| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York
Ave., Rm. C278D, New York, NY 10021.
2 Department of Radiology, New York University Medical Center, New York, NY
10016.
3 Department of Surgery, New York University Medical Center, New York, NY
100016.
Received August 4, 2005;
accepted after revision September 4, 2005.
Address correspondence to J. Zhang
(zhangj12{at}mskcc.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Time-resolved contrast-enhanced 3D MR angiography with parallel imaging has the potential to provide a rapid and comprehensive evaluation for the surveillance and diagnosis of hemodialysis access malfunctions. This technique may function as an important complement to conventional digital subtraction angiography and may be able to help guide medical management. The MR angiography protocol we present is a noninvasive, versatile, and time-efficient technique, without the need of direct graft puncture or flow interruption, and can be performed using a single injection of contrast material at a single station.
Keywords: hemodialysis • MR angiography • MR technique • parallel imaging • renal failure • time-resolved angiography
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Some researchers have advocated MRI as an alternative to conventional angiography for the evaluation of hemodialysis access [7-10]. Whereas, to date, most investigators [7-9] use only a small field of view to cover the vascular access only, Han et al. [10] and Froger et al. [11] described an approach to cover the vascular access region and central inflow-outflow tracts by performing two overlapping 3D contrast-enhanced MR angiography acquisitions. This approach can potentially provide a comprehensive evaluation of the central arteries and veins in addition to the vascular access, but two separate injections at two different stations are needed.
We optimized a time-resolved 3D contrast-enhanced MR angiography sequence with integrated parallel acquisition technique (iPAT, Siemens Medical Solutions) that can provide a large field of view with high temporal and spatial resolution, by which the hemodialysis access and the entire course of the inflow and outflow vessels can be imaged at a single anatomic station. We sought to evaluate the feasibility and usefulness of this method in the evaluation of patients referred for possible abnormalities in hemodialysis access.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
An axial HASTE sequence was performed of the chest and arm of interest to
localize the hemodialysis access and the inflow and outflow tracts using the
following parameters: TR/TE, 
/46; refocusing flip angle, 150°;
field of view, 375 mm; matrix, 256 x 192; slice thickness, 8 mm;
interslice gap, 2 mm; number of slices, 30-40; and acquisition time, 30-40
sec. The acquired HASTE data set was then used to prescribe all 3D sequences.
For 3D time-resolved contrast-enhanced MR angiography, an oblique coronal slab
was positioned to cover the arm of interest to include the AVF or AVG and the
contiguous chest. The arterial inflow and venous outflow tracts were also
included. Typical imaging parameters for the 3D fast low-angle shot sequence
were as follows: 3.4/1.3; flip angle, 25°; field of view, 450 mm on the
Symphony or 500 mm on the Avanto; rectangular field of view; matrix, up to 384
x 160 on the Symphony and 448 x 239 on the Avanto; interpolated
slice thickness, 1.25-1.7 mm; slab thickness, 72-122 mm; iPAT factor, 2 in
left-right direction; and acquisition time, 6-9 sec.
In all patients, the 3D gradient-echo images described earlier were acquired before IV administration of contrast material first; subsequently, 10-15 consecutive data sets were initiated simultaneously with the start of an IV injection of 20 mL of gadopentetate dimeglumine (Magnevist, Schering) injected using an MR-compatible power injector (Spectris, Medrad) followed by 20 mL of normal saline solution, both at a rate of 2.0 mL/sec. Injection was performed through a peripheral 22-gauge IV catheter placed in the opposite arm. A test bolus timing examination was not performed. During the acquisition, the patient was asked to hold his or her breath as long as possible and then gently breathe.
|
|
|
After the 3D time-resolved contrast-enhanced MR angiography acquisition, an axial 3D fat-suppressed gradient-echo sequence was performed of the entire chest and arm of interest with a volumetric interpolated breath-hold examination (VIBE) for a general anatomic survey during suspended respiration using the following parameters: 3.5/1.5; flip angle, 12°; field of view, 400 mm; matrix, from 256 x 86 to 320 x 156; interpolated slice thickness, 2-3 mm; slab thickness, 350-480 mm; iPAT factor, 2 in anteroposterior direction; and acquisition time, 17-20 sec.
All MR studies were prospectively interpreted by an experienced MR fellowship-trained radiologist and a board-certified MR fellow in consensus; they were informed of each patient's clinical history, but they were blinded to other correlative imaging findings. In all patients, the aortic arch, feeding artery, AVF or AVG, venous outflow tract, and superior vena cava were assessed using subtraction MR angiograms and source data. All vessels were categorized as patent, stenosed, occluded, or aneurysmal.
Two authors reviewed the patients' medical records together and recorded the conventional angiography or Doppler sonography findings. MRI results were correlated with findings at conventional angiography (n = 6) or Doppler sonography (n = 2). No confirmatory data were available in one patient who died soon after the MR study (patient 2).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the six patients who underwent both MR angiography and conventional angiography, six venous stenoses in three patients, one venous obstruction in one patient, and two venous aneurysms in one patient were detected by both techniques, with intermodality agreement in all six patients.
In the two patients who underwent Doppler sonography, one patient was shown to have a focal venous stenosis 2 cm from the AVF anastomosis on MR angiography (Figs. 1A, 1B, 1C, 1D, and 1E). This finding was confirmed on Doppler sonography. The other patient was referred for abnormal focal physical examination after AVF construction. Gadolinium-enhanced MR angiography showed focal venous stenosis adjacent to the AVF anastomosis and additional three outflow venous stenoses in the mid arm and shoulder (Figs. 3A, 3B, and 3C). Doppler sonography performed 2 weeks before MR angiography detected only a focal venous stenosis adjacent to the AVF anastomosis site.
|
|
|
In three patients, six venous stenoses were located in the venous outflow tract (shoulder and upper arm) and would not have been identified if only the vascular access in the lower arm had been imaged (Figs. 3A, 3B, and 3C).
Overall, MR angiography findings helped determine the clinical diagnosis of hemodialysis access malfunctions and influenced medical care in most of the patients. In one patient (patient 6), a venous stenosis was apparent, but we thought that this finding was probably due to artifact from adjacent surgical clips. However, because the patient presented with a failing AVF clinically, conventional angiography was performed for further evaluation, which did not show any venous stenosis. In no patient was treatment misguided on the basis of MR angiography results.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Conventional angiography has been the technique of choice for anatomic hemodialysis access evaluation. It has the advantage that therapeutic procedures such as angioplasty or thrombectomy can be performed in the same setting. However, conventional angiography has several disadvantages including its invasiveness; exposure of the patient to ionizing radiation; and risks associated with the use of iodinated contrast material, such as allergic reactions. As a diagnostic technique, conventional angiography is a projectional technique. Interpretation of images can therefore be difficult due to overlap of the vessels. In addition, the arterial part of the vascular access is visualized using a proximal cuff to achieve flow interruption and retrograde filling. However, the arterial part of the vascular access is not always clearly depicted because of incomplete retrograde filling of the feeding artery [7, 17].
The commonly used noninvasive method to assess the hemodialysis vascular access is Doppler sonography [18, 19]. Sonographic evaluation can localize and characterize focal vascular access complications, but it is limited in the detection of central vascular abnormalities. For example, in patient 4 of our series, Doppler sonography detected a focal stenosis adjacent to the AVF but missed all three additional outflow stenoses in the mid arm and shoulder. In addition, it cannot provide an angiographic map for percutaneous or surgical angioplasty. Another limitation of sonography is that it is highly operator-dependent.
With technologic advances that allow faster and higher-spatial-resolution imaging, MR angiography has been increasingly performed for the evaluation of vascular anatomy and disease. MR angiography is a noninvasive technique, not accompanied by radiation exposure, and gadolinium contrast agents are rarely associated with side effects. MR angiography also has the advantage over conventional angiography of providing multiplanar and 3D reconstruction capabilities. The use of MR angiography for the detection of hemodialysis access complications has been reported in several studies [7-10]. Although excellent initial results comparable to digital subtraction angiography have been reported, to date the reported techniques have certain limitations, such as a small field of view, long acquisition times, or technical complexity.
Our aim was to improve the contrast-enhanced MR angiography technique for vascular access visualization by using multicoil arrays that enable imaging over a large field of view and by using multichannel MR systems that facilitate parallel imaging methods for faster and higher-spatial-resolution imaging [20, 21]. Using this technique, we improved the application of 3D contrast-enhanced MR angiography for the visualization of vascular access complications in the following ways: First, we were able to achieve a large field of view covering the entire course of the inflow-outflow tracts in addition to the AVF or AVG. Because vascular access complications can occur distant from the access site, it is important to include evaluation of the feeding arteries and venous outflow as an integrated part of vascular access assessment. In our study, six of the 11 venous stenoses found in three of five patients were remote from the access sites, including the cephalic and brachial veins in the shoulder and upper arm, and would not have been identified if only the area around the vascular access was imaged. Among the published contrast-enhanced MR angiography techniques, Han et al. [10] and Froger et al. [11] are the only authors to have achieved coverage of the central vascular structures in addition to the region of vascular access, and they did so only by using two overlapping 3D contrast-enhanced MR angiography volumes with two separate injections at two separate stations. Our technique allows coverage of a large field of view with only one injection of a single dose of gadopentetate dimeglumine at a single station.
Second, despite the large field of view, with a parallel factor of 2, we could achieve a temporal resolution of 6-9 sec for each interpolated 3D MR angiography acquisition while maintaining a high spatial resolution (typical interpolated voxel size, 1.1 x 1.2 x 1.2 mm). In our limited study of nine subjects, eight of whom had digital subtraction angiography or sonography studies for comparison, we found the time-resolved technique to be robust and accurate. The typical acquisition time that has been reported is 32-48 sec for each station [7, 8, 10, 11]. Planken et al. [9] achieved a temporal resolution in a similar range (9.2-11 sec); however, to do so, their fields of view were limited to cover the forearms only (300 x 90 mm). In our experience, the time-resolved technique using a relatively high temporal resolution helps differentiate the inflow and outflow tracts. A 6- to 9-sec temporal resolution has been proven diagnostic in our limited patient population. However, a potential advantage of reaching an even higher temporal resolution is that information about flow direction may be obtained, which has been a drawback of conventional gadolinium-enhanced MR angiography compared with traditional time-of-flight (TOF) techniques. Further improvements in temporal resolution are possible with k-space-sharing techniques such as keyhole imaging [22] or the TRICKS (time-resolved imaging of contrast kinetics) technique [23], which were not available during this study.
Third, we substantially reduced the complexity of the procedure. Instead of using TOF for localization, which can take more than 2 min just to cover the vascular access in the forearm [9], we used HASTE, a single-shot technique that can be performed in less than 1 min for full coverage of the chest and arm. Because we used a single-station time-resolved technique, no test bolus timing examination was necessary. IV access was required for contrast administration, but the vein accessed could be anywhere in the body. Accessing the hemodialysis conduit was not necessary, and neither was flow interruption.
Bos et al. [7] and Smits et al. [8] reported using 2D phase-contrast surveys to monitor filling of the vascular access with contrast material, and a two-stage hand-injection protocol was used to optimize adequate filling of the vascular access both downstream and upstream of the puncture site with a flow-interruption technique. This process necessitates significant physician involvement throughout the examination. Our technique has eliminated the need to monitor filling of the vascular access with contrast material and to place a needle in the hemodialysis access, and the injection was performed automatically with a power injector. Because of the simplicity of the procedure, the whole study can be completed within 10 min of scanning time with little physician time involved.
Last, we substantially reduced the amount of contrast material administered compared with other recirculation techniques. We were able to obtain high-quality diagnostic images using approximately a single dose of gadolinium contrast material (20 mL), as compared with previous studies that used 30-40 mL [9, 10].
Our study contained a relatively small number of patients. We also had conventional angiographic correlation in only six of the nine patients; however, in all six patients, MR angiography findings were confirmed. Nevertheless, we have shown the feasibility of our technique and its capability to provide appropriate guidance for medical management. Further improvement in temporal resolution may be achieved with k-space-sharing techniques that recently became available, which may lead to improved delineation of the effects of stenoses or obstruction on blood flow, such as through delayed patterns of enhancement or reversal of flow direction. In addition, higher spatial and temporal resolutions may be achieved in the future with an increased number of coil elements and parallel factors greater than 2.
One potential diagnostic pitfall in the evaluation of vascular access on MRI is that surgical clips adjacent to the hemodialysis access site may cause susceptibility artifacts, leading to either false-positive results or over-estimation of the degree of vascular stenosis. Reviewers should therefore be aware of this pitfall to avoid unnecessary procedures or misguide a patient's management.
Froger et al. [11] found that MR angiography can accurately depict stenoses in patients with dysfunctional hemodialysis access but has limited clinical value as a result of the current inability to perform MR-guided access interventions and, therefore, should be considered only when digital subtraction angiography is nondiagnostic [11]. However, in our patient population, we found a significant number of abnormalities other than stenosis causing hemodialysis access dysfunction, which often required surgical revision. MR angiography is helpful in this scenario and would become an even more clinically attractive alternative with further advancements in MR-guided endovascular interventions. Froger et al. also agreed that MR angiography, if combined with MR velocity mapping, may be especially useful for follow-up over time because both the anatomy and the function of dialysis fistulas and grafts can be assessed [11, 24, 25]. Therefore, we believe MR angiography has merit in the evaluation of hemodialysis access.
In conclusion, time-resolved contrast-enhanced 3D MR angiography with parallel imaging has the potential to provide a rapid and comprehensive evaluation for the surveillance and diagnosis of hemodialysis access malfunctions. This technique may function as an important complement to conventional digital subtraction angiography and may be able to help guide medical management. The MR angiography protocol we presented is a noninvasive, versatile, and time-efficient technique, without the need of direct graft puncture or flow interruption, and can be performed using a single injection of contrast material at a single station. In our institution, it is well accepted by the vascular surgeons and is well tolerated by the patients. Our preliminary results suggest that time-resolved MR angiography with parallel imaging provides accurate results for assessing hemodialysis access dysfunction, although further studies are necessary.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Stepansky, E. M. Hecht, R. Rivera, L. E. Hirsh, B. Taouli, M. Kaur, and V. S. Lee Dynamic MR Angiography of Upper Extremity Vascular Disease: Pictorial Review RadioGraphics, January 1, 2008; 28(1): e28 - e28. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
