| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 Department of Radiology, University Hospital of Basel, Petersgraben 4, 4031
Basel, Switzerland.
2 Department of Angiology, University Hospital of Basel, 4031 Basel,
Switzerland.
3 Medical Physics, Department of Radiology, University Hospital of Basel, 4031
Basel, Switzerland.
Received July 9, 2005;
accepted after revision October 10, 2005.
Supported by the Swiss National Science Foundation under grants PP00B-68783
and 3100A0-100633.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Mastering IA near-real-time CE-MRA, including road map reconstruction, with an SSFP technique in the lower extremity of patients with PAOD is an important building block toward successfully performing endovascular catheter MR-guided interventions.
Keywords: angiography • extremities • fluoroscopy • MR angiography • MRI • MR technique • peripheral vascular disease
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A next milestone in the general concept toward MR-guided endovascular interventions is near-real-time MRA in conjunction with reconstruction of road maps using the same data set. The road map technique is a tracking utility to guide interventional devices such as catheters and guidewires and allows exact positioning of angioplasty balloons and stents. Near-real-time MRA might also offer passive or active tracking of catheters and guidewires in a manner analogous to X-ray fluoroscopy [2].
To our knowledge, this study is the first patient study to show the feasibility of IA near-real-time CE-MRA and the calculation of road maps from the same data set in patients with peripheral arterial occlusive disease (PAOD). A nonbalanced 2D steady-state free precession (SSFP) sequence with a nonselective excitation pulse and a frequency-selective fat-saturation pulse was implemented to achieve good background suppression. To further improve image quality, k-space data were subtracted from contrast-enhanced images using an unenhanced mask image.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients were referred to the interventional unit of the radiology department for DSA-guided percutaneous transluminal angioplasty (PTA). The vascular access was antegrade with puncture of the common femoral artery. A 4-French introducer sheath was inserted with the tip in the proximal superficial femoral artery. IA DSA examinations were performed by experienced interventional radiologists using a standard angiography unit (Multistar, Siemens Medical Solutions).
After the intervention, the patients were transferred on a stretcher with the IA sheath in situ from the DSA to the MR suite. In the course of transit, the sheath was continuously flushed with 5,000 IU liquaemin/L (B. Braun Medical AG) in an isotonic saline solution at a flow rate of 1 mL/min to ensure system patency.
IA Near-Real-Time Contrast-Enhanced MRI
All examinations were performed on a 1.5-T clinical scanner (Magnetom
Sonata, Siemens Medical Solutions) equipped with a high-performance gradient
system operating at a gradient strength of 40 mT/m and a slew rate of 200
T/ms. A phased-array peripheral vascular surface coil was used for signal
reception. The coils covered the entire vascular tree of both legs.
|
|
To provide a fast and rough visualization of the arterial run, axial scout slices were applied using a native 2D fast low-angle shot (FLASH) sequence (TR/TE, 11.0/5.6; flip angle, 35°). Based on the axial images, a 2D thick-slab T1-weighted SSFP sequence with a slab thickness of 40 mm was angulated to cover the entire vascular tree at either the thigh or the calf level.
For each examination, a run of 25 consecutive SSFP measurements was performed. After the fifth measurement, IA injection of contrast medium was administered. The first two measurements of the unenhanced images were discarded, and the following three measurements were averaged and served as the background mask [5, 6] (Fig. 1B). Subsequent measurements were complex subtracted from the mask image to provide additional background signal suppression. The images of the pure arterial filling phase were added up to create the road map image (Figs. 2A, 2B, 2C, and 2D), which included in our study 8-10 frames corresponding to an acquisition time of 10.4-13.0 seconds. The road map reconstruction itself took less than 1 second.
|
|
|
|
The side port of the IA access sheath was extended by a 20-cm plastic tube to enable comfortable gadolinium administration. Gadopentetate dimeglumine was applied via a power injector (MR Spectris Injector, Medrad) at an injection rate of 1 mL/s. The thigh and calf stations were examined separately. For each station, a 20-mL bolus of a 50-mmol/L gadopentetate dimeglumine solution was administered starting with the thigh station. The total dose for each examination was 2 mmol of gadopentetate dimeglumine [1].
After MRI, the patients were transferred back to the interventional radiology unit for sheath removal and received a sterile pressure bandage.
Image Analysis
The signal intensities (SIs) were evaluated by regions of interest (ROIs):
three in the femoropopliteal artery and another three in the infrapopliteal
arteries. The rectangular ROIs covered the contrast-enhanced segment of the
investigated vessel and were in the range of 0.5-1 cm2 for the
femoropopliteal axis and 0.3-0.5 cm2 for the infrapopliteal axis.
In the femoropopliteal station, the ROIs were positioned within the proximal
and distal femoral superficial arteries and the popliteal artery. In the
infrapopliteal station, the ROIs were placed in patent vessels. The
contrast-to-noise ratio (CNR) was calculated as the ratio of the mean SI of
the three ROIs (SIROI) minus the SI of the adjacent tissue
(SIadjacent tissue) divided by the SD of the background noise of
the air:
 |
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Qualitatively, the fast SSFP acquisition technique provides good separation between pure arterial and venous filling. Exemplary IA near-real-time CE MR images of the femoropopliteal and infrapopliteal arteries are shown in Figures 2A and 2B. For overview, only every second near-real-time image is displayed.
Quantitatively, sequential mean CNR (± SD) values at the thigh and calf levels are shown in Figure 1B. In general, a continuous CNR increase during injection is observed in both stations until a steady-state condition is achieved. This condition is reached for the femoropopliteal artery after approximately 4 seconds and for the infrapopliteal arteries after approximately 6 seconds. The CNR slope of the femoropopliteal artery is steeper and achieves earlier and higher mean values than that of the infrapopliteal arteries. The mean CNR values under steady-state conditions range from approximately 20 to 25 for the femoropopliteal artery. The mean CNR values are distinctively lower for the infrapopliteal arteries, ranging from 5 to 10.
An exemplary road map image calculated from the data set of Figure 2B is shown in Figure 2C compared with an IA DSA image (Fig. 2D).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Conventional projection CE-MRA images are based on the repetitive application of a rapid T1-weighted FLASH sequence [5, 6]. A major drawback of this technique is the partial volume effect along the slice direction [4]. The MR vessel signal is partially overlaid and is obscured by background tissue signal, especially from poorly saturated low-flip angle regions close to the slice edges when thick slabs are applied. This problem cannot be overcome by a simple mask subtraction technique. Researchers have suggested that a highly improved saturation of background tissue at the slice boundaries can be achieved with a nonselective excitation-saturation pulse applied before a conventional 2D FLASH acquisition [4]. The T1-weighted contrast of such a magnetization prepared sequence is mainly produced and determined by the flip angle and TR of the nonselective excitation-saturation pulse.
A further improvement of intravascular signal gain was also recently proposed using a balanced SSFP technique [7]. Its value was shown in phantom and animal experiments that revealed improved background suppression and increased image contrast compared with the T1-weighted FLASH sequence technique. However, the balanced SSFP techniques are prone to perturbations, such as flow or motion artifacts, eddy currents, and imperfect shim [8]. The nonbalanced SSFP technique used in our study shows a slightly reduced SI compared with the balanced version; however, it is more robust and less vulnerable to imperfections [8]. A direct comparison of these competing SSFP techniques is still under debate.
Because gadolinium chelate is injected directly into the femoral artery, the total dose of gadolinium can be reduced significantly, thereby allowing multiple injections during a single intervention [2]. We used a low-dose injection protocol with only 1 mmol of gadopentetate dimeglumine per injection [1]. In an average patient, more than 20 IA near-real-time CE MR runs may be performed without exceeding the approved U.S. Food and Drug Administration limitations of gadolinium chelate dose, which is 0.3 mmol/kg/d (6 mL of a 50-mmol/L gadopentetate dimeglumine solution/kg/d). The fast SSFP image acquisition technique provides clear separation of arterial, arteriovenous, and beginning venous filling (Figs. 2A and 2B). This allows an online calculation of pure arterial road maps, but its practical value still has to be shown.
CNR values are distinctively lower in the infrapopliteal arteries, which might be due to a dilution effect and the smaller vessel diameter, a phenomenon also observed for diagnostic 3D CE-MRA [1]. Compared with the CNR values of IV CE-MRA, the CNR values of IA CE-MRA are low. Therefore, the near-real-time MR images might not be sufficient for dedicated diagnostic purposes but seem to provide sufficient information about the arterial flow condition, serving as a tracking technique for near-real-time guidance and reconstruction of road maps.
Similar to X-ray fluoroscopy, IA near-real-time CE-MRA reveals the location of a stenosis but might not be sufficient to enable an observer to grade that stenosis. However, road maps are used to guide catheters and to evaluate the dynamic blood flow situation.
In conclusion, this study is, to our knowledge, the first study showing the technical impact of IA near-real-time CE-MRA, including road map reconstruction, with SSFP technique in the lower extremity of PAOD patients. Mastering this technique is an important building block toward successfully performing endovascular catheter MR-guided interventions.
In the future, even faster sequences will allow faster imaging with better spatial and temporal resolution and lead to MR-guided interventions.
Acknowledgments
We thank Tanja Haas and Philipp Madoerin for their effort and continuous
support in MRI acquisition.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Thurnher, S. Miller, G. Schneider, C. Ballarati, G. Bongartz, C. U. Herborn, S. Schoenberg, M. A. Cova, G. Morana, K. Niazi, et al. Diagnostic Performance of Gadobenate Dimeglumine Enhanced MR Angiography of the Iliofemoral and Calf Arteries: A Large-Scale Multicenter Trial Am. J. Roentgenol., November 1, 2007; 189(5): 1223 - 1237. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) than calf level (•).
Because contrast medium was injected in both cases at proximal femoral level,
CNR curve is observably higher in femoropopliteal axis.
