| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 Department of Radiology, Division of Emergency Radiology, University of
Geneva, 24 rue Micheli-du-Crest-14, Geneva 1211, Switzerland.
2 Department of Radiology, University Hospital, Geneva 1211, Switzerland.
3 University Institute for Applied Radiophysics, Lausanne 1007,
Switzerland.
4 Department of Internal Medicine, University Hospital, Geneva 1211,
Switzerland.
5 Clinic of Cardiovascular Surgery, University Hospital, Geneva 1211,
Switzerland.
Received August 4, 2003;
accepted after revision March 16, 2004.
Address correspondence to P.-A. Poletti
(pierre-alexandre.poletti{at}hcuge.ch).
Introduction
 Top Introduction Subjects and MethodsResultsDiscussionReferences |
|---|
|
TopIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Therapeutic management was based on DSA findings only. The protocol of this study was approved by the ethics review board of our hospital, and written informed consent was obtained from each of the patients.
MDCT Angiographic Images
MDCT angiographic images were obtained on a high-speed MDCT unit (MX-8000,
Marconi Medical Systems). Helical CT was initially performed without IV
contrast medium using the following parameters: nominal section thickness of
3.2 mm, pitch of 1.75 mm, gantry rotation period of 0.5 sec, reconstruction
field of view of 40 cm, reconstruction intervals of 1.6 mm, X-ray tube
potential of 120 kV, and current of 240 mA.
A second CT examination was performed after IV injection of a 150-mL bolus of 400 mg I/mL of iomeprol (Iomeron 400, Bracco) at a rate of 3.5 mL/sec.
Contrast-enhanced nonsubtracted MDCT images were processed on a 3D workstation (Vitrea 2 imaging software, Vital Images). Anteroposterior and oblique views of the aorta and iliac vessels were acquired using the workstation preprogrammed 3D volume rendering for angiographic procedures. No other computed proceeding maneuver was performed in order to simplify and standardize the procedure for any patient and for each technique (nonsubtraction MDCT, subtraction MDCT, or DSA) to exclude potential biases. An optimal window level setting was defined on the first nonsubtraction MDCT and subtraction MDCT examinations and was applied to all subsequent MDCT examinations. Bones were removed using the automated workstation's 3D region-growing tool; no manual segmentation was performed to exclude operator-dependent biases.
Subtracted Images
The subtraction MDCT series were obtained by subtracting the
contrast-enhanced MDCT series from the unenhanced MDCT series using basic
software developed to process this task. This software is written in ANSI C
(American National Standards Institute) and subtracts each axial DICOM image
from the other at the same level to produce a new axial DICOM series. A manual
rigid image shift was applied if necessary. This manual shift was possible in
only the x- and y-axes and was applied to the entire series
to reduce operator-dependent biases. The subtracted DICOM series was then
introduced in our PACS and was processed by the same investigator using the
same workstation and the same method as used for nonsubtraction MDCT except
for the bone-removing tool (Figs.
1A,
1B,
1C and
2A,
2B,
2C).
|
|
|
|
|
|
Conventional Angiography
Conventional angiography was performed with a digital subtraction technique
(Integris 3D-RA, Philips Medical Systems) using the standard protocol at our
institution. Images of the lower limbs were obtained sequentially from the
iliac arteries through the feet with five separate contrast medium injections.
Images were obtained at the rate of two images per second of the pelvic and
femoral arteries and one image per second from the popliteal arteries to the
arteries in the feet. A dose of 320 mg I/mL of ioxaglate meglumine (Hexabrix
320, Guerbet) was used.
Acquisition Protocols
Data collection and analysis.—Nonsubtracted MDCT, subtracted
MDCT, and DSA hard-copy films were reviewed by two senior radiologists and one
cardiovascular surgeon blinded to clinical information. The arterial supply
was divided into eight anatomic areas on each side. For each area, the
reviewers graded the severity of the stenosis on a 4-point scale: less than
20% narrowing, 0; 20–49% narrowing, 1; 50–99% narrowing, 2; and
100% narrowing (occlusion), 3.
If more than one stenosis was found in an anatomic area, only the most severe one was the only one considered. If the arterial vessel in an anatomic area was inadequately visualized with one of the techniques (nonsubtraction MDCT, subtraction MDCT, or DSA), findings were reported as indeterminate.
If the difference between the gold standard and an MDCT technique was 1 point or more, the MDCT result was considered a false-negative; if the difference was less than 1 point, the MDCT result was considered a false-positive. The percentages of true-positive and false-negative findings were reported for segments for which comparison was feasible. For this preliminary analysis, we decided to remove indeterminate segments. Therefore, it was not possible to calculate a sensitivity and specificity. These percentages were also calculated for stenoses of 50% or greater, defined as significant.
Dosimetry.—The normalized weighted CT dose index (nCTDIw) of the unit was measured using a standard 32-cm-diameter CTDI test object and a 10-cm-long CT pencil ionization chamber. The effective dose delivered during MDCT was estimated for the pelvic and hip regions. The dose–length product was calculated by multiplying the scanning length by the volume CTDI according to the International Electrotechnical Committee [4]. The dose–length product was then converted into effective dose values by means of a conversion factor of 0.019 mSv / mGyx cm [5].
Estimates of the effective dose of catheter DSA were calculated on the basis of the dose–area product quantity that corresponded to the acquisition protocol used. A conversion factor of 0.15 mSv / Gyx cm2 was used to evaluate the effective dose [6].
|
TopIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Radiologic Findings
Thirty-seven (26%) of the 144 arterial segments were classified as
indeterminate on DSA, 17 (12%) on nonsubtraction MDCT, and 11 (8%) on
subtraction MDCT (p < 0.003, for both nonsubtraction and
subtraction MDCT compared with DSA). Most of the indeterminate results were
for evaluation of the arteries of the feet.
For the segments that could be compared between the three techniques, the percentage of true-positive findings was 82.0% (50/61) and 95.1% (58/61) for nonsubtraction and subtraction MDCT, respectively. These percentages were similar for significant stenosis (81.6% [40/49] and 95.9% [47/49], respectively). Some streak artifacts were linked to the absence of synchronization between the two spirals and to the rigid manual shift between the two series to be subtracted. They resulted in incomplete removal of the bone structures (Fig. 2A, 2B, 2C).
Clinical Considerations
Of the nine patients who underwent both DSA and MDCT, the DSA results alone
indicated that surgery was needed in six patients. After reviewing the DSA and
both nonsubtracted and subtracted MDCT images of these six patients, the
cardiovascular surgeon thought that surgical planning would have been
(wrongly) different in two patients (33%) if nonsubtraction MDCT had been the
only technique available (Figs.
1A,
1B,
1C and
2A,
2B,
2C).
Concordance in the surgical indication between subtraction MDCT and DSA was obtained for all patients.
Dosimetry Results
The average effective dose of MDCT was 6.8 mSv for the two acquisitions.
The whole effective dose of catheter DSA was estimated to be 16.0 mSv.
|
TopIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
However, some limitations should be highlighted: First, subtraction MDCT required a high level of patient collaboration, which may limit its use in critically sick patients. In addition, subtraction MDCT was not feasible in almost 20% of the patients in spite of good collaboration. Finally, this method generates two times more images than nonsubtraction MDCT, which can result in PACS overload.
Future developments to improve subtraction MDCT could include the use of helical synchronization to reduce artifacts during subtraction [7], development of new tools to automatically shift and subtract two 3D series of CT images [8], and precise identification of clinical predictors for selection of patients for subtraction MDCT.
Acknowledgments
We thank Nicolas Murith for his help in the clinical interpretation of the
DSA and MDCT images and Alexandra Platon for her help in preparing the
manuscript.
|
TopIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. H. Heijenbrok-Kal, M. C. J. M. Kock, and M. G. M. Hunink Lower Extremity Arterial Disease: Multidetector CT Angiography Meta-Analysis Radiology, November 1, 2007; 245(2): 433 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. P. Marsman, P.-A. Poletti, and D. Didier Value of IV Contrast Medium in Subtraction CT Angiography Am. J. Roentgenol., April 1, 2005; 184(4): 1360 - 1361. [Full Text] [PDF]
|
|
|
|
 |
|
 G H Roditi and G Harold Magnetic resonance angiography and computed tomography angiography for peripheral arterial disease Imaging, August 1, 2004; 16(3): 205 - 229. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
