| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Depts of Radiology of Univ Hosp, Währinger Gürtel 18–20, A-1093 Vienna, Austria (S.A.T., C.L.); Osatek Hosp de Galdácano, Spain (A.C.); Hosp Ruber Internacional, Madrid, Spain (F.H.D.O.); Centre Hosp Univ, Liège, Belgium (R.F.D.); Donostia-San Sebastián, Spain (C.G.); Martin Luther Univ, Halle, Germany (A.G.J.); Univ Central Hosp, Helsinki, Finland (P.K.); Univ Hosp, Nottingham, United Kingdom (C.N.L.); Nycomed Amersham Imaging, Munich, Germany (M.M.); Hosp Univ Dr Peset, Valencia, Spain (J.P.d.C., L.M.B.); Centre Hosp Régional, Univ de Lille, France (J.P.P.); Hosp General Univ, Valencia (V.M.S.); and Univ Hosp Rudolf Virchow, Berlin, Germany (T.V.). Supported by Nycomed Amersham. Received Jan 4, 2000; revision requested Feb 28; final revision received Aug 21; accepted Sep 14. Address correspondence to S.A.T. (e-mail: siegfried.thurnher@univie.ac.at).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: One hundred five patients were included in the randomized, double-blind, phase III multicenter trial. Results of MR angiography with 0.1 mmol/kg and 0.3 mmol/kg doses of gadodiamide were compared with those of digital subtraction angiography (DSA) and according to dose.
RESULTS: No serious adverse events were observed. The mean contrast index at the region proximal to the primary stenosis was significantly higher in the triple-dose group (P = .03). Mean 95% CI values for the difference in depicted degree of stenosis between DSA and postcontrast MR angiography improved from -3.4% ± 4.7 (SD) in the single-dose group to -1.2% ± 4.7 in the triple-dose group. Mean values for overall image quality on the visual analogue scale improved with the triple dose (P = .02). Confidence in diagnosis was high at postcontrast MR angiography in 88% and 96% of cases in the single- and triple-dose groups, respectively.
CONCLUSION: Gadodiamide-enhanced MR angiography performed with single and triple doses is safe and effective for assessing major abdominal arterial stenoses. Although high agreement between MR angiography and DSA was achieved with both doses, triple-dose MR angiography was superior in the evaluations of image quality, degree of arterial stenoses, and confidence in diagnosis.
Index terms: Arteries, stenosis or obstruction, 95.72, 96.72, 98.72 • Contrast media • Gadolinium • Magnetic resonance (MR), vascular studies, 95.12942, 95.12943, 96.12942, 96.12943, 98.12942, 98.12943
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Most experiences with contrast-enhanced MR angiography have involved gadopentetate dimeglumine, which is an ionic agent and, to our knowledge, the first contrast material approved for clinical use. Gadodiamide (Omniscan; Nycomed Amersham, Oslo, Norway) is an approved nonionic gadolinium chelate with low osmolality and viscosity that is intended for intravascular use. The safety, tolerance, and pharmacokinetic profile of gadodiamide have been evaluated, and it has proved to be safe and effective at doses of 0.1 and 0.3 mmol/kg (2–6). The overall adverse event rate in this clinical situation has been 1%–2%, and the most commonly reported adverse events have been headache, nausea, dizziness, and vomiting (3). However, in routine clinical practice and postmarket surveillance trials, the adverse event rate of gadodiamide injection has been about 0.01%.
To ensure adequate arterial enhancement during image acquisition, the contrast material timing and dose must be optimized (7–11). Until recently, few studies with comparisons of contrast material doses for MR angiography had been published (1,12,13). Breath-hold MR angiography with fast bolus injection of gadolinium chelates at doses of 0.1–0.3 mmol/kg has been developed (14,15). However, doses greater than 0.1 mmol/kg have not been approved for MR angiography of the abdomen.
The aim of this randomized, double-blind, multicenter trial was to evaluate the effectiveness and safety of gadodiamide-enhanced MR angiography with two dose levels—0.1 and 0.3 mmol/kg—in the assessment of stenoses of the abdominal aorta, its major branches, and the iliac arteries. Intraarterial digital subtraction angiography (DSA) was the standard of reference.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Patients were excluded from the study if they met any of the following criteria: (a) were pregnant or breast feeding, (b) were previously included in this trial or currently participating in other clinical trials, (c) had a life-threatening disease with a life expectancy of less than 1 month, (d) had received or were scheduled to receive another contrast material (MR or iodinated) within a minimum of 12 hours before or after the proposed MR examination, (e) had severely impaired renal function (ie, creatinine level, >2.5 mg/dL [221 µmol/L]), and/or (f) had any contraindication to MR imaging (ie, absolute or relative contraindications according to the hospital routine, such as pacemakers, implanted pumps, drains, stents, metallic prostheses, or surgical clips). After being evaluated for inclusion and exclusion criteria, the patients were placed into one of the two dose groups, 0.1 or 0.3 mmol/kg. Commercial randomization software (CLINPRO; Clinical System, Garden City, NY) was used for the randomization process. Approval from the institutional review board was obtained at each site.
The effectiveness of MR angiography was evaluated in 105 patients. Six patients were excluded from this analysis because of noncompliance with the protocol: in three patients, DSA was not performed; one patient underwent angioplasty before MR angiography; one patient received the wrong contrast material; and another patient had claustrophobia, so MR angiography was cancelled. Fifty-three patients (nine women, 44 men; mean age, 64.9 years; age range, 45–83 years) were in dose group 1 (0.1 mmol/kg). Dose group 2 (0.3 mmol/kg) consisted of 52 patients (eight women, 44 men; mean age, 61.9 years; age range, 38–85 years). All patients were white, and there were no major differences in mean weight between the two groups: 40–115 kg (mean, 72.5 kg) in group 1 and 43–96 kg (mean, 73.3 kg) in group 2. The mean height of the patients ranged from 151 to 183 cm in group 1 and from 147 to 187 cm in group 2. The mean body mass index was similar in both dose groups: 25.7 in group 1 and 25.5 in group 2.
At trial initiation, there were 80 patients—43 (81%) in the single-dose group and 37 (71%) in the triple-dose group—with a history of ischemic and coronary heart disease, thrombosis, arterial hypertension, hypercholesterolemia, and/or diabetes mellitus.
Of 105 patients, 71 (67.6%) were known to have a single or multiple risk factors, including cardiovascular disease (51.4%), diabetes mellitus (19.1%), allergy or hypersensitivity (17.1%), renal disease (6.7%), asthma (1.0%), and other factors (10.5%). There were no major differences in the distribution of risk factors or the medical history between the two dose groups.
A total of 103 patients—all 53 (100%) in group 1 and 50 (96%) in group 2—received medication that was not related to MR imaging 24 hours before that examination took place.
Study Design
This trial was designed as a double-blind, randomized, phase III study performed at 12 centers in seven European countries to compare MR angiography performed with two gadodiamide injection doses—a single 0.1 mmol/kg dose and a triple dose totaling 0.3 mmol/kg—and to compare the results of these examinations with those of DSA. After DSA and MR angiography, blinded copies of the images were made and assessed by an independent reader, whose main objective was to compare the degree of stenosis between DSA and MR angiography rather than determine the detection rate.
To ensure a direct comparison of stenosis degree, a subinvestigator, who was not otherwise involved in the trial, marked and numbered the stenoses that were to be measured on the digital subtraction and MR angiograms to ensure that the same stenoses were measured by using both methods. These images were then forwarded to two independent readers (J.P.d.C., L.M.B.) (one for DSA and one for MR angiography), who were blinded with regard to patient information, gadodiamide injection administration, and contrast material concentration. The two independent readers received the images in a randomized order and were unaware of any clinical information.
Each patient was monitored for clinical reactions to imaging. The patients remained at the hospital for 2 hours for close surveillance. They were checked again for possible adverse events 24 hours after MR angiography. Adverse events were classified as (a) discomfort or (b) adverse events other than discomfort. All adverse events reported by the patient or observed by the investigator or hospital personnel were reported. To be considered discomfort, the experience had to be temporarily associated with the injection, transient, self-limiting, and be one of the following: (a) a sensation of heat, warmth, cold, or coolness or (b) pain or pressure at the injection site.
Imaging
All patients enrolled in the study underwent MR angiography before and after gadodiamide injection (Figs 1, 2). Commercially available MR imaging devices with field strengths of 1.0 or 1.5 T were used: Magnetom Expert 1.0 T, MultiStar Top, Vision 1.5 T, and Magnetom 1.5 T (Siemens, Erlangen, Germany); Gyroscan 1.0 T and Gyroscan 1.5 T (Philips, Eindhoven, the Netherlands); and Echo Speed 1.5 T (GE Medical Systems, Milwaukee, Wis). Each study center used identical sequences for all vessel areas. A phased-array, quadrature, or body coil was used. An initial two-dimensional time-of-flight MR angiographic sequence, with or without cardiac triggering, was performed in all patients before contrast material administration. The following parameters were used: 7.5–48.0/2.5–9.0, 30°–70° flip angle, 91–256 x 128–256 matrix, and 2–5-mm section thickness.
|
|
|
|
|
The imaging delay time was calculated as follows: D = TT - (AT/4), where D is the delay time; TT, the transit time; and AT, the acquisition time. The contrast material was administered as a bolus, either manually (at four sites) or by using a power injector (at seven sites), and followed by a saline flush. The application time was 5–20 seconds for the single-dose group and 6–33 seconds for the triple-dose group.
The contrast-enhanced MR examination consisted of three-dimensional spoiled gradient-echo sequences (eg, fast field echo, fast imaging with steady precession, and fast spoiled gradient echo) with the following parameters: 3.8–9.0/1.3–3.0, 30°–60° flip angle, 106–512 x 163–512 matrix, and 1.3–5.9-mm section thickness. With this sequence, the high- or low-frequency k-space lines were acquired at the beginning of the acquisition consistently at each center.
Routine intraarterial DSA techniques (Seldinger technique and femoral approach) were used at each site. A catheter was advanced into the abdominal aorta, and a nonionic contrast material (iopromide [Ultravist 300]; Schering, Berlin, Germany or iohexol [Omnipaque 300]; Nycomed-Amersham, Oslo, Norway) was injected. At least one projection of the abdominal aorta and iliac arteries was obtained for measurement of the stenosis. DSA was performed before MR angiography in the majority of patients. The minimum time between DSA and MR angiography was 1 day (range, 1–29 days; mean, 5 days).
Quantitative Image Analysis
The main effectiveness parameter for comparison of the two dose groups analyzed was the contrast index (CIx), which was calculated from the signal intensities measured in the regions of interest in the abdominal aorta (at the level of the renal arteries), at the aortic bifurcation, and 2 cm proximal and 2 cm distal to the main stenosis. As an anatomic reference, the signal intensity of the perivascular fat (SIREF) was determined. Primary stenosis was defined as the most severe and proximal stenosis greater than 50%, as diagnosed at DSA.
The CIx values obtained at doses of 0.1 mmol/kg and 0.3 mmol/kg were compared. The CIx was calculated as follows: CIx = (SIROI - SIREF)/SIREF, where SIROI is the signal intensity of the region of interest. Signal-to-noise ratio was defined as the ratio of mean signal intensity in the aorta to the SD of the signal intensity in the air (ie, noise) (SDNOISE) lateral to the abdomen. The contrast-to-noise ratio (CNR) was calculated according to the formula CNR = (SIROI - SIREF)/SDNOISE.
The CIx, CNR, and signal-to-noise ratio were calculated from the pre- and postcontrast signal intensities and compared for each dose group separately. The signal intensities were measured on the raw-data images.
Degree of stenosis was defined as the ratio of the minimum transverse diameter of the stenosis to the diameter of the normal vascular segment distal to the stenosis. These measurements were performed on maximum intensity projection and source images combined, reformatted images, or multiplanar reconstruction images. When the stenosis was at the iliac bifurcation, the other side (or if stenosed, the arterial segment proximal to the stenosis) was used as a reference for the normal diameter of that segment. In cases of stenosis in the celiac trunk or the renal arteries, the reference measurement was performed in a normal segment distal to the stenosis. The two dose groups were compared to determine the dose with MR angiography that yielded the most precise estimation of the degree of stenosis, as compared with DSA. Thus, the differences between DSA and MR angiography were calculated for each dose group.
The difference in degree of stenosis was analyzed overall and for three categories. Category 1 consisted of stenoses of 69% or less; category 2, stenoses between 70% and 99%; and category 3, occlusions. The independent investigators compared the precontrast MR angiographic, postcontrast MR angiographic, and DSA data.
In addition, the accuracy, sensitivity, and specificity of contrast-enhanced MR angiography for the detection of different cutoff points of stenosis degree based on DSA results were calculated: 50% or less versus greater than 50% stenosis; 70% or less versus greater than 70% stenosis, and 99% or less stenosis versus occlusion.
Qualitative Image Analysis
The two-dimensional time-of-flight and contrast-enhanced three-dimensional MR angiograms were analyzed to determine whether there was a gain in diagnostic information with the gadolinium-enhanced images (less, equal, more [marginal, moderate, great]); better depiction of smaller vessels (less, equal, more); and better depiction of collateral vessels or accessory renal arteries (less, equal, more); the confidence in diagnosis (low, moderate, high); the possible changes in diagnosis and/or patient treatment (on-site investigator); and the reasons for these changes. The quality of vessel delineation was assessed on a four-point scale: 1 indicated very unsatisfactory and not useful for diagnostic purposes; 2, unsatisfactory, with serious compromise of diagnostic image quality; 3, satisfactory, with good diagnostic image quality; 4, very satisfactory, with very good diagnostic image quality.
Overall image quality was measured on a visual analogue scale with a length of 10 cm; ratings ranged from insufficient (0 cm) to optimal (10 cm). The influence of artifacts on the images and the quality of vessel delineation were evaluated. Artifacts were classified as (a) signal loss, indicated by susceptibility and flow; (b) focal signal intensity increase, indicated by bright spots; (c) blurring of borders between the enhancing and nonenhancing compartments; and (d) other artifacts (eg, motion). When artifacts were present, their overall effect on image quality was indicated on a three-degree scale: (a) images not useful for diagnostic purposes, (b) artifacts seriously compromised diagnostic image quality, or (c) minor artifacts that did not affect diagnostic image quality.
Statistical Analyses
Summary statistical and data analysis tabulations were performed by using SAS software (SAS, Cary, NC). The value of significance (ie, 
level) for statistical tests was .05.
The CIx, CNR, and signal-to-noise ratio were calculated for the primary stenosis in each patient. All enhancement characteristics were determined in four regions of interest. The CIx of the region of interest 2 cm proximal to the stenosis was evaluated as the main parameter, and the other three regions (2 cm distal to the stenosis, the aorta, and the aortic bifurcation) also were evaluated. Signal intensity data were logarithmically transformed to achieve normal distribution and equal variances in the two dose groups, and two-way analysis of variance was applied by including dose group and testing center as fixed factors in the linear model. The correlations between center and dose were tested. Because correlations were not significant, it could be concluded that dose effects were similar among centers. Correlations between center and dose were therefore excluded from the model when center and dose effects were analyzed.
Assessments of percentage stenosis at DSA and MR angiography were performed by the independent reader. The percentage of primary stenosis was measured at DSA and MR angiography, and the difference between these values was calculated for each patient. The optimal gadodiamide injection dose with MR angiography was that which resulted in the most consistent determination of percentage stenosis (with respect to mean differences and CIs), as compared with DSA. MR angiography and DSA were regarded as equally efficient when the CI of the mean difference between DSA and MR angiography for the optimal dose group was within the plus or minus 10% range. In addition, the mean differences in percentage of stenosis between the values recorded on the precontrast MR angiograms and DSA images and the corresponding 95% CIs were calculated separately for each dose group.
The overall quality of the MR angiograms was assessed by using a visual analogue scale. Because the data were not normally distributed and the testing center effect was not significant, the Mann-Whitney test was performed. To compare the pre- and postcontrast MR angiograms and DSA images, a Student t test for matched pairs was performed, because the distribution of the differences was close to a normal Gaussian distribution. By using both tests, two-sided hypotheses were tested to show the difference between the dose groups, modalities, and pre- and postcontrast MR assessments.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Adverse events.—No serious adverse events were observed in this study. After DSA, two patients (1%) experienced adverse events: vertigo and urticaria. After MR angiography, no patient in the single-dose group and two (4%) in the triple-dose group reported adverse events: fatigue for 4 hours and vomiting. It was uncertain whether the fatigue was caused by the contrast material injection. The vomiting was due to an underlying disease and thus unrelated to the gadodiamide administration.
Discomfort.—Ninety-two of the 109 patients did not experience discomfort. Six patients (11%) in the single-dose group felt a sensation of heat or warmth in the abdomen, lumbar spine, upper legs, hip, or groin. Five of these events were mild, and one was moderate. One patient (2%) in this dose group felt a mild coldness or coolness at the injection site.
Six patients (11%) in the triple-dose group felt a warm discomfort in the abdomen, all over the body, in the pelvis, or in the buttocks. Two of these events were mild, and four were moderate. Coldness or coolness at the injection site or in the arm was experienced by four other patients (7%). Three of these events were mild, and one was moderate.
Effectiveness of MR Angiography
CIx, signal-to-noise ratio, and CNR values.—The CIx values are summarized in Table 1. The precontrast CIx values were very similar, regardless of stenosis location or dose group. Because the postcontrast CIx values varied considerably among testing centers (P = .001), the center effect was included in this statistical model. Comparison of the pre- and postcontrast values showed that the CIx was much higher after gadodiamide injection in all vascular segments. With the exception of the aortic bifurcation, this increase correlated with the concentration of contrast material (Table 1).
|
The CNRs were nearly identical in all four vascular segments at precontrast MR angiography (Table 2). The injection of gadodiamide in the single-dose group resulted in a mild increase in CNR in the aorta, aortic bifurcation, and area distal to the main stenosis and a marked increase in the area proximal to the primary stenosis (Table 2). The CNR markedly increased with the triple dose of contrast material in all four vascular regions of interest (P = .020–.026).
|
|
|
An important clinical parameter was the difference in percentage of primary stenosis at DSA and MR angiography in each patient. To determine the difference in degree of primary stenosis, only those patients in whom all examinations were performed (33 patients in dose group 1 and 31 patients in dose group 2) were included in the analysis. In the comparison of pre- and postcontrast MR angiograms, the independent investigator scored category 2 stenoses (70%–99%) higher on the precontrast MR angiograms (difference of 5.7% ± 20.5 for dose group 1 and of 15.1% ± 21.5 for dose group 2) (Table 5). For category 3 stenoses (ie, occlusions), no differences between pre- and postcontrast MR angiographic assessments were found.
|
The mean differences in percentage of stenosis between DSA and precontrast MR angiography and corresponding 95% CIs were calculated. The readers found no significant difference between the dose groups: a difference of 0.9 in dose group 1 and of 0.8 in dose group 2. The 95% CI was 8.6 (mean, 0.9 ± 4.3) for dose group 1 and 12.7 (mean, 0.8 ± 6.35) for dose group 2.
The CIs for the difference between DSA and postcontrast MR angiography were smaller: -1.2% ± 4.7 for dose group 2 compared with -3.4% ± 4.7 for dose group 1. Both CIs were within the predefined range of plus or minus 10%, which indicated an equivalence between DSA and MR angiography. Basically, the occlusions seen at DSA were visualized also at postcontrast MR angiography in most instances. Category 2 stenoses were judged to be similar to DSA in patients in the triple-dose group (mean difference of -3.1% ± 10.2 for the single-dose group and of 1.7% ± 7.0 for the triple-dose group).
The accuracy, sensitivity, and specificity of MR angiography were calculated for three cut-off points of degree of primary stenosis based on DSA findings, the standards of reference (Table 6). For a cut-off point of 70% stenosis, analysis results revealed the sensitivity, specificity, and accuracy of MR angiography to be 88%, 67%, and 84%, respectively, for dose group 1 and 88%, 57%, and 84%, respectively, for dose group 2 (Table 6). In the assessment of occlusions, the sensitivity, specificity, and accuracy of MR angiography were 94% each for dose group 1 and 100%, 97%, and 98%, respectively, for dose group 2.
|
When comparing pre- and postcontrast overall image quality, the mean postcontrast image quality values were 3.07 and 3.87 higher on the visual analogue scale for dose groups 1 and 2, respectively (P = .001 for both groups). The 95% CIs for these differences were 2.41, 3.73 and 3.21, 4.53 for dose groups 1 and 2, respectively.
In addition, the mean differences between DSA and precontrast MR angiographic visual analogue scale values were calculated and ranged from 3.03 to 4.02. The overall image quality at precontrast MR angiography was worse than that at DSA in both dose groups (P < .001 for both groups). The mean differences in visual analogue scale values between DSA and postcontrast MR angiography were small: -0.04 and -0.25 in dose groups 1 and 2, respectively. Therefore, the overall image quality of DSA and postcontrast MR angiograms did not differ significantly (P = .90 and .19 for dose groups 1 and 2, respectively).
Artifacts at MR angiography and DSA.—The readers found no artifacts on 16% and 14% of the precontrast MR angiograms in dose groups 1 and 2, respectively. On the remaining images, 45% of the artifacts did not affect the image quality, 30% of the artifacts seriously affected image quality, and 10% of the images were not useful for diagnosis.
Assessments of the postcontrast MR angiograms revealed no artifacts on 70% of images in dose group 2 and on 90% of images in dose group 2. The reader judged 1% of the images to be seriously compromised by artifacts. However, none of the artifacts precluded analysis of the postcontrast MR angiograms. At interpretation of DSA images, artifacts seriously affected the image quality on 3% of images, and 1% of the images were not useful for diagnosis.
Assessment of vessel delineation.—The quality of vessel delineation on the precontrast MR angiograms was rated as very satisfactory in 11% and 2%, satisfactory in 47% and 64%, unsatisfactory in 36% and 18%, and very unsatisfactory in 7% and 16% of images in dose groups 1 and 2, respectively. The quality of vessel delineation increased after gadodiamide injection such that the images were rated very satisfactory in 38% and 64%, satisfactory in 60% and 36%, and unsatisfactory in 2% and 0% of images in dose groups 1 and 2, respectively.
Confidence in diagnosis.—Confidence in diagnosis was judged by the reader to be low in 33% and 30%, moderate in 47% and 45%, and high in 20% and 25% of precontrast MR angiograms in dose groups 1 and 2, respectively. In contrast, confidence in diagnosis was high in 88% and 96% of postcontrast MR angiograms in the single- and triple-dose groups, respectively.
Diagnostic information from MR angiograms.—Compared with the precontrast MR angiograms, the postcontrast images, as judged by the investigators, provided more diagnostic information in 79% and 85%, less information in 8% and 2%, and equal information in 13% and 14% of patients in dose groups 1 and 2, respectively. Four postcontrast MR angiograms in dose group 1 provided less diagnostic information owing to problems that occurred during image acquisition (n = 2) and to mistiming of the bolus arrival and imaging delay (n = 2). In dose group 2, one postcontrast MR angiogram was inferior to the precontrast MR angiogram owing to acquisition problems.
In dose group 1, 43% of postcontrast MR angiograms provided new diagnostic information that influenced patient management: Three patients (6%) were scheduled for surgery after evaluation of the postcontrast MR angiograms; in 13 patients (24%), the planned surgery was canceled in favor of angioplasty; and in 15% of patients, more detailed information was available. In dose group 2, postcontrast MR angiography provided new diagnostic information in 50% of patients. Two of these patients (2%) were scheduled for surgery; angioplasty was recommended instead of surgery for 14 patients (27%); and/or in 21% of patients, more detailed information was obtained.
The reader detected more smaller vessels on 72% of postcontrast MR angiograms in the single-dose group. This effect was pronounced in the triple-dose group, in which 90% more smaller vessels were found.
In summary, postcontrast MR angiography resulted in a diagnosis modification in 70% and 72% of patients in dose groups 1 and 2, respectively (Table 7). There was no diagnostic gain in 8% and 2%, detection of additional stenoses in 24% and 42%, better vessel delineation in 66% and 71%, better definition of stenosis degree in 30% and 33%, exclusion of a stenosis in 8% and 14%, and confirmation of the diagnosis in 47% and 42% of patients in dose groups 1 and 2, respectively.
|
At precontrast MR angiography, no accessory renal arteries were detected in dose group 1, and three accessory renal arteries were detected in dose group 2. At postcontrast MR angiography, accessory renal arteries were seen in six patients in dose group 1: Five patients each had one accessory renal artery, and one patient had two accessory renal arteries. In dose group 2, one accessory renal artery each was found in eight patients, with two seen in two patients each.
In summary, injection of gadodiamide resulted in a higher accessory renal artery detection rate, regardless of the dose. The investigators detected more accessory renal arteries in the triple-dose group (18%) than in the single-dose group (12%).
Additional abnormalities.—The additional abnormalities found on the pre- and postcontrast MR angiograms and DSA images are summarized in Table 8. Compared with the precontrast MR angiographic findings, 13 and 19 additional findings were detected after gadodiamide injection in dose groups 1 and 2, respectively. However, DSA was superior to contrast-enhanced MR angiography in depicting additional abnormalities.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Safety
The results of the present study indicate that gadodiamide-enhanced MR angiography used in single and triple doses is safe and effective for assessing stenoses of the major abdominal arteries. The excellent safety profile of nonionic gadodiamide injection for numerous indications has already been proved, and the safety of triple-dose administration has been shown to be similar to that of single-dose administration (4,5). No serious adverse events were observed in the current trial. The overall frequency of adverse events after contrast-enhanced MR angiography was very low (2%), and it is unknown how or whether this low frequency was related to the gadodiamide injection. In a minority of patients (16%), mild discomfort (ie, sensations of heat, warmth, or coldness at the injection site) was noted in our study.
Contrast Index
The main effectiveness parameter for the comparison of the two dose groups analyzed in this study was the CIx of the most severe and proximal arterial stenosis in the abdomen. The multicenter design of the present trial included a wide range of MR angiographic techniques and resulted in considerable variation in CIx values among the study centers. Despite this strong testing center effect, the mean CIx of the region of interest was considerably higher after gadodiamide injection in all vascular segments. Because the increase in CIx correlated with the concentration used, the CIx values in the aorta and proximal and distal to the primary stenosis were significantly higher in the triple-dose group than in the single-dose group. In addition, the CNR was markedly increased by using a triple dose of contrast material in all four vascular regions of interest compared with a mild increase in the single-dose group. These findings have been reported by other authors (15,16), whose results showed that a double dose resulted in a substantial increase in aortic and renal arterial enhancement at contrast-enhanced MR angiography.
Degree of Stenosis
An important finding in our study was that the 95% CIs for the difference between DSA and postcontrast MR angiography in the assessment of degree of primary stenosis were smaller for the triple-dose group than for the single-dose group, and these CIs were high at precontrast MR angiography. The occlusions found at DSA were visualized also at MR angiography in most instances. Category 2 (70%–99%) stenoses, which are considered to be hemodynamically significant, were judged to be almost equal to DSA in patients who received the triple dose of contrast material. On the basis of these results, the triple dose tends to be more effective for assessment of the degree of arterial stenosis in the abdomen.
These results contradict the findings of recently published studies (10,13,17,18), in which excellent data were obtained by using sophisticated MR angiographic techniques and a single-dose bolus. An injection dose of 0.1–0.2 mmol/kg has proved to be sufficient for breath-hold gadolinium-enhanced MR angiography of the aorta and its major branches (13,18–23). Although there has been a tendency to use lower dose strategies with high-end MR units, the limitation of such examinations in several studies (18,24–26) has been the lack of correlation with conventional angiography. In addition, in the majority of these studies, different contrast material doses were not evaluated, and, therefore, a direct comparison with our results is difficult to perform. On the other hand, the results of our multicenter trial reflect the experiences at several sites, with use of a variety of MR imaging and MR angiographic techniques and equipment, with some lack of optimization compared with the results obtained in a single-center study.
The accuracy of MR angiography in the assessment of degree of primary stenosis also was improved by using a triple dose of contrast material. In dose group 2, the accuracy for a cutoff point of 50% stenosis and occlusion was superior to that achieved with the single dose and equal to that achieved for a cutoff point of 70% stenosis.
Overall Image Quality
The overall image quality scores, as obtained by using a visual analogue scale, were significantly higher at contrast-enhanced MR angiography compared with those at precontrast MR angiography. The triple-dose group had significantly higher visual analogue scale values than did the single-dose group, which is in agreement with a previous report on the use of a triple dose for MR angiography of non–central nervous system lesions (27).
Artifacts at MR Angiography and DSA
It is well known that conventional MR angiography may cause numerous flow-related artifacts, which eventually result in image degradation (28). In the present study, 85% of cases showed artifacts on precontrast MR angiograms. In these cases, image quality was seriously affected or did not allow interpretation of the images in more than 40% of patients. After gadodiamide injection, no artifacts were noted on 70% of the MR angiograms in the single-dose group and on 90% in the triple-dose group. Image quality was seriously affected owing to artifacts on only 1% of all the postcontrast MR angiograms. This observation confirms findings from other studies (1,13–26) in which contrast-enhanced MR angiography was reported to improve image quality substantially. Along with the reduction of artifacts, the quality of vessel delineation and the confidence in diagnosis were highly dependent on the administration and dose of gadodiamide.
Diagnostic Information from MR Angiograms
A substantial gain in diagnostic information was evident with the use of contrast-enhanced MR angiography compared with the information obtained with the precontrast examinations. About 80% of all postcontrast MR angiograms yielded more diagnostic information than the nonenhanced images, and this increase was positively related to the dose of contrast material administered. In nearly half of the patient population, postcontrast MR angiography provided new diagnostic information (43% and 50% gains in the single- and triple-dose groups, respectively) that influenced patient management, and again the gain was dose dependent. Compared with the precontrast examinations, the postcontrast studies facilitated a change in diagnosis in approximately 70% of all patients.
The additional advantages of using the triple-dose regimen were an improved rate of detecting collateral vessels in patients with occlusion and a higher rate of detecting accessory renal arteries and additional abnormalities (eg, aneurysms). In summary, the use of a triple dose could improve patient treatment in 7% of cases but increase the costs for contrast material by a factor of three.
Several limitations of the present trial should be addressed. The multicenter design of this trial included a wide range of MR angiographic techniques and resulted in a strong testing center effect, with considerable variation in the values measured among the study centers. Furthermore, only 64 precontrast MR angiograms were assessable by the independent reader. The comparison between precontrast MR angiography and DSA with respect to degree of stenosis was therefore biased because a considerable number of MR images were excluded from analysis owing to very poor image quality. However, because nonenhanced MR angiography of the abdominal vessels has been of limited value in the evaluation of vascular disease, this drawback may be of limited clinical relevance.
In conclusion, the results reported herein reflect the experiences at several sites, with use of different equipment and a wide range of MR angiographic techniques in a large number of patients. Both doses of gadodiamide—0.1 and 0.3 mmol/kg—had an excellent safety profile. High agreement between MR angiography and DSA can be achieved in the assessment of stenoses by using both a single and a triple dose. However, the triple dose may have advantages. Results of analysis of the CIx as the primary effectiveness parameter showed significant testing center effects, but the overall mean CIx was significantly higher in the triple-dose group, and the dose effect was similar across all centers—that is, no interactions between dose and centers were observed. Image quality, vessel delineation, and confidence in diagnosis increased in the triple-dose group, and this resulted in more diagnostic information than that obtained by using a single dose.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CIx = contrast index, CNR = contrast-to-noise ratio, DSA = digital subtraction angiography
Author contributions: Guarantors of integrity of entire study, S.A.T., M.M.; study concepts and design, M.M.; literature research, S.A.T., M.M.; clinical studies, data acquisition, manuscript editing, and manuscript revision/review, S.A.T., A.C., F.H.D.O., R.F.D., C.G., A.G.J., P.K., C.L., C.N.L., L.M.B., J.P.d.C., J.P.P., V.M.S., T.V.; data analysis/interpretation, J.P.C.; statistical analysis, M.M.; manuscript preparation, M.M., S.A.T.; manuscript definition of intellectual content, S.A.T.; manuscript final version approval, S.A.T., M.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Remy-Jardin, J. Bahepar, J.-J. Lafitte, P. Dequiedt, O. Ertzbischoff, J. Bruzzi, V. Delannoy-Deken, A. Duhamel, and J. Remy Multi-Detector Row CT Angiography of Pulmonary Circulation with Gadolinium-based Contrast Agents: Prospective Evaluation in 60 Patients Radiology, March 1, 2006; 238(3): 1022 - 1035. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Schaefer, F. P. Boudghene, H. J. Brambs, M. Bret-Zurita, J. L. Caniego, R. A. Coulden, H. B. Gehl, R. Hammerstingl, A. Huber, R. J. Mendez, 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, March 1, 2006; 238(3): 827 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Chicoskie and R. Tello Gadolinium-Enhanced MDCT Angiography of the Abdomen: Feasibility and Limitations Am. J. Roentgenol., June 1, 2005; 184(6): 1821 - 1828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.D. Wolff, J. Schwitter, R. Coulden, M.G. Friedrich, D.A. Bluemke, R.W. Biederman, E.T. Martin, A.J. Lansky, F. Kashanian, T.K.F. Foo, et al. Myocardial First-Pass Perfusion Magnetic Resonance Imaging: A Multicenter Dose-Ranging Study Circulation, August 10, 2004; 110(6): 732 - 737. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kluge, T. Dill, O. Ekinci, J. Hansel, C. Hamm, H. F. Pitschner, and G. Bachmann Decreased Pulmonary Perfusion in Pulmonary Vein Stenosis After Radiofrequency Ablation: Assessment With Dynamic Magnetic Resonance Perfusion Imaging Chest, August 1, 2004; 126(2): 428 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Willmann, S. Wildermuth, T. Pfammatter, J. E. Roos, B. Seifert, P. R. Hilfiker, B. Marincek, and D. Weishaupt Aortoiliac and Renal Arteries: Prospective Intraindividual Comparison of Contrast-enhanced Three-dimensional MR Angiography and Multi-Detector Row CT Angiography Radiology, March 1, 2003; 226(3): 798 - 811. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||
