| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
1 From the Institute of Diagnostic Radiology, University Hospital Zurich, Rämistrasse 100, 8091 Zurich, Switzerland (J.K.W., B.B., T.S., S.W., T.P., B.M., T.B.); University Institute of Applied Radiophysics, Lausanne, Switzerland (F.R.V.); and Department of Biostatistics, University of Zurich, Switzerland (B.S.). From the 2004 RSNA Annual Meeting. Received May 18, 2004; revision requested July 29; revision received August 23; accepted October 1. Supported in part by the National Center of Competence in Research, Computer Aided and Image Guided Medical Interventions (CO-ME Project 12) of the Swiss National Science Foundation. Address correspondence to J.K.W. (e-mail: juergen.willmann{at}usz.ch).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: This study was approved by the institutional review board, and informed consent was obtained. A total of 39 consecutive patients (27 men [mean age, 66 years] and 12 women [mean age, 64 years]) with peripheral arterial disease underwent both conventional DSA and 16–detector row CT angiography. For data analysis, the arterial vascular system was divided into 35 segments. A total of 1365 arterial segments were analyzed for arterial stenosis by two independent blinded readers using a four-point grading system (grade 1, <10% luminal narrowing; grade 2, 10%–49% luminal narrowing; grade 3, 50%–99% luminal narrowing; grade 4, occlusion). Interobserver agreements were calculated by using 
statistics. A third independent blinded reader assessed possible reasons for disagreements between 16–detector row CT angiographic findings and conventional DSA findings. Effective radiation dose was calculated for both imaging modalities.
RESULTS: Sixteen–detector row CT angiographic and conventional DSA findings were diagnostic in all vascular segments. Compared with conventional DSA, the sensitivity and specificity of 16–detector row CT angiography with regard to detection of hemodynamically significant stenosis in all 35 arterial segments were 96% and 97%, respectively, for both readers. Readers 1 and 2 overestimated arterial stenosis in 42 (3%) and 34 (2%) arterial segments, respectively, and underestimated arterial stenosis in 13 (1%) and 10 (1%) arterial segments, respectively. Interobserver agreement was excellent ( = 0.84–1.00). Presence of anteroposteriorly located luminal narrowing and extensive vascular wall calcification were considered main reasons for disagreements between imaging modalities. Effective radiation dose was lower for 16–detector row CT angiography (1.6–3.9 mSv) than for conventional DSA (6.4–16.0 mSv).
CONCLUSION: Sixteen–detector row CT angiography is an accurate and reliable noninvasive alternative to conventional DSA in the assessment of aortoiliac and lower extremity arteries in patients with peripheral arterial disease.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Computed tomographic (CT) angiography is increasingly used for noninvasive imaging of various vascular territories. The introduction of multi–detector row CT scanners has substantially improved CT angiography by offering increased volume coverage, decreased dose of contrast medium, decreased acquisition time, and improved spatial resolution for assessment of smaller arterial branches, including the aortoiliac and lower extremity arteries (4–8). However, because of the limited spatial resolution along the z-axis, four–detector row CT angiography is challenged in the assessment of small vessels, including the internal iliac arteries (6) and the peripheral arteries of the calves (5,9). The rate of diagnostic agreement between four–detector row CT angiography and conventional DSA has been shown to be significantly lower in the assessment of arterial stenosis of the peripheral arteries of the calves when compared with the assessment of the larger proximal arteries of the thigh (9). Sensitivity of four–detector row CT angiography was compared with that of conventional DSA and shown to be lower in the assessment of arterial stenosis of small internal iliac arteries when compared with larger common and external iliac arteries in another study (6). In a study performed by Ofer et al (5), 64% of clinically important mismatches between conventional DSA and four–detector row CT angiography occurred in the small renal arteries and peripheral arteries of the calves.
By offering up to threefold improved z-axis resolution when compared with that of four–detector row CT scanners, last-generation 16–detector row CT scanners may overcome this limitation of CT angiography and further improve diagnostic accuracy of CT angiography when compared with conventional DSA in patients with peripheral arterial disease.
The purpose of this study, therefore, was to prospectively compare the accuracy of 16–detector row CT angiography with that of conventional DSA in the assessment of aortoiliac and lower extremity arteries in patients with peripheral arterial disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Exclusion criteria were a history of renal insufficiency or adverse reactions to iodinated contrast agents. None of the 39 consecutive patients were excluded from the study. The study was approved by the institutional review board at the University of Zurich, Switzerland, and informed consent was obtained from all patients.
All patients underwent conventional DSA and 16–detector row CT angiography within 6 days. Sixteen–detector row CT angiography was performed after conventional DSA in all 39 patients. The mean delay between conventional DSA and 16–detector row CT angiography was 4 days (range, 1–6 days).
Conventional DSA
In all 39 patients, intraarterial conventional DSA was performed transfemorally with a 4-F pigtail catheter (AngiOptic; Angiodynamics, Queensbury, NY) by one of two vascular radiologists (either T.P. or a second vascular radiologist, with 12 and 3 years of experience, respectively) who used one of two units (Integris V3000 or Integris V5000; Philips Medical Systems, Best, the Netherlands). For evaluation of the abdominal aorta, the catheter tip was positioned between the 12th thoracic and first lumbar vertebral body, and 30 mL of the nonionic iodinated contrast material iopromidum (Ultravist 300; Schering, Berlin, Germany; 300 mg iodine per milliliter) was injected. Subsequently, the catheter tip was positioned above the aortic bifurcation for conventional DSA of the pelvic and lower extremity arteries, and 20 mL of contrast material was administered. On average, a total of 140 mL of contrast material was injected during conventional DSA. At the level of the iliac arteries, 30° left and 30° right anterior oblique projections were obtained in all patients. Lower extremity arteries were assessed by using the stepping-table DSA technique in a posteroanterior projection. Oblique or lateral projections of the femoral and popliteocrural region were obtained only if they were deemed necessary by one of the vascular radiologists performing the examination.
Since the angiography suite is not connected to a picture archiving and communication system, images obtained with conventional DSA were printed on film with customized window width and level settings to allow clear delineation of the enhanced lumen.
Sixteen–Detector Row CT Angiography
All 39 patients were examined with a 16–detector row CT scanner (Sensation 16; Siemens, Forchheim, Germany). All patients were placed in the supine position, with their feet entering the gantry first. Each patient's extremities were positioned with the knee and ankle joints in the neutral position. After an initial scout image (tube voltage, 120 kV; tube current, 50 mAs) was obtained, the scanning range was planned for each individual to encompass the aortoiliac and lower extremity arteries. The mean scanning coverage was 1140 mm (range, 1032–1245 mm).
The delay time between the start of contrast material administration and the start of scanning was obtained for each patient individually by using a bolus-tracking technique (CARE-Bolus software; Siemens) for optimal intraluminal contrast enhancement. First, a single unenhanced low-dose (20 mAs) image was obtained at the level of the distal infrarenal abdominal aorta. On the basis of this transverse image, a technologist set a region of interest with an area of 10–15 mm2 in the lumen of the distal infrarenal abdominal aorta. This region of interest served as a reference for the following dynamic measurements of contrast enhancement. Subsequently, a nonionic iodinated contrast medium (Visipaque; Amersham Health, Buckinghamshire, England; 320 mg iodine per milliliter) was administered via a 20–22-gauge needle, which had been placed in a superficial vein located in the antecubital fossa. The volume of contrast medium (mean, 100 mL; range, 85–110 mL) was adjusted for the scanning length of each patient to establish a bolus duration that was equivalent to the scanning duration (10). The contrast medium was administered with an automated injector (Ulrich Medical, Ulm-Jungingen, Germany) at a flow rate of 4 mL/sec and followed by a 30-mL flush of saline administered at the same flow rate.
Repetitive low-dose monitoring examinations (120 kV, 10 mAs, 0.5-second scanning time, 1-second interscan delay) were performed 10 seconds after contrast medium injection began. After reaching the preset contrast enhancement level of 100 HU (mean number of repetitive scans, eight), 16–detector row CT scanning was initiated automatically 2 seconds later.
Data acquisition was performed in a craniocaudal direction with a nominal section thickness of 0.75 mm, a table feed of 18 mm per rotation, and a 0.5-second gantry rotation time (pitch, 1.5). The x-ray tube voltage setting was 120 kV, and mean tube current was 210 mA (range, 170–260 mA). An on-line modulation of the tube current was used for all 16–detector row CT scans to reduce the radiation dose. In short, tube current was modulated according to rotation angle–dependent x-ray attenuation without a concomitant loss in image quality. This means that the tube current was reduced in projections with low attenuation and increased in projections with high attenuation (11,12).
Transverse sections were reconstructed for each extremity separately on a workstation, with a section thickness of 0.75 mm at an interval of 0.4 mm, which resulted in a mean of 5700 transverse images (range, 5160–6225 transverse images). The reconstruction field of view for each reconstruction was 25 cm. The field of view, matrix size of 512 x 512, and section thickness of 0.75 mm resulted in a voxel size of 0.18 mm3. The mean total room time, which was defined as the time from patient entry into the CT suite until scanning was finished, was 12 minutes. Sixteen–detector row CT angiography was performed in all 39 patients without any complications, and none of the examinations needed to be repeated because of technical problems.
All 16–detector row CT angiographic data were transferred to a dedicated workstation (Advantage Windows 4.2; GE Medical Systems, Milwaukee, Wis) with commercially available software that allows generation of different three-dimensional reconstructions, including maximum intensity projections and volume renderings. Three-dimensional reconstructions were generated by an independent technician who was not involved in the study. The technician was blinded to clinical history and results of conventional DSA for all patients. For better localization of vascular abnormalities on the three-dimensional reconstructions during image analysis, segmentation of bone structures was not performed in this study. For each patient, 36 volume-rendered images and 36 maximum intensity projection images were created perpendicular to the superoinferior axis covering 360° of rotation in 10° increments. Three-dimensional reconstructions were stored on the hard-disk memory of the workstation for subsequent image analysis.
Image Analysis
Conventional DSA findings were interpreted on hard-copy images by the two vascular radiologists who performed conventional DSA. Interpretation disagreements were resolved by means of consensus review. Both radiologists were blinded to CT and clinical data, and they were not involved in the further course of the study. Conventional DSA served as reference standard.
Analysis of 16–detector row CT angiograms was performed separately by two independent blinded radiologists (T.B. and J.K.W., both with 4 years of experience in reading multi-detector row CT angiograms). Images were analyzed on the basis of the transverse CT source data and the maximum intensity projections and volume renderings available for both readers on the workstation. Interactive reformatting was also available on the workstation. Interactive reformatting included interactive viewing of the transverse CT source images and interactive generation of reconstructions in other planes (including coronal, sagittal, and curved reconstructions) by the readers themselves. Both readers were blinded to patient data, including clinical history and findings at conventional DSA. The readers analyzed the multi–detector row CT angiograms of all 39 patients in random order. Both readers were allowed to individually adjust window center and level settings of the 16–detector row CT angiograms for image analysis.
For analysis of both conventional DSA images and 16–detector row CT angiograms, the arterial vascular system being considered was divided into the following 35 arterial segments: (a) the infrarenal aorta; (b) the common iliac arteries; (c) the external iliac arteries, which were divided into a proximal and a distal segment; (d) the internal iliac arteries; (e) the common femoral arteries; (f) the deep femoral arteries; (g) the superficial femoral arteries, which were divided into a proximal and a distal segment; (h) the popliteal arteries, which were divided into a proximal and a distal segment; (i) the tibiofibular trunks; (j) the anterior tibial arteries, which were divided into a proximal and a distal segment; (k) the peroneal arteries, which were divided into a proximal and a distal segment; and (l) the posterior tibial arteries, which were divided into a proximal and a distal segment. This resulted in a total of 1365 evaluated arterial segments. Each arterial segment was analyzed with regard to image quality (ie, nondiagnostic vs diagnostic) and for the presence of arterial stenosis and aneurysmal changes. Image quality of an arterial segment was considered nondiagnostic if diagnostic information could not be derived because of inadequate vessel enhancement or blurring of the arterial segment. Image quality was considered diagnostic if all clinically relevant diagnostic information could be obtained with good differentiation of arterial vasculature from background tissue.
Stenosis of the arterial segment was graded by using a four-point Likert scale. Grade 1 indicated a normal vessel or mild vessel irregularities (<10% luminal narrowing). Grade 2 indicated moderate arterial stenosis (10%–49% luminal narrowing). Grade 3 indicated severe arterial stenosis (50%–99% luminal narrowing). Grade 4 indicated occlusion. Grading of the arterial stenosis was performed with an electronic caliper. Arterial stenosis with a grade of 1 or 2 (<50% luminal narrowing) was considered to be hemodynamically insignificant, whereas arterial stenosis with a grade of 3 or 4 (50%–100% luminal narrowing) was considered hemodynamically significant. In case of concurrent arterial stenosis in a single arterial segment, only the stenosis with a higher grade was evaluated.
Evidence of the presence and location of aneurysmal changes was noted separately. An aneurysmal change was diagnosed in the presence of a focal increase in arterial diameter that exceeded the diameter of the adjacent arterial segment by more than 50%.
Four weeks after analysis of 16–detector row CT angiograms by readers 1 and 2, a third independent blinded vascular radiologist performed further subanalysis of all disagreements between conventional DSA images and 16–detector row CT angiograms in the assessment of arterial stenosis. Arterial segments were assessed if the findings of both readers agreed with each other but disagreed with the findings of conventional DSA. Possible explanations for disagreements between conventional DSA and 16–detector row CT angiograms were noted and included extensive arterial wall calcifications and configuration of arterial stenosis.
Radiation Dose Estimation
Effective radiation dose delivered during 16–detector row CT angiography and conventional DSA was calculated by a physicist (F.R.V., 11 years of experience) for the regions of the pelvis and hip, since radiation exposure of the extremities minimally contributes to the effective dose. The distance from the pelvic crest to the proximal third of the thighs, including the testicles in men, was measured in each patient. The mean distance was 26 cm for men and 20 cm for women. The weighted CT dose index indicated by the CT scanner was verified by using a 32-cm-diameter CT dose index test object and a 10-cm-long CT pencil ionization chamber (model 1035–10.3; Radcal, Monrovia, Calif) with an electrometer (MDH model 1015; Radcal) calibrated in RQR9 and RQA9 beams according to the International Electrotechnical Committee (13). Dose length products were calculated by using a normalized weighted CT dose index of 0.085 mGy/mAs, and they were converted into effective radiation dose by means of a conversion factor of 0.019 mSv/mGy · cm according to the guidelines for quality criteria for CT of the Commission of the European Communities (14).
Estimates of the effective dose of conventional DSA were calculated on the basis of the dose area product quantity corresponding to the acquisition protocol used. The dose area product is displayed by the fluoroscopy system itself, and it is representative of the total energy deposited in the examined volume. The dose area product displayed by the unit was verified according to Bochud et al (15). To convert dose area products into effective dose, a unique averaged conversion factor of 0.20 mSv/Gy · cm2 was used for both men and women according to Hart et al (16).
Statistical Analysis
Sensitivity, specificity, positive and negative predictive values, and accuracy for determination of hemodynamically significant arterial stenosis were calculated for all 35 segments together and for each of the following three vascular regions separately: (a) the aortoiliac region (including the distal aorta, common iliac, external iliac, and internal iliac arteries); (b) the femoral region (including the common femoral, superficial femoral, and deep femoral arteries); and (c) the popliteocrural region (including the popliteal, peroneal, and anterior and posterior tibial arteries, as well as the tibiofibular trunks). The 95% confidence intervals (CIs) were calculated and based on binominal probabilities. Interobserver agreement of grading vascular lesions between both readers and intermodality agreement between 16–detector row CT angiography and conventional DSA were determined by calculating values together with 95% CI (poor agreement, = 0.00; 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) (17).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
Reader 1.—Overall, reader 1 identified 973 (71.3%) of 1365 arterial segments as hemodynamically insignificant arterial stenoses (<50% luminal narrowing). A total of 392 (28.7%) significant arterial stenoses (50%-100% luminal narrowing) were depicted. Of these 392 arterial segments, 146 (37.2%) were occluded. Table 1 demonstrates the breakdown of findings by reader 1 for all 1365 arterial segments in all 39 patients.
Reader 2.—Overall, reader 2 identified 979 (71.7%) of 1365 arterial segments as hemodynamically insignificant arterial stenoses (<50% luminal narrowing) and 386 (28.3%) as significant arterial stenoses (50%–100% luminal narrowing). Of these 386 arterial segments, 144 (37.3%) were occluded. Table 1 summarizes the findings of reader 2 for all arterial segments in all patients.
Reader 1 versus Reader 2.—Table 2 summarizes interobserver agreement between readers 1 and 2 for (a) all grades of arterial stenosis (grades 1–4), (b) hemodynamically significant versus insignificant stenosis (grades 1–2 vs grades 3–4), and (c) occlusion versus nonocclusion (grades 1–3 vs grade 4) for all 35 segments together and for each of the three vascular regions separately. For all degrees of arterial stenosis for all 35 segments combined, there was excellent interobserver agreement ( = 0.89; 95% CI: 0.87, 0.91) between readers 1 and 2. There was excellent agreement between the two readers for determination of hemodynamically insignificant versus hemodynamically significant arterial stenosis ( = 0.93; 95% CI: 0.91, 0.95) and excellent agreement for diagnosis of nonocclusion versus occlusion ( = 0.97; 95% CI: 0.95, 0.99).
|
= 0.72; 95% CI: 0.69, 0.75) between 16–detector row CT angiography and conventional DSA for all degrees of arterial stenosis (grades 1–4). For diagnosis of hemodynamically insignificant (grades 1–2) versus hemodynamically significant (grades 3–4) arterial stenosis, there was excellent intermodality agreement ( = 0.90; 95% CI: 0.87, 0.93) for all three vascular regions combined. Findings at 16–detector row CT angiography and conventional DSA agreed in 341 (97.2%) of 351 arterial segments in the assessment of hemodynamically insignificant versus hemodynamically significant arterial stenosis of the aortoiliac region and in 298 (95.5%) of 312 arterial segments of the femoral region. When assessing hemodynamically insignificant versus hemodynamically significant arterial stenosis of the popliteocrural region, 16–detector row CT angiography and conventional DSA agreed in 671 (95.6%) of 702 arterial segments. Reader 1 overestimated arterial stenosis of the aortoiliac region in six (1.7%) of 351 arterial segments and underestimated arterial stenosis in four (1.1%) arterial segments when 16–detector row CT angiographic findings were compared with conventional DSA findings. In the femoral region, arterial stenosis was overestimated by reader 1 in 12 (4.2%) of 312 arterial segments and underestimated in two (0.6%) arterial segments when 16–detector row CT angiographic findings were compared with conventional DSA findings. Reader 1 overestimated arterial stenosis in 24 (3.4%) of 702 arterial segments in the popliteocrural region (Fig 2) and underestimated arterial stenosis in seven (1.0%) arterial segments. In total, reader 1 disagreed with conventional DSA findings in 55 (4.0%) of 1365 arterial segments on the basis of 16–detector row CT angiographic findings.
|
|
|
|
= 0.77; 95% CI: 0.74, 0.80) for all degrees of arterial stenosis. There was excellent intermodality agreement ( = 0.92; 95% CI: 0.90, 0.94) for all three vascular regions combined in the assessment of hemodynamically insignificant versus hemodynamically significant arterial stenosis. Conventional DSA and 16–detector row CT angiography agreed in 343 (97.7%) of 351 arterial segments of the aortoiliac region and in 301 (96.5%) of 312 arterial segments of the femoral region in the assessment of hemodynamically insignificant versus hemodynamically significant arterial stenosis. When assessing hemodynamically insignificant versus hemodynamically significant arterial stenosis of the popliteocrural region, findings at 16–detector row CT angiography and conventional DSA agreed in 677 (96.4%) of 702 arterial segments. When compared with conventional DSA findings, reader 2 overestimated arterial stenosis of the aortoiliac region in seven (2.0%) of 351 arterial segments and underestimated arterial stenosis in one (0.2%) arterial segment on the basis of 16–detector row CT angiograms. In the femoral region, reader 2 overestimated arterial stenosis in eight (2.6%) of 312 arterial segments and underestimated arterial stenosis in three (1.0%) arterial segments when compared with conventional DSA findings. Reader 2 overestimated arterial stenosis in 19 (2.7%) of 702 arterial segments in the popliteocrural region (Fig 2) and underestimated arterial stenosis in six (0.9%) arterial segments. In total, reader 2 disagreed with conventional DSA findings in 44 (3.2%) of 1365 arterial segments on the basis of 16–detector row CT angiograms.
In the subanalysis of the 55 and 44 instances in which readers 1 and 2, respectively, disagreed with 16–detector row CT angiographic and conventional DSA findings with regard to grading of arterial stenosis, both readers agreed with each other in the grading of 29 (53%) of 55 and 29 (66%) of 44 arterial segments on the basis of 16–detector row CT angiograms. Of the 29 agreements between readers, 26 (90%) were overestimations of arterial stenosis by one (n = 16, 62%) or two (n = 10, 38%) grades compared with conventional DSA findings. Three (10%) of the 29 agreements between readers were underestimations by one grade compared with conventional DSA findings. In 14 (54%) of these 26 overestimated arterial segments, there was hemodynamically significant (grade 3) posteroanterior narrowing of the arterial lumen, which was not noted on single-projection conventional DSA images (Figs 3, 4). Seven (50%) of these 14 arterial segments with posteroanterior narrowing were located in the popliteocrural region (Fig 3), five (35%) were located in the femoral region (Fig 4), and two (14.3%) were located in the aortoiliac region. In 20 (77%) of the 26 overestimated arterial segments, extensive arterial wall calcifications were noted (Fig 2). In 12 (46%) of 26 overestimated arterial segments, both extensive arterial wall calcification and posteroanterior narrowing were present. In nine (31%) of 29 arterial segments, no possible reason for disagreement between conventional DSA and 16–detector row CT angiographic findings was noted.
|
|
|
|
|
|
|
For conventional DSA, an excellent correlation was found between the displayed and measured dose area product (<5% difference). The mean dose area product delivered in the region of the pelvis and hip was 55 Gy · cm2 (range, 32–80 Gy · cm2). Thus, the whole mean effective radiation dose of conventional DSA was estimated to be 11 mSv (range, 6.4–16.0 mSv).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
When compared with conventional DSA, duplex US has shown sensitivities of 87%–92% and specificities of 95%–99% in the assessment of aortoiliac and lower extremity arteries (18–20). Duplex US is, however, operator dependent, the examinations are time consuming, and obese patients or patients with excessive bowel gas or calcified arteries are difficult to examine (21). In addition, duplex US does not provide a road map equivalent to that obtained with conventional DSA, or MR or CT angiography.
Contrast material–enhanced three-dimensional MR angiography has been proved useful in the assessment of aortoiliac and lower extremity arteries. Sensitivities of 77%–99% and specificities of 84%–100% in the detection of hemodynamically significant arterial stenosis in aortoiliac and lower extremity arteries have been reported (22–25). Although the development of moving table techniques for contrast-enhanced three-dimensional MR angiography allows coverage from the midabdominal aorta to the foot during a single injection of contrast material, covering a large volume over several vessel territories with sufficient intraarterial contrast enhancement and minimal venous overlay remains a challenge for MR angiography. In addition, the limited spatial resolution of MR angiography is another disadvantage of MR angiography, particularly when evaluating vessels with small diameters. Furthermore, MR angiography has been proved to be more uncomfortable for patients than multi–detector row CT angiography, mostly because of the noise, the need to keep still during the examination, and the lengthy imaging time of MR angiography (6).
With the introduction of multi–detector row CT scanners, CT angiography of the aortoiliac and lower extremity arteries is becoming more important. CT angiography can be performed more efficiently with multi-detector row CT scanners because of faster scanning speed and higher spatial resolution than that which was possible with single-detector helical CT scanners. In a comparative study of CT angiography of the aorta and iliac arteries that used a four–detector row CT scanner and a single-detector CT scanner, Rubin et al (10) showed that CT angiography with a four–detector row CT scanner was faster, and scanning was possible with thinner collimation and a reduced dose of contrast medium.
Recent studies have proved the high level of diagnostic accuracy of four–detector row CT angiography in the assessment of the aortoiliac and lower extremity arteries. In a comparison with conventional DSA, the study of Ofer et al (5) yielded an overall sensitivity and specificity of 90.9% and 92.4%, respectively, for the assessment of hemodynamically significant arterial stenosis (>50% luminal narrowing) from the renal to the lower extremity arteries by using a four–detector row CT scanner with a 3.2-mm effective section thickness. Through the use of a four–detector row CT scanner with an effective section thickness of 5 mm, Martin et al (7) calculated an overall sensitivity and specificity of 92.2% and 96.8%, respectively, for the detection of severe arterial stenosis (>75% luminal narrowing) for the iliac and lower extremity arteries. In another study, in which a four–detector row CT scanner with an effective section thickness of 1.25 mm was used, overall sensitivity and specificity of 92% and 99%, respectively, in the detection of hemodynamically significant arterial stenosis (>50% luminal narrowing) were calculated for the aortoiliac and renal arteries (6). However, because of the limited spatial resolution along the z-axis, evaluation of vessels with small luminal diameter by using four–detector row CT angiography is still challenging. In the study of Ofer et al (5), 14 (64%) of the 22 clinically important mismatches between conventional DSA and four–detector row CT angiography occurred in the small renal and peripheral arteries of the calves. In the study of Martin et al (7), all of the 22 segments that could not be evaluated with four–detector row CT angiography were located in the small peripheral arteries of the calves. In the study of Willmann et al (6), sensitivity in the detection of hemodynamically significant arterial stenosis of the aortoiliac arteries was lowest for the small internal iliac arteries (83%).
In this study, by using a 16–detector row CT scanner with a 0.75-mm effective section thickness, a small field of view for image reconstruction of each extremity separately, and optimized arterial contrast enhancement with a computer-assisted bolus-tracking technique, high-quality CT angiograms of the aortoiliac and lower extremity arteries could be obtained. In all patients, both readers considered image quality of all arterial segments to be sufficient for diagnosis. In this study, the improved spatial resolution obtained with a 16–detector row CT scanner is reflected in the total sensitivity and specificity (96% and 97%, respectively, for both readers) in the detection of hemodynamically significant arterial stenosis of aortoiliac and lower extremity arteries. In particular, excellent sensitivities (ie, 96% and 97% for readers 1 and 2, respectively) and specificities (ie, 95% and 96% for readers 1 and 2, respectively) for grading small popliteocrural arteries were obtained in this study.
The high level of reliability of 16–detector row CT angiography in grading arterial stenosis of aortoiliac and lower extremity arteries is reflected in the high level of interobserver agreements in this study. For all degrees of arterial stenosis, the agreement between readers 1 and 2 was excellent ( = 0.89). When assessing insignificant versus significant arterial stenosis and nonocclusion versus occlusion of arterial segments, interobserver agreement was even better ( = 0.93 and 0.97, respectively). These values are greater than the reported interobserver agreements for four–detector row CT angiography (6,7), and they suggest an improved robustness and reliability of 16–detector row CT angiography when compared with four–detector row CT angiography in the evaluation of aortoiliac and lower extremity arteries.
In a comparison with conventional DSA, readers 1 and 2 disagreed in 55 and 44 arterial segments, respectively, on the basis of 16–detector row CT angiograms. Subanalysis of these disagreements demonstrated that both readers agreed in the same 29 arterial segments with regard to grading of arterial stenosis, which suggests a correct diagnosis on the basis of 16–detector row CT angiograms in at least some of these 29 arterial segments. Agreements in 26 of 29 arterial segments were overestimations of arterial stenosis when compared with conventional DSA findings. Further subanalysis of these 26 overestimations demonstrated that in a substantial number of overestimated arterial segments (54%), there was a posteroanterior luminal narrowing that was missed on single-projection conventional DSA images. For lack of a more accurate technique, we used conventional DSA as a reference standard for evaluation of arterial stenosis in this study. However, as demonstrated in this study, limitation of this projectional technique may result in underestimation of arterial stenosis in posteroanterior luminal narrowing if additional oblique or lateral projections are not obtained at conventional DSA, particularly for evaluation of the femoral and popliteocrural vascular region. Further studies are needed to evaluate whether disagreements between conventional DSA and 16–detector row CT angiography may be reduced in a study design with routine additional oblique or lateral projections in conventional DSA for all three vascular regions. As shown in this study, image analysis of 16–detector row CT angiograms may be reduced in the presence of extensive arterial wall calcifications because of the blooming artifact (Fig 2c). Thus, we now routinely use a bone window setting in the evaluation of arterial segments when calcium is present.
Conventional DSA may also be limited in the assessment of arterial stenosis distal to arterial occlusion. Ten (2.1%) of the 480 arterial segments that we were able to assess on four–detector row CT angiograms could not be evaluated with conventional DSA in the study of Ota et al (8). In the study of Martin et al (7), 91 (86.7%) of 105 arterial segments that could not be evaluated with conventional DSA were sufficient for diagnosis on four–detector row CT angiograms. This may be explained by the fact that peripheral administration of contrast material in multi–detector row CT angiography allows better opacification of collateral circulation and, therefore, better opacification of arteries distal to an occlusion site than does central aortic administration of contrast material in conventional DSA. In our study, however, all conventional DSA images obtained in all patients were considered diagnostic by the two vascular radiologists.
The major drawback of 16–detector row CT angiography in the assessment of aortoiliac and lower extremity arteries is the potential for hazardous radiation exposure. For 16–detector row CT angiography, an effective radiation dose of 3.0 mSv for men and 2.3 mSv for women was calculated. The mean effective radiation dose caused by 16–detector row CT angiography was lower by a factor of about four for men and about five for women when compared with the mean effective radiation dose values calculated for conventional DSA. Reduction of radiation exposure was possible in our study through the use of an online modulation of tube current implemented with the 16–detector row CT scanner for all 16–detector row CT angiograms. Greess (26) demonstrated a reduction of the effective dose from 26% to 43% in children on the basis of the patient's body shape and weight by using a multi–detector row CT scanner with an attenuation-based on-line modulation of the tube current. For the pelvis and extremities, a mean effective dose reduction of 25% and 39%, respectively, could be obtained in another study by using a single–detector row CT scanner and attenuation-based on-line modulation of the tube current (27). The lower amount of administered effective radiation dose associated with 16–detector row CT angiography might be an advantage that may foster the use of 16–detector row CT angiography as an alternative to conventional DSA in the assessment of aortoiliac and lower extremity arteries in patients with peripheral arterial disease.
We acknowledge several limitations of our study. Conventional DSA images and 16–detector row CT angiograms were evaluated only for the presence of arterial stenosis and aneurysmal changes. Additional criteria of peripheral arterial disease, including the diffuseness of the disease, the length of the arterial stenosis, and the location and extent of arterial wall calcification, were not systematically included in image analysis. All these criteria may be important for therapeutic care. A possible limitation relates to the fact that electronic calipers were used for image analysis of 16–detector row CT angiograms but not for image analysis of conventional DSA images.
In this study, we did not evaluate the added diagnostic value of the three-dimensional reconstructions and interactive reformatting for image analysis of the 16–detector row CT angiograms separately. To our knowledge, a rational use of three-dimensional reconstruction techniques for interpretation of 16–detector row CT angiograms has not been defined in the literature. In a recent report, Ota el al (8) stressed the importance of using the orthonormal cross-sectional vessel diameter for analysis of tortuous iliac arteries on the basis of four–detector row CT angiograms. Further studies are warranted to address the diagnostic importance of different three-dimensional reconstruction techniques in the evaluation of 16–detector row CT angiograms.
We did not address cost issues of conventional DSA or 16–detector row CT angiography in our study. In a recent study, a possible new imaging modality was found to be cost-effective when compared with conventional DSA in patients with intermittent claudication if the costs of the new imaging modality were $300 or less, both angioplasty and bypass surgery were considered as treatment options, and the sensitivity of the new imaging modality was higher than 94% (28). In our study, sensitivity of 16–detector row CT angiography was higher than 94% for all three vascular regions, including the aortoiliac, femoral, and popliteocrural regions. Further prospective studies are needed to compare conventional DSA and 16–detector row CT angiography in the evaluation of aortoiliac and lower extremity arteries in terms of cost-effectiveness, since this analysis may also guide referring physicians to select conventional DSA or 16–detector row CT angiography.
In conclusion, this study has demonstrated in a prospective blinded comparison that 16–detector row CT angiography is feasible, accurate, and reliable in the assessment of aortoiliac and lower extremity arteries. Because of its noninvasive nature and lower effective radiation dose, 16–detector row CT angiography is an alternative to conventional DSA in the evaluation of aortoiliac and lower extremity arteries in patients with peripheral arterial disease.
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval • DSA = digital subtraction angiography
Authors stated no financial interest to disclose.
Author contributions: Guarantor of integrity of entire study, J.K.W.; study concepts, J.K.W., T.B.; study design, J.K.W.; literature research, J.K.W.; clinical studies, B.B., T.S., J.K.W., T.P., F.R.V.; data acquisition, B.B., T.S., J.K.W., T.P., F.R.V.; data analysis/interpretation, J.K.W., T.B.; statistical analysis, J.K.W., B.S.; manuscript preparation and definition of intellectual content, J.K.W.; manuscript editing, revision/review, and final version approval, all authors
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
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]
|
|
|
|
 |
|
 R. Met, S. Bipat, D. A. Legemate, J. A. Reekers, and M. J. W. Koelemay Diagnostic Performance of Computed Tomography Angiography in Peripheral Arterial Disease: A Systematic Review and Meta-analysis JAMA, January 28, 2009; 301(4): 415 - 424. [Abstract] [Full Text] [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]
|
|
|
|
 |
|
 W. Lee, J. W. Chung, Y. H. Yin, H. J. Jae, S. J. Kim, J. Ha, and J. H. Park Three-Dimensional CT Venography of Varicose Veins of the Lower Extremity: Image Quality and Comparison with Doppler Sonography Am. J. Roentgenol., October 1, 2008; 191(4): 1186 - 1191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Covey Invited Commentary RadioGraphics, May 1, 2008; 28(3): 665 - 666. [Full Text] [PDF]
|
|
|
|
|
|
J. E. Lopera, C. K. Trimmer, S. G. Josephs, M. E. Anderson, S. Schuber, R. Li, B. Dolmatch, and B. Toursarkissian Multidetector CT Angiography of Infrainguinal Arterial Bypass RadioGraphics, March 1, 2008; 28(2): 529 - 548. [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]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
T. Albrecht, E. Foert, R. Holtkamp, M. A. Kirchin, C. Ribbe, F. K. Wacker, M. Kruschewski, and B. C. Meyer 16-MDCT Angiography of Aortoiliac and Lower Extremity Arteries: Comparison with Digital Subtraction Angiography Am. J. Roentgenol., September 1, 2007; 189(3): 702 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Roos, D. Fleischmann, A. Koechl, T. Rakshe, M. Straka, A. Napoli, A. Kanitsar, M. Sramek, and E. Groeller Multipath Curved Planar Reformation of the Peripheral Arterial Tree in CT Angiography Radiology, July 1, 2007; 244(1): 281 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Nael, S. G. Ruehm, H. J. Michaely, R. Saleh, M. Lee, G. Laub, and J. P. Finn Multistation Whole-Body High-Spatial-Resolution MR Angiography Using a 32-Channel MR System Am. J. Roentgenol., February 1, 2007; 188(2): 529 - 539. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
