| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of Regensburg, Klinikum, Franz Josef Strauss Allee 11, Regensburg 93042, Germany.
Received June 7, 2004;
accepted after revision November 11, 2004.
Address correspondence to N. Zorger.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifteen patients underwent digital subtraction angiography and intraarterial MR angiography before and after balloon angioplasty. For intraarterial MR angiography, 30 mL of diluted contrast agent (5 mL of gadodiamide diluted in 55 mL of 0.9% saline solution) was injected through a sheath in the superficial femoral artery using a flow rate of 2.5 mL/sec. A 3D gradient-echo sequence was performed. Four independent blinded observers assessed differences in the quantitative measurement of stenoses and localization of lesions between digital subtraction angiography and intraarterial MR angiography. The overall impression of the intraarterial MR angiography images was documented on a 4-point scale (1 = excellent, 4 = poor). Interobserver variability was calculated.
RESULTS. Intraarterial MR angiography from the upper leg to the
trifurcation was feasible in all 30 examinations with a mean overall
impression of all segments of 1.3 (SD, 0.68). For the detection of significant
stenoses (
50% stenosis), the overall sensitivity and specificity for the
femoropopliteal and crural vessels were 92.4% and 91.7% and 91.9% and 87.8%,
respectively. For the complete leg, sensitivity and specificity were 92.2% and
88.6%, respectively. Interobserver variability for intraarterial MR
angiography of the crural vessels exceeded that of the femoropopliteal
arteries.
CONCLUSION. Intraarterial MR angiography of the infrainguinal arteries is feasible in humans using injections of diluted contrast agent at concentrations as low as 8%. It has a high sensitivity for detecting stenoses and an acceptable interobserver variability.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The objectives of this study were to show the feasibility of intraarterial MR angiography of the infrainguinal arteries and to determine the accuracy of intraarterial MR angiography for the detection of femoral artery stenoses before and after balloon angioplasty using digital subtraction angiography as the standard of reference. We also aimed to show that a low dose of contrast agent could be used with this new technique.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
Inclusion and Exclusion Criteria
Inclusion criteria were focal symptomatic arteriosclerotic lesions of the
infrainguinal arteries suitable for angioplasty. Exclusion criteria were
occlusions or stenoses that extended more than 8 cm and contraindications for
MR tomography such as pacemakers or incompatible ferromagnetic implants.
Study Protocol
At the beginning of the procedure, an 11-cm 5-French sheath was introduced
antegrade after puncture of the ipsilateral common femoral artery with an
18-gauge needle (Surflo, Terumo) without radiographic guidance. An initial
digital subtraction angiography examination of the whole leg with
half-strength contrast material (Ultravist 150 [iopromide], Schering) was
performed. After the administration of heparin (5,000 IU) through the sheath,
the patient was transferred to a 1.5-T MR scanner (Magnetom Sonata, Siemens
Medical Solutions) with a 40 mT/m gradient capability and a 200 mT/m/sec
maximum slew rate for intraarterial MR angiography using a dedicated coil
system. A 3D gradient-echo sequence with elliptical centric phase encoding and
with a subtraction technique to suppress background tissue was used. The
scanning parameters for 3D MR angiography are presented in
Table 2.
|
Because of intraarterial application, nonionic gadodiamide (Omniscan,
Amersham) was used. An injection protocol was followed that adheres to a
published procedure based on the following equation, which characterizes the
injected concentration, (Gd)inj, as a function of blood
flow rate, Q; injection rate, Qinj; and the
desired blood concentration, (Gd)arterial
[2]:
 |
Preliminary studies in animals showed that an arterial gadolinium concentration, (Gd)arterial, of 16.2 mmol/L was optimal for MR angiography [1, 2]. If the blood volume flow rate (Q) of the superficial femoral artery is 3.3 mL/sec [3], the injection rate (Qinj) is 2.5 mL/sec, and the arterial concentration of injected gadolinium, (Gd)arterial, is 16.2 mmol/L, then the resulting concentration of injected gadolinium, (Gd)inj, is approximately 37.6 mmol/L (7.5%). Our preexaminations showed that a mix of 5 mL of gadolinium (0.5 mol/L) and 55 mL of 0.9% saline solution (41.7 mmol/L = 8.3% concentration) results in a good contrast level, which is easy to achieve in the clinical routine.
|
|
|
After the MR angiography examination, the patient was transferred back to the angiography room and the intervention was continued by passing the lesion with a 0.035-inch guidewire (Radiofocus, Terumo) and performing balloon angioplasty over a period of 3 min [4]. Finally, digital subtraction angiography and further intraarterial MR angiography of the peripheral arteries to the level of the lower leg were performed.
Analysis
Statistical analysis was performed with SPSS software (version 10.0,
Statistical Package for the Social Sciences) for Windows (Microsoft). MR
angiograms and digital subtraction angiography images, including source images
and maximum-intensity-projection images, were transferred in a random order to
a computer workstation (Magic-View, Siemens Medical Solutions). The images
obtained before and after the intervention were evaluated by four blinded
observers with extensive experience in peripheral angiography. No other
clinical information was provided with the images.
For image analysis, the individual vasculature was divided into the following segments: superficial femoral artery, popliteal artery, tibioperoneal trunk, proximal anterior tibial artery, proximal posterior tibial artery, and proximal peroneal artery. The grading system of the American College of Radiology multiinstitutional trial of peripheral MR angiography was used [5]: 0 was considered normal; 1, minimal stenosis of less than 50%; 2, one lesion with 50% or greater stenosis; 3, more than one lesion with 50% or greater stenosis; and 4, occlusion. Manual quantitative measurements at the stenosis and the proximal uninvolved artery were performed. The proximal uninvolved artery was used to represent normal vessel caliber. Diameter percentage stenosis was calculated as follows: [1 – (stenosis caliber/normal caliber) x 100]. Image quality for the evaluation of stenosis was subjectively graded as 1, excellent; 2, good; 3, fair; or 4, poor. Digital subtraction angiography results were considered to be the reference standard, with significant stenosis indicated by a score of greater than 1 and occlusion indicated by a score of 4. In the case of a disagreement among the four observers evaluating digital subtraction angiography, a consensus decision was ultimately agreed on. The sensitivity and specificity were computed for descriptive purposes using all interpretations.
To determine the interobserver variability in assessing the lesion detection with each imaging sequence, we used Cohen's kappa value to measure the degree of agreement among the four observers. Interobserver agreement was considered slight at a value equal to or less than 0.2; fair, 0.21–0.40; moderate, 0.41–0.60; substantial, 0.61–0.80; or almost perfect, 0.81–1.00 [6]. The two-tailed Wilcoxon's rank sum test was applied for the comparison of digital subtraction angiography versus intraarterial MR angiography. The level for a statistically significant difference was set at a p value of less than 0.05.
Complications were defined according to the classification system of the Standards of Practice Committee of the Society of Cardiovascular and Interventional Radiology [7] in which complications are categorized by outcome. Minor complications were defined as A, no therapy, no consequence; or B, nominal therapy, no consequence, including overnight admission for observation only. Major complications were defined as C, require therapy, minor hospitalization (< 48 hr); D, require major therapy, unplanned increase in level of care, prolonged hospitalization (> 48 hr); E, have permanent adverse sequelae; or F, result in death.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Correlation with Angiography
All 180 segments (15 patients x 6 segments x 2 examinations)
were examined on both intraarterial MR angiography and digital subtraction
angiography. In digital subtraction angiography, 116 (64.4%) of these segments
were judged to be patent (grade 0–1), 32 (17.8%) were found to have
grades 2 and 3 lesions (> 50% stenosis), and 32 (17.8%) were considered
occluded (grade 4).
The mean values for the femoropopliteal vessels were 92.4% for sensitivity and 91.7% for specificity for all four observers. The mean values for sensitivity and specificity for the crural vessels were 91.9% and 87.8%, respectively. For femoropopliteal and crural vessels, the mean values for sensitivity and specificity were 92.2% and 88.6%. The individual results of the four observers are shown in Table 3.
|
The kappa values for each pair of observers ranged from fair to substantial for both digital subtraction angiography and intraarterial MR angiography. For intraarterial MR angiography, the interobserver variability was less for the femoropopliteal vessels than for the crural arteries (mean: 0.567 vs 0.459, respectively). The difference between the kappa values of all observers for digital subtraction angiography and intraarterial MR angiography was statistically significant when assessed with the two-tailed Wilcoxon's rank sum test for the femoropopliteal and crural arteries, respectively (p < 0.046 and p < 0.016). The kappa values of interobserver agreement for the evaluation of the MR angiograms are listed in Table 4.
|
Complications
In digital subtraction angiography and intraarterial MR angiography, an
examination to the level of the lower leg was performed to exclude peripheral
embolization. However, no such complication was documented for either method.
A complication of the puncture site, a groin hematoma, occurred in one
patient. Surgical treatment was not necessary (type A). No major complications
occurred. Angioplasty was technically successful in all patients.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our protocol, we used a solution of 5 mL of gadolinium (0.5 mol/L) diluted with 55 mL of 0.9% saline solution, which is equivalent to an injected gadolinium concentration of 41.7 mmol/L (8.3% gadolinium). Our rationale for using an intraarterial concentration of approximately 3–4% (injection rate, 2.5 mL/sec; injected gadolinium concentration, 8.3%) was to make a compromise on the reduction of contrast agent dose and the results of animal and in vitro experiments showing an optimal arterial gadolinium concentration of 1–3% [1, 2]. Using a slower injection rate of 1 mL/sec, Omary et al. [1] injected 30 mL of 15–22% gadolinium concentration, equivalent to an injected concentration of 75–110 mmol/L, which was more than double the concentration we used. Thus, our results suggest that, under the described parameters, a lower injected gadolinium concentration is possible, a finding that is supported by previous studies showing that a wide range of arterial gadolinium concentrations is acceptable while still maintaining sufficient signal-to-noise ratio [1, 2, 9, 10]. Confirming this thesis, Green et al. [11] found an optimal injected contrast agent concentration of 6% for MR coronary angiography in dogs, which is comparable to the gadolinium concentration we used.
Compared with digital subtraction angiography, intraarterial MR angiography achieved mean values of 92.2% for sensitivity and 88.6% for specificity, which are similar to those values with IV MR angiography. For IV MR angiography of the infrainguinal vessels, sensitivities from 75% to 100% and specificities from 95.8% to 98% [12–16] have been reported. Omary et al. [1] supposed a lower accuracy of intraarterial MR angiography of the renal arteries; however, they found no significant differences in their small sample size compared with digital subtraction angiography [1].
A relationship between the location of the stenosis and diagnostic confidence was seen in our study. Particularly, stenoses in vessels of the lower legs seemed to be detected with less accuracy, with a mean value for specificity of 87.8% compared with 91.7% for the femoropopliteal arteries. One reason for the slightly worse detection of hemodynamically significant stenoses in the lower legs compared with the upper legs could be the relatively large effective slice thickness. Despite technical developments, the voxel size in the lower leg seems to be too large compared with the small diameter of these arteries [15]. Another reason could be the reduced concentration of contrast material traveling from the sheath in the femoral artery to the lower leg combined with a small vessel diameter. Finally, sensitivity and specificity are probably lower than historical MR angiography rates due to selection bias. All patients included in this study had hemodynamically significant disease. If a random patient population sample was used, then eventually the sensitivity and specificity would be higher. However, although a different grading of stenoses compared with the digital subtraction angiography reference standard was achieved, this kind of procedure would be sufficient for treatment planning and control while performing MR angioplasty of the femoropopliteal arteries.
For multiple contrast injections to be administered during MR angioplasty, contrast agent must be conserved so that background tissue enhancement is minimized and limits mandated by the U.S. Food and Drug Administration are not exceeded. Special techniques that reduce single contrast agent dosage would be an advantage, providing the operator with a larger number of images for the procedure. In our study, for one acquisition of a defined vascular region only, 2.5 mL of nonionic 0.5 mol/L of gadodiamide (Omniscan, Amersham) diluted with 27.5 mL of saline solution had to be injected intraarterially through the sheath. Based on an 80-kg patient, theoretically 19 separate acquisitions of the treated vascular region are possible (injecting a maximum dose of 0.3 mmoL/kg = 0.6 mL gadodiamide per kilogram of body weight), enough for most interventional procedures.
Other studies report significant reduction of intraarterial contrast agent injection by limiting injection time to only part of the imaging acquisition time. Excellent images of all vessels were obtained at 50% injection duration in one study [17], despite several limitations of this study using an in vitro vascular phantom without pulsatile blood flow and no background tissue mimic. Omary et al. [1] also reported that intraarterial MR angiography used smaller doses of injected gadolinium (mean, 5.6 mL) than IV MR angiography (mean, 9 mL) in five pigs.
Despite the many advantages of using intraarterial injection of gadolinium chelates in angiography and interventional radiology [8, 18, 19], there are critical reports as well. Nyman et al. [20] discourage the use of gadolinium for digital subtraction angiography because of the lack of randomized studies, potentially toxic hyperosmolality of gadolinium, and the need for high doses of up to 50 mL of 0.5 mol/L gadolinium chelates. With respect to the dosage, intraarterial MR angiography in our study was performed with less gadolinium than that needed for fluoroscopy. Subsequent potential toxicity as a function of the concentration of gadolinium chelates would not be significant in MR angiography.
In conclusion, we showed that intraarterial MR angiography of the infrainguinal arteries of humans is feasible using injections of diluted contrast agent with concentrations as low as 8%. We confirmed the accuracy of intraarterial MR angiography for detecting stenoses and treatment success after angioplasty despite moderately higher interobserver variability compared with digital subtraction angiography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Poschenrieder, O. W. Hamer, T. Herold, T. Schleicher, I. Borisch, S. Feuerbach, and N. Zorger Diagnostic Accuracy of Intraarterial and IV MR Angiography for the Detection of Stenoses of the Infrainguinal Arteries Am. J. Roentgenol., January 1, 2009; 192(1): 117 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Paetzel, N Zorger, O W Hamer, J Seitz, T Schleicher, S Feuerbach, W R Nitz, M Lenhart, and T Herold Intra-arterial MR angiography in the iliac system: initial clinical experience with 25 patients. Br. J. Radiol., April 1, 2006; 79(940): 298 - 302. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
