| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Ishinomaki Red Cross Hospital, 1-7-10 Yoshino,
Ishinomaki, Miyagi 986-8522, Japan.
2 Department of Diagnostic Radiology, Tohoku University Graduate School of
Medicine, 1-1 Seiryo, Aoba, Sendai, Miyagi 980-8574, Japan.
Received February 10, 2003;
accepted after revision July 23, 2003.
Address correspondence to H. Ota.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twenty-four patients (representing 27 cases) with symptomatic lower extremity peripheral arterial occlusive disease underwent both MDCT angiography and digital subtraction angiography of the aortoiliac and lower extremity arteries. For data analysis, the arterial system was divided into 10 segments. Each segment was classified as normal, mildly stenotic, moderately stenotic, severely stenotic, or occluded. In evaluating MDCT angiographic findings, cross-sectional images were mainly observed by scrolling. The diagnostic accuracy of MDCT angiography was determined, using digital subtraction angiography as the standard reference. The extent of calcification in each segment was also assessed on MDCT angiography and was classified as absent, mildly calcified, or severely calcified.
RESULTS. Of the 480 segments studied, 470 were assessable on both digital subtraction angiography and MDCT angiography. On digital subtraction angiography, 142 stenoocclusive segments (20 mildly stenotic, 14 moderately stenotic, 25 severely stenotic, and 83 occluded) were detected. With regard to the detection of segments that had more than mild stenosis, the sensitivity, specificity, and accuracy of MDCT angiography were 99.2%, 99.1%, and 99.1%, respectively. In the 421 noncalcified and mildly calcified segments, the sensitivity, specificity, and accuracy of MDCT angiography for the detection of more-than-mild stenosis were 100%, 100%, and 100%, respectively.
CONCLUSION. MDCT angiography is a reliable method for evaluating the aortoiliac and lower extremity arteries.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MDCT, a recently developed technique, has made it possible to cover a long segment using thin collimation and a short acquisition time [11, 12]. Rubin et al. [13] used MDCT for the imaging of the entire lower extremity arterial system and found excellent concordance with the findings of digital subtraction angiography. They assessed the arterial system using axial reconstruction, curved planar reconstruction, and maximum-intensity-projection images. The adverse effects of mural calcification on assessment were not discussed. Martin et al. [14] also studied the accuracy of MDCT angiography, reporting that 5-mm collimation revealed arterial atheroocclusive disease with a reliability similar to that of digital subtraction angiography. They evaluated MDCT findings using axial reconstruction, maximum-intensity-projection, and volume-rendered images.
Cross-sectional images obtained using intravascular sonography have been reported to be useful for assessing the diameter of the vascular lumen [15, 16]. We therefore speculated that cross-sectional image analysis of MDCT data would also be effective. Recently, the stand-alone workstation (Zio M900 Quadra, Amin, Tokyo, Japan) that we have used for processing CT images was updated to automatically process multiplanar reconstruction images with a slice thickness of a few millimeters, providing true cross-sectional images of the arteries at any point along the vessel (Cross-Sectional Vascular Analyzer, Amin).
The purpose of this study was to evaluate the efficacy of MDCT angiography with thin collimation compared with digital subtraction angiography in the assessment of lower extremity arterial occlusive disease, mainly by analyzing true cross-sectional images of the vascular system. We also investigated whether the presence of mural calcification might hinder accurate evaluation of arterial stenoocclusive changes on MDCT angiography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subjects
Before undergoing MDCT and digital subtraction angiography, all patients
gave informed consent. Our hospital did not require us to have the approval of
the institutional review board in this study because all MDCT and digital
subtraction angiograms were obtained using our usual protocols. Between
October 1999 and May 2002, MDCT angiography was performed in 120 consecutive
patients with suspected lower extremity arterial occlusive disease. Of these
patients, 24 presented with intermittent claudication or rest pain and had
less than 1.0 on the ankle-brachial blood pressure index, along with MDCT
angiographic findings of stenoocclusive change that could explain the clinical
features. Because these patients were considered to be candidates for invasive
treatment, digital subtraction angiography was undertaken by an angiographer
for the planning of surgical or radiologic intervention. As a result, all 24
consecutive patients (23 males, one female; mean age, 68.9 years; age range,
17–88 years) were included in this study. Three of these patients
underwent both MDCT and digital subtraction angiography twice within the study
period because they again experienced symptoms a few years after the initial
angioplasty. Each of the two sets of examinations in these three patients was
counted as a separate study subject, making a total of 27 cases. One patient
had previously undergone unilateral amputation below the hip joint.
The arterial system was divided into 10 segments in the manner described by Sueyoshi et al. [2]: the common iliac artery, external iliac artery, common femoral artery, superficial femoral artery, popliteal artery, anterior tibial artery, tibioperoneal trunk, posterior tibial artery, peroneal artery (the trifurcation of the vessels was examined above the ankle joint), and bypass graft (Fig. 1). MDCT angiography and digital subtraction angiography were used to examine 479 arterial segments and one bypass graft. Because one patient had undergone amputation of the right lower limb below the hip joint, segments of unilateral lower extremity arteries were excluded from the evaluation. Two segments of common iliac arteries and two segments of external iliac arteries containing implanted metallic stents were included in this series.
|
In addition, the nine arterial segments (the segment containing a bypass graft was excluded) were divided into three groups according to the level of the arterial axis: the iliac group (common iliac and external iliac arteries), femoral group (common femoral and superficial femoral arteries), and calf group (popliteal arteries, anterior tibial artery, tibioperoneal trunk, posterior tibial arteries, and peroneal artery).
MDCT Angiography
The scanner used was a four-channel MDCT system (Aquilion, Toshiba, Tokyo,
Japan). Scanning was performed using the following parameters: 0.5 sec per
table rotation; collimation, 2 mm; table increment, 22 mm/sec; and pitch, 5.5.
X-ray tube voltage was 120 kV, and tube current was 300 mA. Patients were
asked to hold their breath for approximately 15 sec at the start of scanning
and then to breathe shallowly to minimize movement of the lower extremities
during the rest of the scanning procedure. Before scanning was started,
75–100 mL of iopamidol (Iopamiron, Schering, Berlin, Germany;
300–370 mg I/mL) was injected via an antecubital vein at a rate of
1.5–2.0 mL/sec. The scanning delay was set by an automatic triggering
system (Sure Start, Toshiba). Continuous low-dose fluoroscopy (120 kV, 50 mA)
at the level of the lower abdominal aorta was initiated 15 sec after the start
of the contrast material injection.
The MDCT value in a circular region of interest in the abdominal aorta was measured three times per second. When the MDCT value matched or exceeded the threshold value (absolute MDCT value, 85 H) at three consecutive sampling points, scanning automatically was begun from the level of the second lumbar vertebra to the calf. The average duration of scanning was 46 sec, and the total examination time, including the time spent in patient preparation, was less than 15 min. Axial slices were reconstructed with a 2-mm slice thickness at 1-mm intervals. Images were processed using a stand-alone workstation.
Digital Subtraction Angiography
MDCT angiography and digital subtraction angiography were performed within
2 months of each other (range, 1–46 days; mean, 20.4 days). MDCT
angiography was performed first in all cases. None of the patients exhibited
changes in symptoms during the interval between examinations.
Digital subtraction angiography was performed using the Angiostar unit (Siemens Medical Systems, Erlangen, Germany). A 4-French pigtail catheter was positioned above the aortic bifurcation from a trans-femoral (20 cases) or transbrachial (six cases) approach. In the one patient in whom Leriche's syndrome was diagnosed on MDCT angiography (resulting in occlusion from aortic bifurcation to proximal iliac arteries), the catheter was positioned in the ascending aorta, using a transbrachial approach for the evaluation of arteries in the lower extremities, possibly through the collateral vessels from the internal thoracic arteries. In all cases, we obtained the posteroanterior view of the lower abdomen and the entire range of the lower extremities using the stepping-table digital subtraction angiography technique. In 22 patients, bilateral oblique views of the pelvis were also obtained, whereas in the remaining five patients who had only minimal tortuosity of iliac arteries, oblique views were not obtained: posteroanterior views alone were considered to be enough for the evaluation of the iliac arteries. All digital subtraction angiographic findings were evaluated on the films by the angiographer, who was unaware of the results of MDCT angiography.
Image Analysis of MDCT Angiography
Evaluation of all the MDCT scans was performed by two radiologists who had
no knowledge of the findings of digital subtraction angiography. In the event
of disagreements in evaluation, a final opinion was reached through consensus.
In addition to overall assessment of the vascular tree using volume-rendered,
maximum-intensity-projection, and multiplanar reconstruction images, we used
original axial and cross-sectional multiplanar reconstruction images of the
vascular lumen for more detailed analysis.
We used the workstation to process the cross-sectional multiplanar reconstruction images of vascular lumina for the iliac arteries. In this reconstruction, the system initially draws a central vascular line connecting the centers of cross-sectional enhanced areas along the targeted vessel in the volume data. Then, cross-sectional multiplanar reconstruction images that are tangential to the central vascular line are automatically generated (Fig. 2). The time required for this process to cover the total length of the unilateral iliac vessels group, comprising about 200 cross-sectional slices at 1-mm intervals, is a few minutes. Cross-sectional vascular images, as well as standard axial and multiplanar reconstruction images, were evaluated using the paging method (i.e., display by scrolling the images as if turning the pages of a book at the workstation) [17]. Evaluation of the entire iliac and lower extremity arterial occlusive system by the paging method was generally completed within 15 min. Volume-rendered and maximum-intensity-projection images in which the bones were subtracted were routinely generated within 30 min by two radiology technologists.
|
Original Axial Images Versus Cross-Sectional Multiplanar Reconstruction Images
For MDCT angiograms, the results of evaluation of stenoocclusive change
using original axial images were compared with those using cross-sectional
multiplanar reconstruction images. Evaluation of the MDCT data of the iliac
group segments was initially performed, with the two radiologists
independently reviewing original axial images with supplemental use of
maximum-intensity-projection and volume-rendered images. More than 2 months
later, the evaluation was repeated by the same radiologists, using
cross-sectional multiplanar reconstruction images in addition to
maximum-intensity-projection and volume-rendered images. The patients' names
and case record numbers could not be blocked out on the workstation, but the
interpretations of the images were undertaken in random order. For the
segments of femoral and calf vessel groups, cross-sectional multiplanar
reconstruction images were not considered mandatory because these arteries are
not generally tortuous. Evaluation of MDCT data of these groups was performed
using original axial images with maximum-intensity-projection and
volume-rendered images.
Assessment of Stenoocclusive Sites
To match the findings of MDCT and digital subtraction angiographic studies,
we measured the diameter of stenosis in the following manner. Evaluation of
MDCT findings was performed at the workstation. Stenoocclusive sites were
initially identified by visual inspection, and the cross-sectional area
(A) of the stenotic lumen was measured using electronic calipers.
Then, the diameter of the lumen (D) was calculated using the formula,
A = 
(D / 2)2. The calculated diameter at the
point of most severe change in any segment was compared with that of the most
normal-looking region (D') immediately proximal or distal to
the point of stenosis in the same segment. When more than a single stenotic
part was in a segment, assessment of the stenoocclusive process was based on
the segmental part showing the most severe change. The degree of stenosis was
calculated as (1 – D) / D' x 100%. On the
basis of the findings of MDCT angiography, the degree of stenosis in each
segment was classified as absent, mildly stenotic (< 50% luminal
narrowing), moderately stenotic (50–74% luminal narrowing), severely
stenotic (75–99% luminal narrowing), or occluded. In assessing the axial
images of the iliac arteries, we allowed some degree of subjectivity because
measurement of cross-sectional areas was often difficult.
With regard to digital subtraction angiographic findings, we measured the diameter at the most stenotic portion in the individual segment using mechanical calipers. The degree of stenosis was calculated in a way similar to that described for analysis of MDCT findings. For the assessment of the segments of the calf group, we allowed some degree of subjectivity for both modalities because caliper measurement of narrow arteries was often difficult [14].
Assessment of the Extent of Mural Calcification
We knew of no previously described method of evaluating the extent of mural
calcification of lower extremity arteries; therefore, we developed a
classification scheme for use in this study. In our classification system, the
extent of calcification was assessed at the most stenotic point within each
segment because information concerning calcification at that point is
generally of the greatest importance in the selection of the therapeutic
approach. If a segment did not contain a stenoocclusive lesion, the most
severely calcified site was assessed. Irrespective of the cross-sectional
thickness of calcified plaque revealed on MDCT angiography, we used the extent
of calcification along the circumference of the vessel in each segment to
grade mural calcification as absent, mildly calcified (
50% mural
calcification), or severely calcified (>50% mural calcification) (Fig.
3A,
3B,
3C).
|
|
|
Statistical Analysis
Statistical analysis was performed using statistical software (Statistical
Program for the Social Sciences 11.0, SPSS, Chicago, IL). The sensitivity,
specificity, and accuracy, with 95% confidence intervals (CIs), in the
identification of stenosis or occlusion on cross-sectional MDCT angiograms
were calculated for each segment and compared against the results obtained
with digital subtraction angiograms. For the segments of the iliac group, we
calculated the sensitivity, specificity, and accuracy in the detection of
stenoocclusive change using original axial images and then using
cross-sectional multiplanar reconstruction images. The statistical differences
in the diagnostic accuracy of MDCT angiography using both types of images were
calculated using the binominal test for paired data. Interobserver agreement
in the grading of arterial lesions in the iliac group on original axial or
cross-sectional multiplanar reconstruction images was also determined by
calculating kappa values (poor agreement, 
= 0; slight agreement,
= 0.01–0.20; fair agreement, = 0.21–0.40; moderate
agreement, = 0.41–0.60; good agreement, =
0.61–0.80; and excellent agreement, = 0.81–1.00).
For segments with at least moderate stenosis, we calculated the diagnostic
performance of MDCT angiography for the subgroups with severe calcification
(> 50% circumferential calcification) or without severe calcification (
50% circumferential calcification). Statistical differences in the diagnostic
performance of MDCT angiography among the subgroups were calculated with
Fisher's exact test. Kappa values were also calculated for interobserver
agreement in the grading of arterial lesions in the subgroups.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 1 shows the classification of arterial lesions based on MDCT angiographic cross-sectional images with maximum-intensity-projection and volume-rendered images and on digital subtraction angiograms for the segments studied. On digital subtraction angiography, 142 diseased segments (30%) were identified: 20 mildly stenotic, 14 moderately stenotic, 25 severely stenotic, and 83 occluded segments. In two patients, the five segments in the unilateral arteries below the level of the knee were not well opacified on digital subtraction angiography because of the occlusion of the superficial femoral artery, although the segments were all visualized on MDCT angiography, probably as a result of opacification via the collateral vessels. Thus, 10 segments were excluded from the comparison of digital subtraction angiography and MDCT angiography because digital subtraction angiography was selected as the standard reference. Four segments containing stents are included.
|
Of the 470 segments that were assessable on both modalities, 448 (95.3%) gave consistent results in both examinations (Figs. 4A, 4B, 4C, 4D, 4E, 4F and 5A, 5B). On MDCT angiography, nine segments (1.9%) were overestimated by one grade, eight (1.7%) were underestimated by one grade, and five (1.1%) were overestimated by two grades compared with the grades determined using digital subtraction angiograms. Of the five segments overestimated by two grades on MDCT angiography, four were segments of the superficial femoral artery with severe and dense calcification, and the other was a segment of the peroneal artery in the patient with popliteal artery entrapment syndrome. Occlusion by thrombus was observed on MDCT angiograms, but digital subtraction angiograms obtained 20 days after the MDCT examinations showed moderate stenosis in the segment. Of the 17 segments that had one-grade discrepancies between the findings of MDCT angiography and digital subtraction angiography, 13 had severe calcification, and the remaining four were arteries in the calf group without calcification.
|
|
|
|
|
|
|
|
With regard to occlusion, the sensitivity, specificity, and accuracy of MDCT angiography were 96.4% (95% CI, 92.4–100.4%), 98.4% (95% CI, 97.2–99.7%), and 98.1% (95% CI, 96.8–99.3%), respectively. If only those segments with stenosis of 50% or more (moderate stenosis or more severe) that were thought to require therapeutic intervention were considered, the sensitivity, specificity, and accuracy of MDCT angiography were 99.2% (95% CI, 97.6–100.8%), 99.1% (95% CI, 98.2–100.1%), and 99.1% (95% CI, 98.3–100.0%), respectively. Of four stent-implanted segments, two were classified as normal, one as mildly stenotic, and the other as moderately stenotic on MDCT angiography, showing 100% concordance with the results of digital subtraction angiography (Fig. 6A, 6B, 6C, 6D). One bypass graft studied was judged to be patent on MDCT angiography. This finding agreed with that on digital subtraction angiography.
|
|
|
|
Segments of the iliac and lower extremity arteries were sorted into three groups, and for each group, we calculated the sensitivity, specificity, and accuracy of diagnosis based on cross-sectional MDCT angiograms for the detection of stenosis of 50% or more (Table 2). Regardless of the level of the arterial axis, the results of MDCT angiography showed excellent agreement with those of digital subtraction angiography.
|
We also calculated the sensitivity, specificity, and accuracy of MDCT angiography for the detection of stenosis of 50% or more when all 108 segments in the iliac group were analyzed using the original axial images (Table 2). The binominal test showed a statistically significant difference between the rate of correct interpretation using cross-sectional multiplanar reconstruction images and that using axial images (p = 0.0313). The kappa values for interobserver agreement in the grading of arterial lesions in the iliac group with the observation of cross-sectional multiplanar reconstruction or axial images were 0.888 (95% CI, 0.808–0.968) and 0.701 (95% CI, 0.579–0.823), respectively.
As determined on MDCT angiography, mural calcification was judged to be absent in 340 (71%) of the total 480 segments, mildly calcified in 91 (19%), and severely calcified in 49 (10%). Excluding 10 segments that were not visualized on digital subtraction angiography but were visualized on MDCT angiography, the 470 segments were sorted into two subgroups with or without severe calcification. For 49 segments with severe calcification and the remaining 431 segments without such calcification, the sensitivity, specificity, and accuracy of MDCT angiography for the detection of stenosis of 50% or more were calculated and are presented in Table 3. Using Fisher's exact test, we found the diagnostic accuracy to be significantly different between the subgroups in terms of specificity (p = 0.0005) and accuracy (p = 0.0001) but not in terms of sensitivity (p = 0.164). The kappa values for interobserver agreement in the grading of arterial lesions in the subgroups with and without severe calcification for MDCT angiography were 0.760 (95% CI, 0.614–0.906) and 0.957 (95% CI, 0.928–0.986), respectively.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, gadolinium-enhanced MR angiography has gained some acceptance as a useful noninvasive vascular imaging technique, providing high-quality images of the vascular system with high sensitivity and specificity [2–7]. However, the acquisition protocol in MR angiography involves some intricate and time-consuming processes, such as careful setting of the scanning position, multiple acquisitions and test contrast injection. With regard to image quality, MR angiography, like digital subtraction angiography, cannot precisely depict the vascular anatomy at completely occluded sites or the condition of the vascular wall, such as mural calcification or wall thickness. In addition, the presence of indwelling stents can cause severe artifacts that may render the results of MR angiography inaccurate or nondiagnostic [3–5].
Compared with MR angiography, our MDCT angiography protocol was simple—a single-acquisition protocol with an automatic triggering system that took an average of only 46 sec for acquisition and approximately 15 min for the total examination. These are the significant advantages of MDCT angiography.
Because it does not have the limitations of single-detector CT, such as suboptimal z-axis resolution and a limited range of coverage [11, 12], MDCT makes it possible to evaluate short segmental stenosis accurately (Fig. 5A, 5B). MDCT angiography permits the precise evaluation of the vascular wall itself. It can clearly depict the courses of vessels not only in patent but also in completely occluded segments; such depictions provide useful information for planning interventional radiology revascularization procedures (Fig. 6A, 6B, 6C, 6D). If a surgical procedure is required, the preoperative information concerning the vascular wall, such as the degree of mural calcification and plaques, obtained with MDCT angiography is important in determining the anastomotic sites for bypass grafting. In addition, as previously reported [18, 19], extravascular causes of occlusion may be detected on CT angiography, as was seen in one patient with popliteal artery entrapment syndrome in our study.
For the evaluation of arteries with indwelling stents, MDCT angiography may be superior to MR angiography. MR angiography cannot depict in-stent patency [3–5]. In MDCT angiography, beam-hardening artifacts caused by the presence of metallic materials, as well as severe calcification, interfere with proper evaluation. However, in vessels with large diameters such as iliac arteries, the effects of such artifacts are relatively minor, so they do not interfere with the evaluation of in-stent stenosis. Indeed, the vascular findings in all four stent-implanted segments in the iliac group in our study showed excellent agreement between MDCT angiography and digital subtraction angiography, although the number of segments was limited.
Rubin et al. [13] described the usefulness of MDCT angiography of the lower extremity arteries, using a 2.5-mm nominal section thickness. Exploiting the advantages of this method, they reported 100% concordance with digital subtraction angiography for stenoocclusive segments of the lower extremity arteries. However, their study did not provide a true test of the diagnostic accuracy of MDCT angiography because the findings of MDCT angiography and digital subtraction angiography were reviewed and directly compared by the same radiologist rather than evaluated in a blinded manner. They stated that independent assessment by multiple interpreters who were unaware of the results of the other examinations remained to be performed.
A prospective study by Martin et al. [14] revealed reliable accuracy of MDCT angiography of the aortoiliac system and lower extremities compared with digital subtraction angiography. Their results showed the sensitivity and specificity of MDCT angiography for depicting occlusions and stenoses of at least 75% were 88.6% and 97.7%, and 92.2% and 96.8%, respectively. However, their MDCT study was performed with a 5-mm collimation, which decreased the z-axis resolution. Their results were based on combinations of axial and reformatted 3D images such as volume-rendered, maximum-intensity-projection, and curved planar reconstruction images. However, evaluation of true cross-sectional images was not performed.
Using reconstructed MDCT acquisition data with a 2-mm collimation at 1-mm intervals, we compared the results of evaluating the iliac arteries using the original axial images with the results of evaluating the arteries using the cross-sectional multiplanar reconstruction images. In the assessment of iliac arteries, our statistical results showed that sufficient diagnostic accuracy was not achieved using the axial images compared with the accuracy achieved using cross-sectional multiplanar reconstruction images and that the axial images required more interobserver discussion to reach a final consensus. Reports on the efficacy of intravascular sonography have mentioned that cross-sectional images are useful for the assessment of arteries, especially in cases of eccentric stenosis [15, 16]. Our study, which we believe is the first comparative evaluation of original axial images versus cross-sectional images in MDCT angiography, disclosed the superior efficacy of cross-sectional images.
MDCT angiography has two advantages over digital subtraction angiography. First, eccentric stenosis can be evaluated accurately with the use of cross-sectional MDCT angiograms, whereas additional views, including oblique and lateral views, are required in digital subtraction angiography. Second, MDCT angiography makes it possible to show segments immediately distal to the point of occlusion, which are not opacified on digital subtraction angiography. Indeed, in our study, MDCT angiography depicted 10 additional segments that could not be visualized at all on digital subtraction angiography. This finding may be due to higher contrast resolution of MDCT, which shows only faintly opacified lumen distal to the complete occlusion, probably via collateral vessels [9, 13, 14, 20].
In five segments, stenotic lesions were overestimated by two grades on MDCT angiography. Four of them were superficial femoral artery segments with severe and dense calcifications, resulting in strong beam-hardening artifacts, which might have hindered precise evaluation. The remaining one was a peroneal artery segment in a patient with popliteal artery entrapment syndrome. Complete occlusion by thrombus was visualized on MDCT, but digital subtraction angiography performed 20 days later showed only moderate stenosis. This discrepancy might have been due to spontaneous recanalization during the interval between the MDCT and digital subtraction angiography examinations: subsequent MDCT angiographic examination (excluded from this study because it was performed after decompression surgery of the popliteal artery) performed 1 month after digital subtraction angiography showed complete patency of the artery and its branches.
In 17 segments, there were one-grade discrepancies between MDCT and digital subtraction angiographic findings in the grading of stenoocclusive change. Of these, 13 segments were severely calcified, which we believed caused the grading discrepancies. The remaining four segments were narrow calf arteries. Two of the four segments were diagnosed as occluded on MDCT angiography but as severely stenotic on digital subtraction angiography, and the other two were diagnosed as severely stenotic on MDCT angiography but as occluded on digital subtraction angiography. The length of the stenoocclusive lesions in all four segments was quite short, measuring slightly more than a few millimeters. Although calcification was absent in all of them, the reason for the discrepancies may have been related to the limited spatial resolution of MDCT.
The limitations of MDCT angiography were related to difficulties in the evaluation of severely calcified lesions due to beam-hardening artifacts. In contrast to the 100% consistency with the results of digital subtraction angiography in the detection of stenosis of 50% or more in arteries without severe calcification, the specificity and accuracy of MDCT angiography in severely calcified segments were significantly different. Even when the grading of stenoocclusive changes was consistent between MDCT angiography and digital subtraction angiography, interobserver discussion was often required to reach consensus in evaluating the results of MDCT angiography in cases with more severe calcification, which was reflected in the difference of kappa values between the subgroups with and without severe calcification. Such cases could not be managed without digital subtraction angiography for accurate evaluation. Compared with the complete concordance described in the report of Rubin et al. [13], the diagnostic accuracy of MDCT angiography for the detection of stenoocclusive lesions was a little worse in our study. However, the extent and significance of mural calcification were not described in their study. Our study population might have included more cases with more severe calcification.
Our study had several limitations. First, digital subtraction angiography was performed only in patients with intermittent claudication or rest pain, a low ankle-brachial blood pressure index, and stenoocclusive findings on MDCT angiography considered to be treatable by an invasive procedure. However, we compared digital subtraction angiograms with the MDCT angiograms that had at least partially determined whether digital subtraction angiography should be performed. The second problem might have been different indications for digital subtraction angiography. In the early period of this study, digital subtraction angiography was always performed immediately before treatment in patients who required surgery or interventional radiology. In the later stage of the study, however, because the correlation between MDCT angiography and digital subtraction angiography was found to be excellent, surgical intervention was undertaken without obtaining digital subtraction angiograms in patients with excellent-quality MDCT angiograms. Therefore, these cases were excluded from this series in the latter period, and digital subtraction angiography was performed in the limited number of cases in which evaluation using MDCT angiography alone was difficult. This factor might have negatively biased the results for the diagnostic capabilities of MDCT angiography in our study.
In conclusion, MDCT angiography performed using our new software was found to be a reasonable alternative to other diagnostic imaging techniques for the evaluation of patients with lower extremity arterial occlusive disease. We emphasize the importance of careful observation of true cross-sectional images using the paging method.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. H. Lin, C. Bechara, P. Kougias, T. T. Huynh, S. A. LeMaire, and J. S. Coselli Assessment of Aortic Pathology and Peripheral Arterial Disease Using Multidetector Computed Tomographic Angiography Vascular and Endovascular Surgery, December 1, 2009; 42(6): 583 - 598. [Abstract] [PDF]
|
|
|
|
 |
|
 S. C. Josephs, H. A. Rowley, G. D. Rubin, and for Writing Group 3 Atherosclerotic Peripheral Vascular Disease Symposium II: Vascular Magnetic Resonance and Computed Tomographic Imaging Circulation, December 16, 2008; 118(25): 2837 - 2844. [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Anderson, B. R. Foster, and J. A. Soto Upper Extremity CT Angiography in Penetrating Trauma: Use of 64-Section Multidetector CT Radiology, December 1, 2008; 249(3): 1064 - 1073. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Sugita, T. Yamazaki, N. Fujita, T. Naitoh, M. Kobari, and S. Takahashi Cystic Artery and Cystic Duct Assessment with 64-Detector Row CT before Laparoscopic Cholecystectomy Radiology, July 1, 2008; 248(1): 124 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. K. Fishman, K. M. Horton, and P. T. Johnson Multidetector CT and Three-dimensional CT Angiography for Suspected Vascular Trauma of the Extremities RadioGraphics, May 1, 2008; 28(3): 653 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Schernthaner, D. Fleischmann, F. Lomoschitz, A. Stadler, J. Lammer, and C. Loewe Effect of MDCT Angiographic Findings on the Management of Intermittent Claudication Am. J. Roentgenol., November 1, 2007; 189(5): 1215 - 1222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
T. Albrecht, E. Foert, R. Holtkamp, M |
