade 2 meant poor visibility: Image quality was degraded owing to low signal intensity and motion-induced blurring artifacts. Grade 3 meant moderate visibility: Image quality was degraded owing to low signal intensity or motion-induced blurring artifacts. Grade 4 meant good visibility owing to high signal intensity and slight motion-induced blurring artifacts. Grade 5 meant excellent visibility owing to high signal intensity and no motion-induced blurring artifacts.

The two readers graded the presence of arterial stenosis in the segments independently by using an electronic caliper: Grade 1 meant normal vessel or vessel irregularity (<10% luminal narrowing); grade 2, mild arterial stenosis (10% to <50% luminal narrowing); grade 3, severe arterial stenosis (50%–99% luminal narrowing); and grade 4, occlusion. In accordance with literature reports (23–27), grades 3 and 4 (50%–100% luminal narrowing) were considered to indicate hemodynamically significant arterial stenosis in our study group. When two or more stenotic lesions were detected in the same arterial segment, the most severe stenosis was used for grading and analysis.

A separate analysis of the two MR angiography techniques was performed to assess the presence of renal arterial variants. Both readers evaluated the presence and number of accessory left or right renal arteries, as well as the presence of early branching of the renal arteries (branching within 2 cm of the origin of the renal artery from the abdominal aorta) (28).

Analysis of DSA Findings
Analysis of the DSA images available on the interactive workstation was performed by one of two radiologists (T.P. or a second vascular radiologist, with 14 and 5 years experience, respectively). Interpretation disagreements were resolved by means of consensus review of five vascular segments. All DSA image cases were assessed in random order, and the readers were blinded to all patient data. To assess for possible arterial stenosis and renal arterial variants at DSA, we used the same classifications applied to evaluate the MR angiography data.

Statistical Analyses
SNR and CNR values are presented as means ± standard deviations. The paired t test was used to assess differences in SNR and CNR between the two MR angiography techniques. After Bonferroni correction to adjust for multiple comparisons, a comparison-wise P value of less than .0028 was considered to be indicative of a statistically significant difference. Differences in age between the women and men were assessed by using the Mann-Whitney test, with P < .05 indicating significance. With regard to the mean subjective image quality for all arterial segments combined, differences between the two MR angiography techniques were assessed by using the Wilcoxon signed rank test (P < .05) in a per-patient analysis. In a per-segment subanalysis, differences between the two techniques in terms of the subjective image quality of each of the 18 arterial segments being assessed were evaluated by using a paired sign test, with P < .0028 indicating significance after Bonferroni correction. For these analyses, the proportion procedure for survey data obtained with Stata software (StataCorp, College Station, Tex), with patient as the primary sample unit, was performed to address dependencies between segments.

Agreement between the two readers and between the two MR angiography techniques for renal arterial variant detection and arterial stenosis grading was determined by calculating {kappa} values. A {kappa} value of 0 indicated poor agreement; 0.01–0.20, slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, good agreement; and 0.81–1.00, excellent agreement (29). Owing to dependencies between segments, confidence intervals for {kappa} values were not calculated.

The sensitivity, specificity, positive and negative predictive values, and accuracy of the two MR angiography techniques (with 95% confidence intervals), as compared with DSA, for the detection of hemodynamically significant arterial stenosis were calculated for all arterial segments combined (total) and for each of two vascular regions separately: renal arteries (main and accessory renal arteries) and aortoiliac arteries (abdominal aorta; both common, external and internal iliac arteries, as well as both common femoral arteries). To address clustering by image analysis within the same patient, 95% confidence intervals were calculated by using the proportion procedure for survey data obtained with Stata software, with patient as the primary sample unit. The significance of differences in sensitivity between the two MR angiography techniques for both readers was assessed by analyzing the true findings per patient with use of the paired sign test, with P < .025 indicating significance after Bonferroni correction. Post hoc power analyses were performed for the comparisons of image quality and sensitivity per patient.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
 References
 
Quantitative Analysis of MR Angiography Results
The SNR and CNR for the aortoiliac and renal arteries depicted on both types of MR angiograms were measured in all 40 patients (Table E1 [http://radiology.rsnajnls.org/cgi/content/full/2451062081/DC1]). For all arterial segments, the SNR (P = .001) and CNR (P ≤ .007) were significantly higher with conventional MR angiography than with fast MR angiography (Fig 2).


Figure 2A
View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Bilateral intermittent claudication in 60-year-old man. Frontal maximum intensity projections of aortoiliac and renal arteries reconstructed from coronal 3D contrast-enhanced (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiography data sets. Although the SNRs and CNRs with fast MR angiography were lower than those with conventional MR angiography, the subjective image quality of angiograms of the aortoiliac and renal arteries depicted with both techniques was graded similarly by both readers. Neither reader detected aortoiliac or renal arterial stenosis with either technique. (c) Corresponding DSA findings confirmed the MR angiography results. A single right accessory renal artery (arrow) at the lower pole of the right kidney was diagnosed with both MR examinations by both readers and confirmed at DSA.

 

Figure 2B
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Bilateral intermittent claudication in 60-year-old man. Frontal maximum intensity projections of aortoiliac and renal arteries reconstructed from coronal 3D contrast-enhanced (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiography data sets. Although the SNRs and CNRs with fast MR angiography were lower than those with conventional MR angiography, the subjective image quality of angiograms of the aortoiliac and renal arteries depicted with both techniques was graded similarly by both readers. Neither reader detected aortoiliac or renal arterial stenosis with either technique. (c) Corresponding DSA findings confirmed the MR angiography results. A single right accessory renal artery (arrow) at the lower pole of the right kidney was diagnosed with both MR examinations by both readers and confirmed at DSA.

 

Figure 2C
View larger version (163K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Bilateral intermittent claudication in 60-year-old man. Frontal maximum intensity projections of aortoiliac and renal arteries reconstructed from coronal 3D contrast-enhanced (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiography data sets. Although the SNRs and CNRs with fast MR angiography were lower than those with conventional MR angiography, the subjective image quality of angiograms of the aortoiliac and renal arteries depicted with both techniques was graded similarly by both readers. Neither reader detected aortoiliac or renal arterial stenosis with either technique. (c) Corresponding DSA findings confirmed the MR angiography results. A single right accessory renal artery (arrow) at the lower pole of the right kidney was diagnosed with both MR examinations by both readers and confirmed at DSA.

 
Qualitative Analysis of MR Angiography Results
Subjective image quality.—The subjective image quality of angiograms of the aortoiliac and renal arteries depicted with both MR angiography techniques was graded by both readers independently (Table E2 [http://radiology.rsnajnls.org/cgi/content/full/2451062081/DC1]). The overall image quality grade for all 18 arterial segments in each patient was significantly higher with fast MR angiography than with conventional MR angiography (P = .001 for both readers). With the assumption that a difference in image quality between the two techniques could be considered clinically relevant when, on average, the image quality grade for two arterial segments was higher by at least one grade or the image quality grade for one segment was higher by at least two grades, a power of 96% was calculated for the difference in mean image quality at per-patient analysis.

At subanalysis of each of the 18 arterial segments, there were no significant differences between the two MR angiography techniques for either reader regarding the image quality of most of the 18 arterial segments (Table E2 [http://radiology.rsnajnls.org/cgi/content/full/2451062081/DC1]). There was a trend toward higher image quality for angiograms of the distal segments of the left (P = .008 for reader 1, P = .006 for reader 2) and right (P = .004, for reader 1) renal arteries with the fast examination compared with the image quality for these segments with the conventional examination. Reader 2 assigned a significantly higher image quality grade for the right distal renal arterial segment with fast MR angiography (P = .0026) (Table E2 [http://radiology.rsnajnls.org/cgi/content/full/2451062081/DC1]) (Fig 3). Both readers noted fewer motion-induced artifacts on the fast MR angiograms, compared with the number of these artifacts on the conventional images, as the main reason for the better image quality for the distal segments with the fast acquisition (Fig 3).


Figure 3A
View larger version (58K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Focused transverse maximum intensity projections of (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiographic data sets in 76-year-old man with generalized atherosclerosis and shortness of breath. Owing to motion-induced artifacts, both readers rated the image quality for the proximal and middle segments of the right renal artery (arrows) as moderate (grade 3) and the image quality for the distal segment as poor (grade 2) with conventional MR angiography. In contrast, both readers rated the image quality for the proximal and middle segments of the right renal artery as excellent (grade 5) and the image quality for the distal segment as good (grade 4) with fast MR angiography.

 

Figure 3B
View larger version (46K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Focused transverse maximum intensity projections of (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiographic data sets in 76-year-old man with generalized atherosclerosis and shortness of breath. Owing to motion-induced artifacts, both readers rated the image quality for the proximal and middle segments of the right renal artery (arrows) as moderate (grade 3) and the image quality for the distal segment as poor (grade 2) with conventional MR angiography. In contrast, both readers rated the image quality for the proximal and middle segments of the right renal artery as excellent (grade 5) and the image quality for the distal segment as good (grade 4) with fast MR angiography.

 
Renal arterial variants.—Both readers identified variant renal arterial anatomy on both types of MR angiograms in 18 (45%) of 40 patients (Table, Fig 2). Agreement between conventional and fast MR angiography and between readers 1 and 2 with both examinations was excellent ({kappa} = 1.00). The diagnoses of all renal arterial variants assigned by both readers with both types of MR angiography were in agreement with the DSA-based diagnoses.


View this table:
[in this window]
[in a new window]

  		Variant Renal Artery Anatomy Assessed by Two Independent Readers on Conventional and Fast MR Angiograms, with DSA Findings as Reference Standard

 
Aortoiliac and renal arterial stenoses.—For all grades of arterial stenosis, there was good interobserver agreement regarding conventional MR angiography findings ({kappa} = 0.69) and excellent interobserver agreement regarding fast MR angiography findings ({kappa} = 0.81) (Fig 4). For the diagnosis of hemodynamically significant (grades 3 and 4) versus nonsignificant (grades 1 and 2) arterial stenosis, there was excellent interobserver agreement regarding both conventional ({kappa} = 0.83) and fast ({kappa} = 0.90) MR angiography findings.


Figure 4A
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Generalized atherosclerosis and left leg claudication in 75-year-old man. On focused frontal maximum intensity projections of (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiographic data sets, both readers noted mild arterial stenosis (grade 2, <50% luminal narrowing) of the proximal segment of the left common iliac artery (large arrow) and grade 2 stenosis of the left internal iliac artery (small arrow). (c) Findings were confirmed on corresponding DSA image. The clinical symptoms of this patient were caused by high-grade stenosis of the distal superficial femoral artery, which was not in the field of view of the MR angiograms obtained in our study.

 

Figure 4B
View larger version (112K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Generalized atherosclerosis and left leg claudication in 75-year-old man. On focused frontal maximum intensity projections of (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiographic data sets, both readers noted mild arterial stenosis (grade 2, <50% luminal narrowing) of the proximal segment of the left common iliac artery (large arrow) and grade 2 stenosis of the left internal iliac artery (small arrow). (c) Findings were confirmed on corresponding DSA image. The clinical symptoms of this patient were caused by high-grade stenosis of the distal superficial femoral artery, which was not in the field of view of the MR angiograms obtained in our study.

 

Figure 4C
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c: Generalized atherosclerosis and left leg claudication in 75-year-old man. On focused frontal maximum intensity projections of (a) conventional (3.5/0.9) and (b) fast (4.2/1.3) MR angiographic data sets, both readers noted mild arterial stenosis (grade 2, <50% luminal narrowing) of the proximal segment of the left common iliac artery (large arrow) and grade 2 stenosis of the left internal iliac artery (small arrow). (c) Findings were confirmed on corresponding DSA image. The clinical symptoms of this patient were caused by high-grade stenosis of the distal superficial femoral artery, which was not in the field of view of the MR angiograms obtained in our study.

 
True-positive, true-negative, false-positive, and false-negative findings, as well as sensitivities, specificities, positive and negative predictive values, and accuracies for the diagnosis of hemodynamically significant arterial stenosis of the aortoiliac and renal arteries were calculated (Table E3 [http://radiology.rsnajnls.org/cgi/content/full/2451062081/DC1]). Differences in sensitivity per patient between conventional and fast MR angiography were not significant for reader 1 (P = .453) or reader 2 (P = .063). For the two readers, the sensitivities of conventional and fast MR angiography were identical for 33 and 35 of the 40 patients. With the assumption of a 15% proportion of discordant pairs, a study sample size of 120 patients would be necessary to detect 10% differences in sensitivity with 80% power.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
 References
 
Differences in sensitivity between conventional and fast MR angiography with regard to diagnosis of arterial stenosis were not significant in our study. Only a trend toward higher sensitivity with fast MR angiography than with conventional MR angiography was noted in the diagnosis of renal and aortoiliac arterial stenoses. In addition, there was no marked difference between conventional and fast MR angiography in the detection of variant renal arterial anatomy in our study. With both types of MR angiography, the findings of both readers for all renal arterial variants were in agreement with those of DSA.

Although we found SNR and CNR values to be significantly lower with the faster acquisition, both readers independently graded the mean image quality for all aortoiliac and renal arterial segments significantly higher with fast MR angiography. In a per-segment analysis of each of the 18 arterial segments assessed, however, only the image quality for the distal right renal artery segment was assigned a significantly higher grade with fast MR angiography by one reader. The image qualities of the proximal and middle renal arterial segments—in terms of visualization of the aortoiliac arteries in particular—were almost identical between the two types of MR angiography.

Fewer motion-induced blurring artifacts on the fast MR angiograms than on the conventional MR angiograms may be one explanation for the higher image quality grades, especially for the distal renal artery segments. Blurring artifacts of the renal arteries are a known limitation of conventional MR angiographic acquisitions. Kidney motion propagates to the renal arteries when the patient starts breathing during relatively long MR examinations, and this can detrimentally affect the image quality for the renal arteries (30). It has been shown that the distal segments of the renal arteries move about 10-fold more than do the proximal segments during normal respiration (31). Therefore, the improved visualization—especially that of the distal renal artery segments in our study—may be explained by a reduction of motion-induced blurring artifacts that results from the shortened mean imaging time (from 29 to 17 seconds in our study) achieved by using parallel acquisition. Our findings suggest that the image quality for the proximal and middle renal artery segments and the aortoiliac arteries may be less affected by a shortened acquisition time owing to the decreased motion of these segments during breathing.

Our results are in agreement with those of a study in which the image quality for the renal arteries depicted on two types of MR angiograms acquired with different acceleration factors (17) were compared. The proximal and middle renal artery segments were seen equally well at both MR angiography examinations in that study. However, the distal renal artery segments were better depicted on the MR angiograms acquired more rapidly—in 19 seconds—than on those acquired with the slower technique involving 26 seconds (17).

To our knowledge, our study is the first to systematically address the clinical value of a reduced acquisition time in MR angiography accelerated by parallel imaging for assessment of the aortoiliac and renal arteries in an intraindividual comparison with conventional MR angiography, with DSA as the reference standard. Fast MR angiography may be a valuable alternative to conventional MR angiography in patients with compromised respiratory function, who may experience difficulty holding their breath during imaging.

Our study had limitations. Our estimation of image noise based on the standard deviation of signal intensities measured in regions of interest in air outside the body did not account for the spatial noise variation that is characteristic of parallel imaging. Therefore, the SNR and CNR calculated for the parallel acquisitions in our study represent only rough estimations. Furthermore, since we did not compare findings between patients with and those without compromised respiratory function, the true differences in image quality, sensitivity, and specificity with a shorter acquisition time in patients with breath-holding difficulties could not be estimated from the results of our study. In addition, the limited prevalence of hemodynamically significant renal arterial stenosis in our study allowed us to draw only restricted conclusions regarding the sensitivity of both MR angiography techniques for the detection of significant renal arterial stenosis. Further studies with larger study samples are warranted.

Furthermore, with our study protocol, only a limited number of patients underwent selective DSA of the renal arteries for detailed assessment of the renal arteries. In addition, because the renal arterial stenoses in our patients were located in only the proximal segment of the renal arteries and because no patients with fibromuscular dysplasia were included in our study, the role of fast MR angiography for assessment of distal renal arterial stenosis and for examination of patients with fibromuscular dysplasia remains unclear.

In conclusion, our prospective study revealed, in an intraindividual comparison, that despite the lower SNR and CNR with fast MR angiography compared with those at conventional MR angiography, the subjective image quality for the aortoiliac and renal arteries is not degraded with the fast acquisition. This is mainly because of the reduced motion-induced artifacts on the fast MR angiograms. Furthermore, our study results show that the sensitivity of fast MR angiography for the diagnosis of aortoiliac and renal arterial stenoses is not significantly different from that of conventional MR angiography. However, the role of fast MR angiography in the assessment of both distal renal arterial stenosis and fibromuscular dysplasia remains unclear.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE
 References
 


   IMPLICATION FOR PATIENT CARE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
References
 


   FOOTNOTES
 

Abbreviations: CNR = contrast-to-noise ratio • SNR = signal-to-noise ratio • 3D = three-dimensional

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, R.S., J.K.W.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, R.S., J.K.W.; clinical studies, R.S., D.N., A.M.L., T.P., A.S., J.K.W.; statistical analysis, B.S., J.K.W.; and manuscript editing, R.S., D.N., A.M.L., J.K.W.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATION FOR PATIENT CARE
 References
 

  1. Glockner JF. Three-dimensional gadolinium-enhanced MR angiography: applications for abdominal imaging. RadioGraphics 2001;21:357–370.[Abstract/Free Full Text]
  2. Korst MB, Joosten FB, Postma CT, Jager GJ, Krabbe JK, Barentsz JO. Accuracy of normal-dose contrast-enhanced MR angiography in assessing renal artery stenosis and accessory renal arteries. AJR Am J Roentgenol 2000;174:629–634.[Abstract/Free Full Text]
  3. Nael K, Laub G, Finn JP. Three-dimensional contrast-enhanced MR angiography of the thoraco-abdominal vessels. Magn Reson Imaging Clin N Am 2005;13:359–380.[CrossRef][Medline]
  4. Vosshenrich R, Fischer U. Contrast-enhanced MR angiography of abdominal vessels: is there still a role for angiography? Eur Radiol 2002;12:218–230.
  5. Willmann JK, Wildermuth S, Pfammatter T, et al. Aortoiliac and renal arteries: prospective intraindividual comparison of contrast-enhanced three-dimensional MR angiography and multi-detector row CT angiography. Radiology 2003;226:798–811.[Abstract/Free Full Text]
  6. Golay X, Brown SJ, Itoh R, Melhem ER. Time-resolved contrast-enhanced carotid MR angiography using sensitivity encoding (SENSE). AJNR Am J Neuroradiol 2001;22:1615–1619.[Abstract/Free Full Text]
  7. Schoenberg SO, Rieger J, Weber CH, et al. High-spatial-resolution MR angiography of renal arteries with integrated parallel acquisitions: comparison with digital subtraction angiography and US. Radiology 2005;235:687–698.[Abstract/Free Full Text]
  8. Vasbinder GB, Nelemans PJ, Kessels AG, Kroon AA, de Leeuw PW, van Engelshoven JM. Diagnostic tests for renal artery stenosis in patients suspected of having renovascular hypertension: a meta-analysis. Ann Intern Med 2001;135:401–411.[Abstract/Free Full Text]
  9. Vasbinder GB, Nelemans PJ, Kessels AG, et al. Accuracy of computed tomographic angiography and magnetic resonance angiography for diagnosing. Ann Intern Med 2004;141:674–682.[Abstract/Free Full Text]
  10. Heiss SG, Shifrin RY, Sommer FG. Contrast-enhanced three-dimensional fast spoiled gradient-echo renal MR imaging: evaluation of vascular and nonvascular disease. RadioGraphics 2000;20:1341–1352.[Abstract/Free Full Text]
  11. Chen Q, Quijano CV, Mai VM, et al. On improving temporal and spatial resolution of 3D contrast-enhanced body MR angiography with parallel imaging. Radiology 2004;231:893–899.[Abstract/Free Full Text]
  12. Kanal E, Borgstede JP, Barkovich AJ, et al. American College of Radiology white paper on MR safety. AJR Am J Roentgenol 2002;178:1335–1347.[Free Full Text]
  13. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med 1997;38:591–603.[Medline]
  14. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 2002;47:1202–1210.[CrossRef][Medline]
  15. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med 1999;42:952–962.[CrossRef][Medline]
  16. Born M, Willinek WA, Gieseke J, von Falkenhausen M, Schild H, Kuhl CK. Sensitivity encoding (SENSE) for contrast-enhanced 3D MR angiography of the abdominal arteries. J Magn Reson Imaging 2005;22:559–565.[CrossRef][Medline]
  17. Michaely HJ, Herrmann KA, Kramer H, et al. High-resolution renal MRA: comparison of image quality and vessel depiction with different parallel imaging acceleration factors. J Magn Reson Imaging 2006;24:95–100.[CrossRef][Medline]
  18. Kroencke TJ, Wasser MN, Pattynama PM, et al. Gadobenate dimeglumine-enhanced MR angiography of the abdominal aorta and renal arteries. AJR Am J Roentgenol 2002;179:1573–1582.[Abstract/Free Full Text]
  19. 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.[Abstract/Free Full Text]
  20. Goyen M, Herborn CU, Kroger K, Lauenstein TC, Debatin JF, Ruehm SG. Detection of atherosclerosis: systemic imaging for systemic disease with whole-body three-dimensional MR angiography—initial experience. Radiology 2003;227:277–282.[Abstract/Free Full Text]
  21. Goldfarb JW. The SENSE ghost: field-of-view restrictions for SENSE imaging. J Magn Reson Imaging 2004;20:1046–1051.[CrossRef][Medline]
  22. Zenge MO, Vogt FM, Brauck K, et al. High-resolution continuously acquired peripheral MR angiography featuring partial parallel imaging GRAPPA. Magn Reson Med 2006;56:859–865.[CrossRef][Medline]
  23. Barber GG, Fong H, McPhail NV, Scobie TK. Hemodynamic assessment of the aortoiliac segment: a prospective study. Can J Surg 1980;23:542–544.[Medline]
  24. Snidow JJ, Johnson MS, Harris VJ, et al. Three-dimensional gadolinium-enhanced MR angiography for aortoiliac inflow assessment plus renal artery screening in a single breath hold. Radiology 1996;198:725–732.[Abstract/Free Full Text]
  25. Quinn SF, Sheley RC, Szumowski J, Shimakawa A. Evaluation of the iliac arteries: comparison of two-dimensional time of flight magnetic resonance angiography with cardiac compensated fast gradient recalled echo and contrast-enhanced three-dimensional time of flight magnetic resonance angiography. J Magn Reson Imaging 1997;7:197–203.[Medline]
  26. Poon E, Yucel EK, Pagan-Marin H, Kayne H. Iliac artery stenosis measurements: comparison of two-dimensional time-of-flight and three-dimensional dynamic gadolinium-enhanced MR angiography. AJR Am J Roentgenol 1997;169:1139–1144.[Abstract/Free Full Text]
  27. Schaefer PJ, Boudghene FP, Brambs HJ, et al. Abdominal and iliac arterial stenoses: comparative double-blinded randomized study of diagnostic accuracy of 3D MR angiography with gadodiamide or gadopentetate dimeglumine. Radiology 2006;238:827–840.[Abstract/Free Full Text]
  28. Rusnack D, Israel GM. Kidney transplantation: evaluation of donors and recipients. Magn Reson Imaging Clin N Am 2004;12:505–514; vi–vii.
  29. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–374.[CrossRef][Medline]
  30. Vasbinder GB, Maki JH, Nijenhuis RJ, et al. Motion of the distal renal artery during three-dimensional contrast-enhanced breath-hold MRA. J Magn Reson Imaging 2002;16:685–696.[CrossRef][Medline]
  31. Draney MT, Zarins CK, Taylor CA. Three-dimensional analysis of renal artery bending motion during respiration. J Endovasc Ther 2005;12:380–386.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
A. Nchimi, D. Brisbois, R. Materne, T. K. Y. Broussaud, I. Mancini, and P. Magotteaux
Free-Breathing Accelerated Gadolinium-Enhanced MR Angiography in the Diagnosis of Renovascular Disease
Am. J. Roentgenol., June 1, 2009; 192(6): 1531 - 1537.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Supplemental Tables
Right arrow All Versions of this Article:
2451062081v1
2451062081v2
245/1/276    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sutter, R.
Right arrow Articles by Willmann, J. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sutter, R.
Right arrow Articles by Willmann, J. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE