| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Institute for Clinical Radiology (O.A.M., J.R., C.W., M.F.R., S.O.S.), Division of Vascular Surgery (B.S.), and Institute of Medical Informatics, Biometry, and Epidemiology (U.S.), Ludwig-Maximilians-University, Marchioninistr 15, 81377 Munich, Germany; and Harvard Center for Risk Analysis, Harvard School of Public Health, Boston, Mass (U.S.). Received October 16, 2003; revision requested January 12, 2004; final revision received July 13; accepted August 16. Address correspondence to O.A.M. (e-mail: oliver.meissner@med.uni-muenchen.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The study was approved by the institutional review board, and written consent was obtained from all patients. Pretreatment DSA and hybrid MR angiography were performed in 19 consecutive patients (15 men, four women; mean age, 69.8 years; range, 44–86 years). Hybrid MR angiography included submillimeter dual-phase three-dimensional gadolinium-enhanced MR angiography in lower calf and foot, and four-station bolus-chase MR angiography in pelvis, thigh, and upper calf. Three readers identified the target lesion and inflow and outflow segments and determined treatment (bypass graft placement, percutaneous transluminal angioplasty, conservative management, amputation). Results of interobserver and intermethod comparisons were expressed as percentage of agreement and 95% confidence interval (CI).
RESULTS: On hybrid MR angiograms, no substantial venous overlay was present and image quality was excellent or adequate in 18 (95%) of 19 limbs. Readers 1, 2, and 3 selected the identical target lesion on the DSA image and the MR angiogram in 18, 17, and 18 of 18 comparable limbs, respectively. Mean percentage of agreement for readers 1 and 3 was 100% (95% CI: 81%, 100%) and for reader 2 was 94% (95% CI: 73%, 100%). Agreement of all three readers was superior with use of MR angiography for determination of inflow segments (13 [72%] of 18 limbs) and outflow segments (17 [94%] of 18 limbs), compared with agreement with use of DSA (13 [68%] of 19 inflow segments, 10 [53%] of 19 outflow segments). Agreement in therapy decisions was higher with DSA (15 [79%] of 19) than with MR angiography (11 [61%] of 18).
CONCLUSION: Preliminary data strongly support the combination of submillimeter dual-phase MR angiography in lower calf and foot with four-station bolus-chase MR angiography to extend the utility of MR angiography to patients with critical limb ischemia.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Treatment planning for patients who are being evaluated for limb salvage surgery requires high-quality images of vessels in the calf and foot. Information about the number, length, and severity of vascular lesions, as well as the inflow and outflow segments, especially in patients in whom infrageniculate bypass surgery may be considered, is essential. The assessment of the arterial runoff prior to therapeutic procedures has traditionally been performed with intraarterial digital subtraction angiography (DSA). DSA is, however, an invasive procedure with an additional risk of morbidity and mortality. Furthermore, the use of iodinated contrast material may harm these critically ill patients, who often suffer from additional renal insufficiency (3,4).
In recent years, contrast material–enhanced magnetic resonance (MR) angiography has emerged as a compelling alternative to intraarterial DSA (5–14). Its noninvasive nature makes peripheral arterial MR angiography an especially attractive alternative. Continuous table movement enables examination of the arteries from the abdominal aorta to the foot in three to four stations, with excellent results in healthy volunteers and patients with peripheral arterial disease of Rutherford-Becker grade I or II (6,7,15,16).
However, venous overlay and motion artifacts, especially in the infrageniculate region, may hinder the accurate depiction of arteries distal to the knee in patients with critical limb ischemia and make therapeutic decisions impossible. Khilnani et al (17) reported a high degree of correlation between their findings with a combined two-dimensional and three-dimensional (3D) MR angiography protocol and findings with DSA for determining the lesion to treat and for visualizing both the inflow and outflow segments in case bypass surgery is necessary.
The purpose of our study was to compare a hybrid MR angiography protocol with selective DSA in patients with critical limb ischemia.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
From July 2002 to February 2003, a total of 21 consecutive patients were eligible for prospective enrollment in the study. Two patients met exclusion criteria (urgent need for treatment, n = 1; cardiac pacemaker, n = 1). The remaining 19 patients (15 men, four women; mean age, 69.8 years; age range, 44–86 years) with 19 symptomatic legs were entered into the study prospectively. At presentation in the Division of Vascular Surgery at Ludwig-Maximilians-University, Munich, Germany, nine patients (47%) had Rutherford stage IV disease. Rutherford stage V disease was found in 10 patients (53%). Thirteen patients (68%) had additional diabetes mellitus with severe microangiopathy. Only symptomatic legs considered for revascularization were included in the analysis, even if an asymptomatic contralateral extremity was examined in its entirety. To determine the performance of the hybrid MR angiography protocol for evaluation of patients in difficult examination conditions (eg, patients with pain at rest, which might result in image artifacts from inadvertent movement, or those with an arteriovenous shunt that causes early venous backflow), all patients were included in the final data analysis. Complete hybrid MR angiography consisted of dual-phase 3D gadolinium-enhanced MR angiography of the lower calf and foot, combined with four-station bolus-chase MR angiography for evaluation of the area from the infrarenal aorta to the ankle. Complete DSA consisted of selective intraarterial angiography in the clinically relevant extremity. DSA and MR angiography were carried out successfully in all 19 patients. In all patients, MR angiography was performed before DSA, and the mean interval between the two angiographic examinations was 1.5 days.
The research study was approved by the institutional review board, and oral and written consent was obtained from all patients.
Hybrid MR Angiography Protocol
MR angiography was performed by using a 1.5-T whole-body MR imaging unit (Magnetom Sonata; Siemens Medical Solutions, Erlangen, Germany). The patient was placed in the supine position with the feet positioned to enter the magnet first. The lower part of the legs was immobilized with foam cushions and stabilizing straps and placed into a vascular array. The precision of contrast material injections was ensured by using an MR-compatible power injector (Spectris; Medrad, Pittsburgh, Pa). The exact protocol for test bolus and timing determination for the MR angiographic studies was described in detail previously (18,19).
Dual-phase 3D contrast-enhanced MR angiography in lower calf and foot.—3D fast spoiled gradient-echo acquisitions were performed with the following parameters: 3.5/1.2 (repetition time msec/echo time msec), flip angle of 20°, field of view of 360 mm2, voxel size of 1.4 x 0.7 x 0.9 mm, matrix of 512 x 240 pixels, and receiver bandwidth of 360 Hz per pixel. To enable the acquisition of central k-space prior to venous enhancement, elliptic centric reordering of k-space was used. For exact timing of contrast agent arrival in the calf, a test bolus of 2 mL of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) was administered with a power injector at a rate of 1.5 mL/sec through an 18–20-gauge intravenous catheter placed in an antecubital vein, followed by 20 mL of normal saline administered at the same rate to facilitate contrast material passage into the central circulation. The section for imaging of the test bolus arrival was positioned in the popliteal artery.
For MR angiography, three continuous acquisitions were performed. The first acquisition, which was performed without contrast material, served as a mask for subsequent subtraction and was immediately followed by contrast-enhanced arterial and venous phase imaging. Arterial phase images were obtained after injection of 15 mL of gadopentetate dimeglumine at the same rate and with the same volume of saline chaser as were used for the test bolus injection. The delay between the contrast agent injection and the start of the MR angiographic acquisition was based on the time of arrival of the contrast agent bolus, which was determined in the test bolus measurement. Arterial phase imaging was followed by a second contrast-enhanced acquisition for depiction of small arterial vessels with delayed enhancement during the venous phase.
3D bolus-chase MR angiography.—Three-station bolus-chase MR angiography was performed with multiphasic 3D fast spoiled gradient-echo sequences by using a vascular surface coil, automatic table movement, and mask subtraction. For signal reception, a 12-element vascular coil was used in combination with a large field-of-view extender and two additional dual-element surface coils, which allowed coverage of the entire volume from the diaphragm to the toes. The 3D MR imaging parameters were as follows: 3.4/1.2, flip angle of 20°, field of view of 380 mm2, voxel size of 1.5 x 0.7 x 1.3 mm, matrix of 512 x 240 pixels, and receiver bandwidth of 540 Hz per pixel. To minimize venous overlay, k-space acquisition in the station from the aorta to the pelvis was sequential, while that in the station from the thighs and calves was centric. Prior to bolus-chase MR angiography, a test bolus sequence was applied at the level of the renal arteries, and 2 mL of gadopentetate dimeglumine was injected at a rate of 2 mL/sec. The three imaging stations were (a) the abdomen and pelvis, (b) the upper thighs to the knees, and (c) the knees and calves. To obtain a mask for subsequent subtraction, the 3D MR angiographic sequence was applied in each station prior to injection of 25 mL of contrast material, which was followed by a 20-mL normal saline chaser, for contrast-enhanced MR angiography. Automatic subtraction of the contrast-enhanced image data from the mask image data was performed at the MR imager console. The raw data were reconstructed into 3D maximum intensity projection images, and all data sets, as well as the raw data, were stored in the digital archive.
The duration of the entire hybrid MR angiographic examination was 50–90 minutes (mean, 63 minutes).
A flowchart of the hybrid MR angiography protocol is shown in Figure 1.
|
Image Analysis
Clinical decision making in regard to therapy was performed in consensus by the interventional radiologists and the vascular surgeons in charge, who were not part of the reading team. The therapy decision was based on the results of DSA image interpretation alone. The image reading for this study was performed separately from clinical decision making and was retrospective. Otherwise, an image analysis performed in a randomized and blinded fashion for comparison of images acquired with two different modalities within a mean interval of 1.5 days would not have been possible without the readers being biased. For our research study, the MR angiographic and DSA images were randomly organized into a work list and displayed at a workstation (IMPAX DS 3000; Agfa-Gevaert, Mortsel, Belgium). Images from each patient were randomly displayed by an independent operator (C.W.), without the readers being able to see or access the work list. On the workstation monitor, the cell containing the patient’s name, medical record number, and examination details was deactivated. MR angiograms and DSA images from the same patient were interpreted in different order and on different days at least 2 weeks apart. According to the method used in the study by Khilnani et al (17), each extremity for which therapy was being considered was divided into the following 11 vascular segments: 1, common femoral artery; 2, superficial femoral artery; 3, above-knee popliteal artery (including the middle genicular branch); 4, below-knee popliteal artery; 5, tibioperoneal trunk and upper half of peroneal artery; 6, lower half of peroneal artery; 7, upper half of anterior tibial artery; 8, lower half of anterior tibial artery; 9, upper half of posterior tibial artery; 10, lower half of posterior tibial artery; and 11, dorsal artery of the foot, either plantar artery, and bypass graft (if present). Three fellowship-trained cardiovascular and interventional radiologists (O.A.M., J.R., S.O.S.), each with more than 6 years of experience in DSA and MR angiographic image analysis, independently evaluated the 38 angiograms. The readers were provided with a grading sheet for each DSA and each MR angiographic image. The readers, who were blinded to clinical information and to the DSA or MR angiographic results, were asked to indicate the target lesion, any additional relevant stenosis, inflow and outflow segments in case of necessary bypass surgery, therapy decision, and image quality. The target lesion was defined as the hemodynamically most important lesion, that is, the one that had to be treated first in order to improve critical ischemia. Relevant stenosis was defined as any additional lesion that had the potential to further affect distal runoff and that could also be treated to further improve revascularization. Vascular lesions were graded as follows: 1, stenosis of less than 50% of luminal diameter; 2, stenosis of more than 50% of luminal diameter; or 3, complete occlusion. As in the study of Khilnani et al (17), the inflow segment was defined as the segment that was free of significant (>50%) stenosis and was located immediately above the segment that contained the target lesion. The outflow segment was defined as the segment immediately below the target lesion. The quality of depiction of the arterial anatomy was graded as inadequate (impossible to confidently determine a treatment plan), intermediate (sufficient for treatment planning), or excellent (more than sufficient for treatment planning). Contamination of arterial phase images by venous enhancement was graded as follows: 1, major (diagnostic evaluation not possible); 2, mild (diagnostic evaluation not affected); or 3, none (no visible venous contamination). To evaluate the amount of venous contamination, the arterial phase images from the preceding section (calf) were compared with those from the standard four-station bolus-chase MR angiography protocol.
Each reader was instructed to provide a therapy decision based on the DSA and MR angiographic examinations. Therapeutic options included endovascular treatment (ie, percutaneous transluminal angioplasty with or without stent placement), bypass surgery, and, in cases in which vascular repair was impossible, medical management with vasodilative agents (eg, alprostadil) or amputation.
Statistical Analysis
The statistical design and subsequent analysis of this study were performed by a dedicated biostatistician (U.S.). Investigated outcomes for interobserver and intermethod comparisons comprised identification of the target lesion, additional relevant stenosis, inflow segment and outflow segment, therapeutic decision, and image quality. Interobserver agreement is described as the fraction of patients for whom all three readers agreed regarding the investigated outcome. In addition, the fraction for the agreement of at least two readers is presented. Agreement was determined for both DSA and MR angiography separately. For each reader, intermethod agreement is described as the fraction of patients in whom the results of DSA and MR angiography were concordant. Overall mean agreement between methods was expressed as the mean of the three readers’ agreement fractions (mean percentage). As this was an exploratory study with multiple comparisons, we present exact binomial 95% confidence intervals (CIs) for proportions, rather than P values, to describe uncertainty in our data. Statistical analyses were performed by using two statistical software packages (SPSS, version 12.0, SPSS, Chicago, Ill; Stata, version 7.0, Stata, College Station, Tex).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Image Quality Evaluation
MR angiography yielded images with excellent or intermediate overall quality in 18 (95%) of the 19 limbs examined (mean for the three readers). There was only one patient for whom the quality of the MR angiogram was consistently graded as inadequate by all three readers. The MR angiogram for this patient was excluded from the intermethod and MR angiographic interobserver comparisons, although the patient’s DSA images were included in the DSA interobserver comparison. Therefore, the denominator for the comparison between MR angiography and DSA is 18. No grade 1 venous contamination was found on the arterial phase images from the MR angiographic examination of the lower calf and midfoot prior to three-station bolus-chase MR angiography in any of the 144 vascular segments in the 18 limbs available for evaluation. Venous contamination of grade 2 was found in 23 (16%) of 144 vascular segments, and that of grade 3, in 112 (78%) of 144 segments (mean for the three readers). Venous contamination on the arterial phase images obtained with the standard four-station bolus-chase MR angiography protocol was grade 1 in 109 (76%) of 144 vascular segments, grade 2 in 19 (13%) of 144 segments, and grade 3 in 16 (11%) of 144 segments (mean for the three readers). Figure 2 illustrates the problem of venous contamination on images obtained in the lower calf and midfoot with standard continuous four-station bolus-chase MR angiography, as opposed to hybrid MR angiography.
|
|
|
|
|
|
On the basis of DSA images in 10 (53%) of 19 patients, all three readers agreed in the determination of relevant stenoses, and on the basis of DSA images in 18 (95%) of 19 patients, at least two readers agreed. On the basis of MR angiograms in 11 (61%) of 18 patients, all three readers agreed in the determination of relevant stenoses, and on the basis of MR angiograms in 17 (94%) of 18 patients, at least two readers agreed.
Inflow Segment Determination
Although overall image quality of MR angiograms was graded as excellent or intermediate in 18 limbs, in one patient no inflow segment determination was possible for two of the three readers. Readers 1, 2, and 3 selected the identical segment for inflow on both the DSA image and the MR angiogram in 15, 12, and 11 of 17 cases, respectively. The mean percentage of agreement for reader 1 was 83% (95% CI: 59%, 96%), that for reader 2 was 64% (95% CI: 38%, 86%), and that for reader 3 was 71% (95% CI: 44%, 90%) (Table 1).
On the basis of DSA images, all three readers agreed in inflow segment determination in 13 (68%) of 19 limbs, and at least two readers agreed in inflow segment determination in 18 (95%) of 19. On the basis of MR angiograms, all three readers agreed in inflow segment determination in 13 (72%) of 18 limbs, and at least two readers agreed in inflow segment determination in 16 (89%) of 18.
Outflow Segment Determination
Determinability of outflow segments for MR angiography and DSA differed between readers, and this difference led to reduced denominators in the comparison between the methods. The three readers selected the identical segment for outflow on both the DSA image and the MR angiogram in nine of 14 (reader 1), 10 of 14 (reader 2), and 14 of 16 (reader 3) cases. Mean percentage of agreement for reader 1 was 64% (95% CI: 34%, 87%), that for reader 2 was 71% (95% CI: 42%, 92%), and that for reader 3 was 88% (95% CI: 61%, 98%) (Table 1).
All three readers agreed in outflow segment determination on DSA images in 10 (53%) of the 19 limbs, and at least two readers agreed in all cases (100%, 19 of 19). All three readers agreed in outflow segment determination on MR angiograms in 17 (94%) of 18 limbs, and at least two readers agreed in all cases (100%, 18 of 18).
Therapy Decision
The three readers selected the identical therapy decision with DSA and MR angiography in 12 (reader 1), 12 (reader 2), and 14 (reader 3) of the 18 comparisons. Mean percentage of agreement for readers 1 and 2 was 67% (95% CI: 41%, 87%), and that for reader 3 was 78% (95% CI: 52%, 94%) (Table 1). In one patient, all three readers were unable to determine the therapy decision because of motion artifacts on the images. Mean agreement between DSA and MR angiography with regard to the therapy decision was 53% (10 of 19) for all three readers and 74% (14 of 19) for at least two readers. With use of DSA images, all three readers agreed on the therapy decision in 15 (79%) of 19 cases, and at least two readers agreed in all cases (100%, 19 of 19). With use of hybrid MR angiograms, all three readers agreed on the therapy decision in 11 (61%) of 18 cases, and at least two readers agreed in all cases (100%, 18 of 18). Table 2 provides an overview of target lesion determinations and therapy decisions based on DSA findings and hybrid MR angiographic findings made by each of the three readers.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Given the continued improvement in vascular surgical and interventional techniques, it is now possible to successfully bypass stenoses and occlusions in the crural and/or pedal circulation in patients with critical limb ischemia. The increasing use of small-vessel balloon catheters makes percutaneous transluminal angioplasty possible even in arteries in the lower calf and foot (eg, segments of the below-knee popliteal artery). Both vascular surgeons and interventional radiologists demand unimpeded visualization of a stenosis in arterial structures in the lower extremity and identification of docking sites for a potential bypass graft.
Patients with critical limb ischemia represent a challenge at pretherapeutic diagnostic imaging. Because of their often substantially reduced general condition and frequent pain at rest, it is difficult to perform necessary diagnostic procedures or to obtain multiple examinations. Intraarterial DSA carries an additional risk because of the use of potentially nephrotoxic contrast material and the invasive nature of the procedure (3,4,21). Gangrenous changes with early arteriovenous shunts result in rapid venous backflow and often substantially hamper evaluation of the arteries in the lower extremity.
Because of limitations intrinsic to the method, such as rapid venous overlay and limited spatial and temporal resolution, MR angiography until now has played only a minor role in the setting of critical limb ischemia. The hybrid MR angiography protocol used in this study permitted a high-quality and low-risk examination of patients with critical limb ischemia, because of its noninvasiveness, the lack of potentially nephrotoxic contrast agents, and the minimal rate of allergic reaction. All patients in our study group experienced pain at rest. Nonetheless, motion artifacts sufficiently severe to preclude evaluation for therapeutic planning were encountered in only one case. The specially designed multiple-element vascular coil permits extensive anatomic coverage and rapid data acquisition with high spatial resolution in small vessels such as the dorsal pedal artery and, thus, enables therapeutic decision making in regard to the critical peripheral arterial supply. The acquisition of pure arterial phase images improved the evaluation of the small-caliber arteries in the lower calf and foot.
Our data suggest a higher agreement between the three readers in determining outflow segments with MR angiography than with DSA. This is in accordance with the results of studies by Dorweiler et al (22) and Kreitner et al (23), in which the MR angiographic depiction of dorsal pedal artery bypass grafts to blood vessels in the foot that were occult at conventional angiography was evaluated in patients with diabetes mellitus and severe arterial occlusive disease. They found a significant superiority of MR angiography to DSA for the detection of patent pedal vessels. One possible explanation for these findings could be the ability of MR angiography to image blood flow at velocities as slow as 2 cm/sec, whereas in DSA the dilution of contrast media after passage through multiple stenosed segments incidentally leads to inadequate opacification of distal vessels (9). On the other hand, the three readers in this study were less successful in obtaining information about outflow segments with MR angiography than with DSA. Outflow segments were determined by readers 1, 2, and 3 in 17, 16, and 17 limbs, respectively, on DSA images but in only 14, 15, and 17 limbs, respectively, on MR angiograms. Using both DSA images and MR angiograms, however, the most experienced reader (reader 3) failed to determine outflow segments in only two patients (one patient each for DSA and MR angiography), meaning that assessment in 16 patients was successful with both methods. These results may indicate one of the most important disadvantages of MR angiography, that is, the lack of experience with this method. We believe that experience will increase, once MR angiography enters routine practice; our results, however, suggest that extensive training in reading MR angiographic images is necessary before a radiologist should use this method routinely, and that supervision by more experienced radiologists should be available in cases with equivocal results.
Although all patients in our study had multilevel arterial occlusive disease, there was a high level of agreement in target lesion determination between readings of DSA images and readings of hybrid MR angiograms. The hybrid MR angiography protocol enabled detection of the hemodynamically most important lesion with the same level of precision as that achievable with DSA.
Concerning the therapy decisions, there was only moderate agreement between readings of DSA images and those of MR angiograms, with better results for readings of DSA images. In this study, in contrast to studies like that performed by Khilnani et al (17), all possible treatment options for critical limb ischemia were compared, including surgical and interventional approaches, as well as conservative options. Although we recognized that specific bypass or endovascular procedure recommendations are in part based on intuitive decisions and on the radiologist’s experience with the particular treatment option, this type of decision making is crucial in any analysis for planning treatment of critical limb ischemia and represents daily clinical practice. One possible explanation for the higher agreement among readers in regard to decision making based on DSA images is that all of the readers were more familiar with DSA than with MR angiography.
One of the major problems in the examination of arteries in the lower part of the leg is contamination of arterial phase images by venous enhancement. With previous MR angiography protocols, venous overlay often drastically limited the interpretation of arterial status even in healthy volunteers or patients with peripheral arterial occlusive disease of grade I or II (15). Recent studies have shown that above-knee popliteal venous compression increases the arteriovenous transit time in the calf and therefore could eliminate venous contamination of angiograms in the calf. With the inclusion of above-knee popliteal venous compression in their four-station MR angiography protocol, Goldman and Lookstein (24) were able to decrease venous contamination on angiograms obtained in the calf from 45% to 0%.
In patients with critical limb ischemia, however, early arteriovenous shunting often leads to venous enhancement even before filling of the distal arteries. Because of that and the severely restricted arterial blood supply, venous compression in these patients would not likely be helpful. The use of a preceding calf acquisition made it possible for us to minimize venous contamination of 3D MR angiograms. In no case did venous artifacts occur that were severe enough to disable decision making, in contrast to the high amount of venous contamination on arterial phase images obtained with the standard four-station MR angiography protocol.
Other studies in which high accuracy in below-knee popliteal vessel detection was demonstrated were performed with a two-station examination only, which made it difficult to see distal vessels like the dorsal pedal artery (11). Various investigators (25–28) have studied an MR angiographic technique that incorporates fast imaging and rapid table motion for evaluation of the entire lower extremity. Initial reports of high sensitivity and specificity with this technique have been followed by more recent reports that describe limitations at levels below the knee, in which image quality may be compromised by venous overlay in 30% of cases (29). Wang et al (5) reported unpredictable venous enhancement that can obscure arteries and motion artifacts; furthermore, the 1.5–2.0-mm spatial resolution inherent in the imaging matrix that is typically used limits the use of 3D bolus-chase MR angiography as a stand-alone procedure. These limitations are most apparent when below-knee popliteal vessels are imaged. In the 89 three-station bolus-chase MR studies reviewed by Wang et al, diagnostic images were obtained in 100% and 96% of the abdominopelvic and thigh stations, respectively, but in only 43% of the calf stations. The inability to reliably and accurately image the below-knee popliteal vessels with bolus-chase MR angiography limits the usefulness of this method for treatment planning. This was demonstrated in a study in which treatment planning with 3D bolus-chase MR angiography was compared with that with use of DSA (6).
Khilnani et al (17) used a combination of multiphase two-dimensional fast gradient-recalled-echo DSA in the calf with three-station bolus-chase MR angiography for defining the inflow and outflow segments for a possible bypass surgery in patients with claudication or limb ischemia. They found a good correlation among three readers in the selection of inflow and outflow segments at MR angiography and DSA. Motion artifacts at two-dimensional MR angiography of the below-knee popliteal vessels contributed to the few remaining discrepancies between the MR angiographic and DSA interpretations. Hany et al (30) used a different approach, in which the low-frequency central parts of the k-space are more frequently acquired than the high-frequency domains. In combination with temporal interpolation, the investigators in that study used a series of time-resolved 3D MR angiographic image reconstructions. Initial results of a comparison of time-resolved contrast-enhanced MR angiography with a bolus-chase technique indicate that more vessels are visualized with the time-resolved technique, without substantial venous overlay (26). Hany et al (30), however, did not perform an evaluation of the ankle vessels, which is essential in determining the appropriate intervention in critical limb ischemia. Furthermore, spatial resolution is restricted, which makes it difficult to exactly determine the degree of stenosis in distal runoff vessels. Modifications of the time-resolved 3D MR angiographic technique, such as floating table isotropic projection imaging, might increase both the spatial resolution and the anatomic coverage available with this approach in the near future (31).
We also acknowledge several limitations of this study. First, one major limitation relates to the small number of patients. Second, cluster effects from analysis of adjacent segments cannot be excluded. We therefore performed our main analysis at the level of the patient, that is, in the context of clinical decision making, which is not vulnerable to any cluster effects. More importantly, segment determination (eg, distance of in- and outflow segments from the target lesion) depends on the individual clinician. Therefore, we reported reader-specific results. Third, as there was no perfect reference standard, our results have limited value regarding the accuracy of the evaluated methods. Fourth, as we evaluated multiple parameters, single results could be distorted due to chance. Therefore, we emphasize that our results have to be interpreted as a whole. Finally, only a randomized diagnostic trial in which the two methods are compared with regard to future clinical outcomes (eg, complications) can provide a definitive answer with respect to the clinical value of MR angiography. Diagnostic accuracy is only an intermediate parameter in clinical decision making. Further studies are needed to show the effect of improved diagnostic accuracy on therapeutic decisions and on patient outcomes such as long-term morbidity and mortality.
In conclusion, the results of this study demonstrate that the combination of dual-phase contrast-enhanced MR angiography in the below-knee popliteal arteries with 3D bolus-chase MR angiography in the abdomen, pelvis, and lower extremities extends the utility of MR angiography for treatment planning in patients with peripheral arterial disease who are being evaluated for limb salvage.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, O.A.M., J.R., S.O.S., M.F.R.; study concepts, O.A.M., J.R., S.O.S., R.S.; study design, O.A.M., J.R., C.W.; literature research, O.A.M., C.W.; clinical studies, O.A.M., J.R., S.O.S.; data acquisition, O.A.M., C.S.; data analysis/ interpretation, U.S., O.A.M.; statistical analysis, U.S.; manuscript preparation, O.A.M., C.W., J.R., S.O.S.; manuscript definition of intellectual content, S.O.S., J.R., B.S.; manuscript editing, O.A.M., C.W.; manuscript revision/review, M.F.R., S.O.S., J.R., B.S.; manuscript final version approval, M.F.R., S.O.S.
O.A.M. and J.R. contributed equally to this work.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. J. Dinter, K. W. Neff, G. Visciani, R. Lachmann, C. Weiss, S. O. Schoenberg, and H. J. Michaely Peripheral Bolus-Chase MR Angiography: Analysis of Risk Factors for Nondiagnostic Image Quality of the Calf Vessels--A Combined Retrospective and Prospective Study Am. J. Roentgenol., July 1, 2009; 193(1): 234 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Klomp, E. Steyerberg, C. Wittens, H van Urk, J. Habbema, and For the ESES study group A prognostic model for amputation in critical lower limb ischemia Vascular Medicine, May 1, 2009; 14(2): 109 - 115. [Abstract] [PDF]
|
|
|
|
|
|
R. Schernthaner, D. Fleischmann, A. Stadler, M. Schernthaner, J. Lammer, and C. Loewe Value of MDCT Angiography in Developing Treatment Strategies for Critical Limb Ischemia Am. J. Roentgenol., May 1, 2009; 192(5): 1416 - 1424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Kramer, M. J. Budoff, Z. A. Fayad, V. A. Ferrari, C. Goldman, J. R. Lesser, E. T. Martin, S. Rajagopalan, J. P. Reilly, G. P. Rodgers, et al. ACCF/AHA 2007 Clinical Competence Statement on vascular imaging with computed tomography and magnetic resonance Vascular Medicine, November 1, 2007; 12(4): 359 - 378. [PDF]
|
|
|
|
 |
|
 C. M. Kramer, M. J. Budoff, Z. A. Fayad, V. A. Ferrari, C. Goldman, J. R. Lesser, E. T. Martin, S. Rajagopalan, J. P. Reilly, G. P. Rodgers, et al. ACCF/AHA 2007 Clinical Competence Statement on Vascular Imaging With Computed Tomography and Magnetic Resonance: A Report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training Developed in Collaboration With the Society of Atherosclerosis Imaging and Prevention, the Society for Cardiovascular Angiography and Interventions, the Society of Cardiovascular Computed Tomography, the Society for Cardiovascular Magnetic Resonance, and the Society for Vascular Medicine and Biology J. Am. Coll. Cardiol., September 11, 2007; 50(11): 1097 - 1114. [Full Text] [PDF]
|
|
|
|
 |
|
 G. Andreisek, T. Pfammatter, K. Goepfert, D. Nanz, P. Hervo, R. Koppensteiner, and D. Weishaupt Peripheral Arteries in Diabetic Patients: Standard Bolus-Chase and Time-resolved MR Angiography Radiology, December 19, 2006; (2006) 2422051111. [Abstract] [Full Text]
|
|
|
|
|
|
K. Nikolaou, H. Kramer, C. Grosse, D. Clevert, O. Dietrich, M. Hartmann, P. Chamberlin, S. Assmann, M. F. Reiser, and S. O. Schoenberg High-Spatial-Resolution Multistation MR Angiography with Parallel Imaging and Blood Pool Contrast Agent: Initial Experience Radiology, December 1, 2006; 241(3): 861 - 872. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
