Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2352040089
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Leiner, T.
Right arrow Articles by van Engelshoven, J. M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Leiner, T.
Right arrow Articles by van Engelshoven, J. M. A.
(Radiology 2005;235:699-708.)
© RSNA, 2005

Vascular and Interventional Radiology

Peripheral Arterial Disease: Comparison of Color Duplex US and Contrast-enhanced MR Angiography for Diagnosis1

Tim Leiner, MD, PhD, Alphons G. H. Kessels, MD, MSc, Patricia J. Nelemans, MD, PhD, G. Boudewijn C. Vasbinder, MD, PhD, Michiel W. de Haan, MD, PhD, Peter E. J. H. M. Kitslaar, MD, PhD, Kai Yiu J. A. M. Ho, MD, PhD, Jan H. M. Tordoir, MD, PhD and Joseph M. A. van Engelshoven, MD, PhD

1 From the Departments of Radiology (T.L., G.B.C.V., M.W.d.H., K.Y.J.A.M.H., J.M.A.v.E.), Clinical Epidemiology and Medical Technology Assessment (A.G.H.K.), Epidemiology (P.J.N.), and Vascular Surgery (P.E.J.H.M.K., J.H.M.T.), Maastricht University Hospital, Peter Debijelaan 25, NL-6229 HX Maastricht, the Netherlands; and Department of Epidemiology, Maastricht University Medical School, Maastricht, the Netherlands (P.J.N.). From the 2002 RSNA Annual Meeting. Received January 17, 2004; revision requested March 31; revision received May 25; accepted July 1. Supported in part by the Dutch Heart Foundation, grant 98–150. Address correspondence to T.L. (e-mail: leiner@rad.unimaas.nl).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
PURPOSE: To prospectively compare the diagnostic accuracies of color duplex ultrasonography (US) and contrast material–enhanced magnetic resonance (MR) angiography and to assess interobserver agreement regarding contrast-enhanced MR angiographic findings in patients suspected of having peripheral arterial disease (PAD).

MATERIALS AND METHODS: The institutional review board approved the study, and all patients provided signed informed consent. Two hundred ninety-five patients referred for diagnostic and preinterventional work-up of PAD with duplex US also underwent gadolinium-enhanced MR angiography. Data sets were reviewed for presence or absence of 50% or greater luminal reduction, which indicated hemodynamically significant stenosis, and to determine interobserver agreement. At duplex US, a peak systolic velocity ratio of 2.5 or greater indicated significant stenosis. Primary outcome measures were differences between duplex US and contrast-enhanced MR angiography in sensitivity and specificity for detection of significant stenosis, as assessed with the McNemar test, and interobserver agreement between the two contrast-enhanced MR angiogram readings, expressed as quadratic weighted {kappa} values. Intraarterial digital subtraction angiography (DSA) was the reference standard.

RESULTS: Two hundred forty-nine patients had at least one hemodynamically significant stenotic lesion at contrast-enhanced MR angiography, duplex US, or both examinations. One hundred fifty-two patients underwent intraarterial DSA. The quadratic weighted {kappa} for agreement regarding the presence of 50% or greater stenosis at contrast-enhanced MR angiography was 0.89 (95% confidence interval [CI]: 0.87, 0.91). Sensitivity of duplex US was 76% (95% CI: 69%, 82%); specificity, 93% (95% CI: 91%, 95%); and accuracy, 89%. Sensitivity and specificity of contrast-enhanced MR angiography were 84% (95% CI: 78%, 89%) and 97% (95% CI: 95%, 98%), respectively; accuracy was 94%. Sensitivity (P = .002) and specificity (P = .03) of contrast-enhanced MR angiography were significantly higher.

CONCLUSION: Results of this prospective comparison between contrast-enhanced MR angiography and duplex US provide evidence that contrast-enhanced MR angiography is more sensitive and specific for diagnosis and preinterventional work-up of PAD.

© RSNA, 2005


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
The diagnosis of peripheral arterial disease (PAD) is made on the basis of the typical medical history, physical examination results, and ankle-brachial index (ABI) measurements (1,2). About 5% of patients who have intermittent claudication when they present for medical attention eventually undergo endovascular or surgical treatment (2). In the United States alone, this has led to more than 100 000 patients undergoing percutaneous or surgical treatment for PAD annually (3), and the number of vascular procedures performed throughout the Western world is rapidly increasing (4,5). The major determinants of available treatment options are the precise location, length, and severity of the atherosclerotic lesion (2).

Duplex ultrasonography (US) is widely used to determine the location, length, and severity of aortoiliac and femoropopliteal stenoses and obstructions. Duplex US was developed in the 1980s as an alternative to invasive intraarterial angiography, which is associated with local and systemic complications, and to acquire direct physiologic information about affected arteries. With duplex US, the severity of a stenosis can be determined by using peak systolic velocity (PSV) measurements in arteries with reduced luminal diameter, the PSV ratio at the site of the stenosis and the adjacent normal artery, the end-diastolic velocity, and other less firmly established criteria (6). The sensitivity and specificity of duplex US for the detection and grading of PAD are generally moderate to high, ranging from 70%–90% (6). Contrast material–enhanced magnetic resonance (MR) angiography has been developed as an alternative imaging modality for the diagnostic work-up of patients suspected of having PAD (7–9). Thus far, mainly studies in which contrast-enhanced MR angiography was compared with intraarterial digital subtraction angiography (DSA) in patients who were already referred for intraarterial DSA have been reported. These studies are summarized in two relatively recently published meta-analyses (10,11). However, there are limited data on the accuracy of duplex US compared with that of contrast-enhanced MR angiography for the diagnosis of PAD in the same patient group.

Our aims in this study were to prospectively compare the diagnostic accuracies of duplex US and contrast-enhanced MR angiography and to assess interobserver agreement regarding contrast-enhanced MR angiography findings in patients suspected of having PAD.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
Study Design and Patients
The study was performed in the radiology and vascular surgery departments of Maastricht University Hospital between September 1997 and June 2001. Patients were eligible for inclusion if they reported having intermittent claudication and/or rest pain or had tissue loss (Fontaine stages II–IV) (12,13), and they were referred for duplex US by vascular surgeons, internists, cardiologists, or any other physician because of recent complaints. Referral for duplex US was based on medical history and physical examination findings, palpation of lower extremity pulses, and diminished or absent pulsations, as determined by using a hand-held continuous wave Doppler device. ABI measurements, duplex US, and contrast-enhanced MR angiography were performed in the included patients. (Hereafter, for brevity, contrast-enhanced MR angiography usually will be referred to as MR angiography.) The sequence of the imaging procedures was determined according to the availability of the MR imaging unit.

If the results of duplex US, MR angiography, or both examinations indicated the presence of a treatable stenosis (ie, ≥50% luminal reduction) in the aortoiliac or superficial femoral arteries, intraarterial DSA was recommended and considered the reference standard. Arteries below the knee were not evaluated in the current study because duplex US of these arteries is not routinely performed at our institution. For medical ethical reasons, patients with negative duplex US and MR angiography results did not undergo intraarterial DSA and were assumed to have no significant lesions in the aortoiliac or superficial femoral arteries.

On the basis of the results of analysis of data in the interventional radiology registry at our hospital, we expected to identify treatable lesions in the aortoiliac and superficial femoral arteries in 55% of the patients and assumed that in 20% of them, duplex US and MR angiography would differ in terms of demonstrating the presence of such lesions. The study was powered for the detection of a minimum difference in sensitivity between duplex US and MR angiography of 10%. We believed that a difference of approximately this magnitude would be clinically relevant. At a significance level of .05 (two sided) and a power of 80%, the target subject enrollment was set at 300 patients. The study was approved by the institutional review board, and all patients signed informed consent forms before being enrolled.

From September 1997 to June 2001, 1430 patients were referred for duplex US of the aortoiliac and/or superficial femoral arteries. Patients who were referred for follow-up of an intervention (n = 327) and those who were previously included in the study for complaints in the contralateral lower extremity (n = 51) were excluded from participation. Of the remaining 1052 patients, 390 patients who were chosen from a random sample of names taken from the vascular laboratory agenda every week were contacted. We were unable to include consecutive patients because our MR imaging unit could accommodate a maximum of five patients per week. Ninety-five of the patients contacted did not participate. Reasons for nonparticipation were claustrophobia (n = 9), presence of a pacemaker (n = 6) or other metal implant (n = 2), lack of availability when the MR imaging unit was available (n = 8), and nonwillingness (n = 70).

A total of 295 of the contacted patients were enrolled in the study and underwent both duplex US and MR angiography. Figure 1 shows the patient flow diagram. There were 95 women and 200 men, and the overall mean age was 63 years ± 10 (standard deviation) (range, 39–87 years). The mean age of the women was 62 years ± 11 (range, 43–80 years), and that of the men was 63 years ± 10 (range, 39–87 years). The difference in age distribution between the men and women was not significant (P = .67, independent samples t test). Disease severities, mean resting and postexercise ABIs for symptomatic extremities, and risk factors are listed in Table 1. None of the 295 patients who underwent MR angiography and duplex US reported having any associated serious adverse events (eg, shortness of breath, nausea, vomiting, or other symptoms of contrast material allergy) or was medically treated or hospitalized as a result of undergoing either imaging procedure.



View larger version (51K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Outline of current study design. CE-MRA = contrast-enhanced MR angiography, DU = duplex US, IA = intraarterial. Positive and negative indicate, respectively, presence or absence of 50% or greater stenosis in any vessel segment.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. Disease Severity, Risk Factors, and Concomitant Disease

 
Duplex US
Four licensed vascular technologists, who had 3–8 years experience performing US and were blinded to the MR angiography findings, performed the duplex US examinations by using Aloka 2000 or Aloka 5500 (Aloka, Tokyo, Japan) color duplex US equipment. The abdominal aorta, common and external iliac arteries, and common femoral arteries were insonated by using a 5-MHz convex transducer. Patients were asked to fast 8–9 hours before the examination. In the presence of excessive bowel gas or when a patient weighed more than 90 kilograms, a 3.5-MHz convex transducer was used. Ninety-two extremities (45 right, 47 left) also were examined at superficial femoral artery duplex US. When the superficial femoral arteries were scanned, a 7.5-MHz linear transducer was used.

In every arterial segment that was accessible and that could be evaluated, the PSV ratio was measured and classified by using a five-category stenosis scale. In the aortoiliac and common femoral arteries, the end-diastolic velocity was also measured. Stenosis scale categories were as follows: A PSV ratio of less than 1.5 indicated 0%–19% stenosis; a PSV ratio of greater than or equal to 1.5 but less than 2.5, 20%–49% stenosis; a PSV ratio of greater than or equal to 2.5, 50%–74% stenosis; a PSV ratio of greater than or equal to 2.5 plus an end-diastolic velocity of greater than 60 cm/sec, 75%–99% stenosis; and no Doppler signal, occlusion (14,15). For the superficial femoral arteries, no differentiation was made between 0%–19% and 20%–49% stenosis or between 50%–74% and 75%–99% stenosis. The total examination time for aortoiliac duplex US was 45–60 minutes; when the superficial femoral artery was also included, the total examination time was 60–75 minutes.

Contrast-enhanced MR Angiography
All MR angiography examinations were performed by using a commercially available whole-body 1.5-T MR imaging system (Gyroscan ACS-NT and Intera; Philips Medical Systems, Best, the Netherlands) with standard hardware and software (releases 6.1, 7.1, and 8.1; Philips Medical Systems). Gradient strengths varied between 21 and 23 mT/m, and slew rates varied between 105 and 115 mT/m/msec. For depiction of aortoiliac and peripheral arteries, a stepping table approach was used (16). With this technique, high-spatial-resolution three-dimensional images of the peripheral arterial system, from the abdominal aorta to the ankles, were obtained during three acquisitions (for aortoiliac arteries and upper and lower leg arteries) in rapid succession. Patients were placed in the MR magnet bore feet first and in the supine position. Before image acquisition, venous access was established in an antecubital or hand vein by using an 18- or 19-gauge intravenous cannula (Venflon; Ohmeda, Helsingborg, Sweden).

To prescribe the imaging volumes for the MR angiogram acquisitions, nonenhanced time-of-flight images were acquired in all three stations (aortoiliac, upper leg, and lower leg arteries). We used a gradient-recalled-echo turbo field-echo pulse sequence with a flip angle sweep and a 180° inversion prepulse for optimal background tissue suppression. Parameters for the time-of-flight sequence were 15/6.9 (repetition time msec/echo time msec), a 50° flip angle, a 512 x 282-mm field of view, a 256 x 100 matrix, 20 transverse sections with a 2.5-mm thickness and a 20-mm intersection gap, and an inferior concatenated saturation band. For signal transmission and reception during the three time-of-flight acquisitions, a standard quadrature body coil was used. The total imaging time for each field of view with this sequence was 30 seconds. No cardiac synchronization was used. From these time-of-flight acquisitions, maximum intensity projections were generated in three orthogonal directions to prescribe the imaging volumes for the subsequent MR angiography examinations.

For all MR angiogram acquisitions, a gradient-recalled-echo fast field-echo pulse sequence was used; the parameters used are listed in Table 2. To enhance intravascular signal intensity, every patient was injected with a standard dose of 35 mL of 0.5 mmol/mL gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany). By using an MR imaging–compatible power injector (Medrad, Indianola, Pa), the contrast medium was administered as a single continuous bolus at rates varying between 0.4 and 1.8 mL/sec. Higher and dual-phase injection rates were used later in the study when the speed of image acquisition increased owing to improvements in gradient performance and newer software releases.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Technical Parameters Used to Perform Contrast-enhanced MR Angiography

 
For the first 43 patients, appropriate synchronization of the arrival of the contrast material with the start of the MR angiogram acquisition was achieved by using a timing sequence. This sequence consisted of a transverse single-section gradient-recalled-echo image acquisition through the abdominal aorta just proximal to the aortic bifurcation during injection of a 2-mL test bolus of contrast medium followed by an injection of 25 mL of normal saline. Nonenhanced arterial and venous blood was saturated by using superior and inferior saturation bands. The injection speeds used for the subsequent MR angiography examinations were adopted for this sequence. For the remaining 252 patients, real-time bolus monitoring software (BolusTrak; Philips Medical Systems) was used. With this monitoring software, coronal two-dimensional thick-section gradient-recalled-echo image acquisition in combination with real-time complex subtraction was used to visualize the arrival of contrast material in the vascular territory of interest. When enhancement was first seen in the descending aorta, the patient was asked to suspend breathing during image acquisition in the aortoiliac station.

Before the acquisition of all contrast-enhanced MR angiograms, identical nonenhanced angiograms were obtained; these images were subtracted from the postinjection images to increase vessel-to-background contrast. Patients were asked to perform breath holding for the duration of the aortoiliac acquisitions (15–25 seconds). After contrast-enhanced MR angiograms were acquired, they were downloaded to dedicated postprocessing workstations (UltraSparc 10; Sun Microsystems, Mountain View, Calif), where they were evaluated by using commercially available image-processing software (EasyVision, releases 3.0 and 4.0; Philips Medical Systems). The total examination time for contrast-enhanced MR angiography was approximately 25–30 minutes.

Intraarterial DSA
We performed DSA by using an x-ray system (Diagnostic Arc; Philips Medical Systems) with a programmable stepping C-arm, a film changer (Puck; Elema-Schonander, Solina, Sweden), an add-on DSA system (Technicare DR 960-B; GE Medical Systems, Milwaukee, Wis), and a power injector (Medrad) until June 1998. After system upgrades, we used Integris V5000 (Philips Medical Systems) and Polystar (Siemens Medical Systems, Erlangen, Germany) digital angiography equipment. All intraarterial DSA examinations were supervised by one of three experienced board-certified interventional radiologists, who were unaware of the duplex US and MR angiography findings. These radiologists had 7–10 years of interventional radiologic experience.

Angiograms were obtained by using 4-F universal flush catheters (Cordis, Miami, Fla) that were placed in the infrarenal aorta. Iohexol (Omnipaque, 300 mg of iodine per milliliter; Amersham Health, Eindhoven, the Netherlands) was injected at flow rates ranging from 4 to 12 mL/sec, for a total of 50–175 mL. In 13 patients, antegrade puncture angiography was performed by using a 5-F sheath (Cordis). The total amount of contrast medium used in these cases varied between 60 and 80 mL.

Time between Imaging Examinations
The mean delay between duplex US and MR angiography was 1 day ± 15 (standard deviation) (range, 60 days before to 63 days after MR angiography). The mean delay between duplex US and intraarterial DSA was 56 days ± 39 (range, 231–27 days before intraarterial DSA), and the mean delay between MR angiography and intraarterial DSA was 51 days ± 43 (range, 225–27 days before DSA).

Image Evaluation
The stenoses depicted at MR angiography and intraarterial DSA that were identified in arterial segments that were also imaged with duplex US were recorded. The following arteries were evaluated: the infrarenal aorta, the left and right common and external iliac arteries, and the common and superficial femoral arteries. Thus, a maximum of seven arterial segments per patient were evaluated. The iliac arteries were evaluated as a single arterial segment because 50% or greater stenosis in either vessel would lead to the same treatment. In each arterial segment, only the most severe stenosis was recorded. To determine the lengths and degrees of the stenoses from the MR angiography data, we used the routine clinical evaluation algorithm that was being used at our hospital at the time of the study. This algorithm involved evaluations of coronal, sagittal, and rotational maximum intensity projections. Rotational maximum intensity projections were generated at 30° increments from 0° to 150° and enabled us to view the vasculature from multiple angles.

The retrieval of MR angiography data and the generation of maximum intensity projections took about 5 minutes per patient. The severity of stenosis at MR angiography was classified by using the same five categories used for classification at duplex US and was calculated by dividing the luminal diameter at maximum stenosis by the luminal diameter of the closest adjacent normal part of the arterial segment.

At intraarterial DSA, anteroposterior views were obtained in all vascular regions. For the aortoiliac region, additional left and right anterior oblique projections at 25° were obtained. When intraarterial DSA data sets were available digitally, they were reviewed at the workstation; if no digital data sets were available, film hard-copy images were reviewed.

Four radiologists (G.B.C.V., M.W.d.H., K.Y.J.A.M.H., J.M.A.v.E.) with 6–8 years of experience interpreting contrast-enhanced MR angiograms determined the grades of the atherosclerotic lesions in the arterial segments, as defined previously. During the first reading, three radiologists evaluated 72 patient cases each, and the fourth radiologist evaluated 79 cases. Image interpretation took about 5 minutes per patient. To determine interobserver agreement, a second reading was performed after the first reading was completed. To avoid bias, the cases evaluated by each radiologist during the second reading were nearly evenly distributed among the three other radiologists. The results of the first reading were used to calculate sensitivity and specificity. Every MR angiogram reader was blinded to the findings of the other readers and to the duplex US and intraarterial DSA findings. Conversely, the duplex US image readers were blinded to the MR angiography and intraarterial DSA findings. The intraarterial DSA data sets were retrospectively read after inclusion for the study had ended, in consensus by two observers (T.L. and another observer, with, respectively, 6 and 10 years of experience in reading DSA images). The intraarterial DSA image readers were blinded to the results of the other imaging examinations. The duplex US images were interpreted by two vascular surgeons (P.E.J.H.M.K., J.H.M.T.) with, respectively, 20 and 15 years of duplex US experience.

Statistical Analyses
The McNemar test was used to test for significant differences in sensitivity and specificity between duplex US and MR angiography on an arterial segment level. Because we evaluated multiple arterial segments per patient, these observations were not independent. To correct for this dependency, we used an analysis in which clustered observations within each patient are accounted for. Accounting for this clustering is necessary for avoiding the underestimation of standard errors and for obtaining valid P values and confidence intervals (CIs) (Stata User’s Guide, 8th ed; Stata, College Station, Tex).

Sensitivity, specificity, accuracy, and corresponding 95% CIs were calculated with the assumption that patients with negative results at both duplex US and MR angiography were considered to be true-negative cases. This assumption has no influence on the testing for differences in sensitivity and specificity between duplex US and MR angiography when the McNemar test is used. Agreement regarding MR angiogram readings was expressed in quadratic weighted {kappa} values (17). To investigate whether the delay between imaging examinations differed between the patients with concordant duplex US and MR angiography results and those with discordant results, the Mann-Whitney test was used. We used a nonparametric test because it was assumed that these delay times were not normally distributed. P < .05 was considered to indicate statistical significance. All statistical analyses were performed by using SPSS 11.0 (SPSS, Chicago, Ill) and STATA 8.0 (Stata, College Station, Tex).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
Duplex US and MR Angiography Results
There was no significant difference (P = .32) in examination delay time between the patients with discordant duplex US and MR angiography results and those with concordant results. In 249 patients, at least one significant stenosis was found; 152 (61%) of these patients underwent intraarterial DSA (Fig 1). Intraarterial DSA was performed in 50 patients by using the Diagnostic Arc system, in 84 patients by using the Integris system, and in 18 patients by using the Polystar system. The remaining 97 patients were not referred for intraarterial DSA because their vascular surgeons prescribed medical and walking exercise therapy (eg, for total superficial femoral arterial occlusion) (n = 70) or because the patients declined to undergo invasive imaging and treatment because their symptoms were not severe enough (n = 27).

Six hundred sixty-eight corresponding aortoiliac and superficial femoral arterial segments were available for the comparison between duplex US, MR angiography, and intraarterial DSA findings in 152 patients. The numbers and grades of lesions identified with duplex US, MR angiography, and intraarterial DSA are listed in Table 3.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Numbers and Grades of Stenotic Lesions Detected with Duplex US, Contrast-enhanced MR Angiography, and Intraarterial DSA in Different Anatomic Regions

 
Accuracy of Duplex US and MR Angiography
The absolute numbers of arterial segments with true-positive, true-negative, false-positive, and false-negative findings of stenosis in the patients who underwent intraarterial DSA are listed in Table 4. Because the group of patients who underwent intraarterial DSA was not necessarily comparable in terms of clinical characteristics with the group of patients who were not referred for intraarterial DSA (although it should have been), the sensitivity and specificity values based on findings in only the group of patients who underwent intraarterial DSA may have been biased. To correct for this possible verification bias, we investigated the reasons that patients were not referred for intraarterial DSA when the results of duplex US, MR angiography, or both examinations indicated 50% or greater stenosis. With step-wise logistic regression analysis, we investigated whether the decision to refer a patient for intraarterial DSA was dependent on duplex US and MR angiography results or on disease characteristics (as listed in Table 1). This analysis revealed that postexercise ABI was the only significant (P = .02) predictor of the referral of patients for intraarterial DSA. Assuming that the patients with ABIs in the different postexercise categories (above the mean, below the mean, and not performed) were randomly referred for intraarterial DSA, we applied sampling weights in our analysis to obtain sensitivity and specificity estimates that were nearly unbiased in the full population (Stata User’s Guide, 8th ed; Stata, College Station, Tex). A detailed explanation of this weighting method is given in the Appendix. All statistical values were calculated by using this weighted analysis.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Results of Different Imaging Examination Combinations Used to Analyze 668 Arterial Segments in 152 Patients Who Underwent Intraarterial DSA

 
With use of the described weighting method, the sensitivity and specificity for detection and grading of hemodynamically significant stenoses were 76% (236 of 312 vessel segments; 95% CI: 69%, 82%) and 93% (933 of 1001 segments; 95% CI: 91%, 95%), respectively, for duplex US and 84% (261 of 312 segments; 95% CI: 78%, 89%) and 97% (967 of 1001 segments; 95% CI: 95%, 98%), respectively, for MR angiography. The accuracy of duplex US was 89% (1169 of 1313 segments), and the accuracy of MR angiography was 94% (1228 of 1313 segments). Differences in sensitivity (P = .002) and specificity (P = .03) between duplex US and MR angiography were significant. Figure 2 shows an example of concordant and discordant duplex US and MR angiography findings.



View larger version (38K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Contrast-enhanced MR angiography, intraarterial DSA, and duplex US images obtained in 56-year-old man with severe intermittent claudication (pain-free walking distance capability, 50 m). (a) Coronal maximum intensity projection image generated from acquired MR angiography data set and (b) corresponding intraarterial DSA image show a high-grade stenosis (80% luminal narrowing) at the border between the left common and external iliac arteries (arrow). The extent and grade of disease seen in a match those seen in b very well. (c) Duplex US scan of the arterial segment with 80% stenosis in a and b shows a PSV of 83.6 cm/sec, which corresponds to less than 50% stenosis. LI AIC M = left common iliac artery. Contrast-enhanced MR angiography (a) and intraarterial DSA (b) images also show a high-grade stenosis at the transition of the left common femoral and superficial femoral arteries (90% luminal narrowing, top arrowhead). (d) Corresponding duplex US scan of this area shows a PSV of 309 cm/sec, which also indicates high-grade stenosis in this arterial segment. LI AIE D = left common femoral artery. In addition, there is a moderate stenosis at the origin of the posterior tibial artery (bottom arrowhead in a and b, not shown at duplex US). The image quality of the lower leg arteries at contrast-enhanced MR angiography is comparable to that at intraarterial DSA. In b, the angiogram of the right lower leg is detached from the coronal image because it is a lateral projection.

 


View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Contrast-enhanced MR angiography, intraarterial DSA, and duplex US images obtained in 56-year-old man with severe intermittent claudication (pain-free walking distance capability, 50 m). (a) Coronal maximum intensity projection image generated from acquired MR angiography data set and (b) corresponding intraarterial DSA image show a high-grade stenosis (80% luminal narrowing) at the border between the left common and external iliac arteries (arrow). The extent and grade of disease seen in a match those seen in b very well. (c) Duplex US scan of the arterial segment with 80% stenosis in a and b shows a PSV of 83.6 cm/sec, which corresponds to less than 50% stenosis. LI AIC M = left common iliac artery. Contrast-enhanced MR angiography (a) and intraarterial DSA (b) images also show a high-grade stenosis at the transition of the left common femoral and superficial femoral arteries (90% luminal narrowing, top arrowhead). (d) Corresponding duplex US scan of this area shows a PSV of 309 cm/sec, which also indicates high-grade stenosis in this arterial segment. LI AIE D = left common femoral artery. In addition, there is a moderate stenosis at the origin of the posterior tibial artery (bottom arrowhead in a and b, not shown at duplex US). The image quality of the lower leg arteries at contrast-enhanced MR angiography is comparable to that at intraarterial DSA. In b, the angiogram of the right lower leg is detached from the coronal image because it is a lateral projection.

 


View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c. Contrast-enhanced MR angiography, intraarterial DSA, and duplex US images obtained in 56-year-old man with severe intermittent claudication (pain-free walking distance capability, 50 m). (a) Coronal maximum intensity projection image generated from acquired MR angiography data set and (b) corresponding intraarterial DSA image show a high-grade stenosis (80% luminal narrowing) at the border between the left common and external iliac arteries (arrow). The extent and grade of disease seen in a match those seen in b very well. (c) Duplex US scan of the arterial segment with 80% stenosis in a and b shows a PSV of 83.6 cm/sec, which corresponds to less than 50% stenosis. LI AIC M = left common iliac artery. Contrast-enhanced MR angiography (a) and intraarterial DSA (b) images also show a high-grade stenosis at the transition of the left common femoral and superficial femoral arteries (90% luminal narrowing, top arrowhead). (d) Corresponding duplex US scan of this area shows a PSV of 309 cm/sec, which also indicates high-grade stenosis in this arterial segment. LI AIE D = left common femoral artery. In addition, there is a moderate stenosis at the origin of the posterior tibial artery (bottom arrowhead in a and b, not shown at duplex US). The image quality of the lower leg arteries at contrast-enhanced MR angiography is comparable to that at intraarterial DSA. In b, the angiogram of the right lower leg is detached from the coronal image because it is a lateral projection.

 


View larger version (95K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d. Contrast-enhanced MR angiography, intraarterial DSA, and duplex US images obtained in 56-year-old man with severe intermittent claudication (pain-free walking distance capability, 50 m). (a) Coronal maximum intensity projection image generated from acquired MR angiography data set and (b) corresponding intraarterial DSA image show a high-grade stenosis (80% luminal narrowing) at the border between the left common and external iliac arteries (arrow). The extent and grade of disease seen in a match those seen in b very well. (c) Duplex US scan of the arterial segment with 80% stenosis in a and b shows a PSV of 83.6 cm/sec, which corresponds to less than 50% stenosis. LI AIC M = left common iliac artery. Contrast-enhanced MR angiography (a) and intraarterial DSA (b) images also show a high-grade stenosis at the transition of the left common femoral and superficial femoral arteries (90% luminal narrowing, top arrowhead). (d) Corresponding duplex US scan of this area shows a PSV of 309 cm/sec, which also indicates high-grade stenosis in this arterial segment. LI AIE D = left common femoral artery. In addition, there is a moderate stenosis at the origin of the posterior tibial artery (bottom arrowhead in a and b, not shown at duplex US). The image quality of the lower leg arteries at contrast-enhanced MR angiography is comparable to that at intraarterial DSA. In b, the angiogram of the right lower leg is detached from the coronal image because it is a lateral projection.

 
Interobserver Agreement Regarding MR Angiogram Readings
The quadratic weighted {kappa} value for agreement between the two MR angiography data readings regarding the five categories of stenosis was 0.89 (95% CI: 0.87, 0.91).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
APPENDIX
 REFERENCES
 
In many hospitals in the Western world, duplex US and contrast-enhanced MR angiography are now used as noninvasive or minimally invasive alternatives to diagnostic intraarterial DSA for the detection and grading of PAD and for treatment planning. Noninvasive or minimally invasive imaging for assessment of PAD is desirable because it may enable a patient to avoid undergoing diagnostic intraarterial DSA, a more expensive procedure that is associated with low but nonnegligible rates of local and systemic complications. The most frequently encountered local complications of intraarterial angiography are hematoma, arterial wall dissection, and thrombosis, which occur in up to 7.3% of patients undergoing the procedure (18). In addition, up to 1.8% of patients may experience systemic complications (18).

However, both duplex US and MR angiography have limitations, and at present it is unclear which modality is more accurate in the diagnostic work-up of patients suspected of having PAD. The fact that there was little available information about the accuracy of duplex US versus that of contrast-enhanced MR angiography for the detection of suspected aortoiliac and superficial femoral PAD—especially in the same patients—motivated us to perform the current study.

In this study, the use of contrast-enhanced MR angiography instead of duplex US enabled better discrimination between patients with and those without significant stenosis in the aortoiliac or femoropopliteal arteries. It was shown that MR angiography is more sensitive than duplex US for the detection of angiographically significant stenosis in the diagnostic work-up of patients who present with intermittent claudication. The higher sensitivity of MR angiography is combined with its higher specificity; this means that the improved detection of stenoses with MR angiography is not achieved at the cost of more false-positive diagnoses of stenotic lesions.

In addition, the agreement between experienced contrast-enhanced MR angiography data readers regarding the classification of atherosclerotic lesions in the aortoiliac and superficial femoral arteries was observed to be high. The agreement between the two MR angiography data readings (quadratic {kappa}, 0.89) compares favorably with the {kappa} value reported in a relatively recent study of duplex US of the aortoiliac arteries (0.53) (19). The high interobserver agreement indicates that dedicated readers are interchangeable, and this interchangeability makes the technique acceptable in clinical practice.

The results of the present study corroborate the findings of the meta-analysis performed by Visser and Hunink (20), in which contrast-enhanced MR angiography was found to have greater discriminatory power over duplex US in the work-up of patients with PAD. In that study, a total of 295 patients were enrolled in 10 MR angiography studies in which examination results were analyzed. For duplex US, 21 studies, in which the examination results for a total of 1291 patients were analyzed, were performed. The pooled sensitivities of contrast-enhanced MR angiography and duplex US were 97.5% and 87.6%, respectively. Specificities were 96.2% for MR angiography and 94.7% for duplex US.

Thus far, two studies (21,22) in which duplex US was compared with contrast-enhanced MR angiography in the same patients have been published. In both studies, results similar to those observed in the current study were reported: sensitivities and specificities of duplex US of 72% and 88%, respectively (21), and of 72% and 97%, respectively (22). Sensitivities and specificities of contrast-enhanced MR angiography were 81% and 86%, respectively (21), and 88% and 92%, respectively (22). In the study by Wikstrom et al (21), however, there was considerable selection bias because nonconsecutive patients with iliac artery stenosis found at duplex US also underwent contrast-enhanced MR angiography and intraarterial DSA. In the study by Lundin et al (22), 39 patients who were already referred for intraarterial DSA also underwent contrast-enhanced MR angiography and duplex US.

In contrast, in the current prospective study, patients were randomly selected from a population of subjects who were referred for noninvasive evaluation because of reports of PAD from their vascular surgeons, and in this respect, this study was more representative of actual clinical practice. Although we were not able to include consecutive patients, the study population represented a random sample of patients who were scheduled to undergo duplex US in a given week during the study inclusion period.

A limitation of this study was that not all of the patients who had 50% or greater atherosclerotic lesions at duplex US, contrast-enhanced MR angiography, or both examinations underwent intraarterial DSA. To adjust the sensitivity and specificity values to account for this bias, we used sampling weights (Appendix). Surprisingly, the results of neither duplex US nor MR angiography, in terms of the presence or absence of greater than 50% stenosis, seemed to have a significant influence in the decision of whether or not to refer patients for intraarterial DSA. In addition to correcting for verification bias, we assumed that patients who had both negative duplex US results and negative contrast-enhanced MR angiography results did not have stenotic lesions, so a small number of segments with 50% or greater stenosis probably were falsely neglected. The effects of this assumption on the results of the study were that sensitivity and specificity values were probably slightly overestimated. However, this assumption influenced only the absolute estimates of sensitivity and specificity and had no influence on the testing of differences in sensitivity and specificity.

In addition, at logistic regression analysis, we found that only postexercise ABI was significantly associated with referral for intraarterial DSA. However, this finding does not exclude the fact that other important variables that were not considered in the present study are associated with referral for intraarterial DSA. Furthermore, the sensitivity and specificity values that we report were not directly derived from observations regarding the patients who underwent intraarterial DSA; rather, they are adjusted estimates, as explained in the Appendix.

Another limitation was the choice of intraarterial DSA as the reference standard. A better reference standard for the evaluation of aortoiliac lesions would have been intraarterial pressure measurements. This is because it is known that the stenosis percentages measured on intraarterial DSA images do not always correlate with either the true three-dimensional morphologic features or the hemodynamic effects of a given stenosis (21). For evaluation of the contrast-enhanced MR angiography data sets, it would have been better if source images or cross-sectional stenosis measurements also had been available. It is known that reading source images derived from contrast-enhanced MR angiography data sets increases diagnostic accuracy (10,23,24). In the current study, this probably would have resulted in MR angiography having higher sensitivity and specificity.

A final limitation is that different US transducers (3.5, 5.0, and 7.0 MHz) were used in different body regions and in patients with different weights. Although this factor may have influenced the results, this protocol was in keeping with current best clinical practices (25).

What are the implications of this study for clinical practice? Patients with PAD are best served when they undergo as little testing as possible to establish a diagnosis and to plan the appropriate therapy. The key clinical question is whether either contrast-enhanced MR angiography or duplex US is useful for predicting which patients are candidates for a simple local procedure (ie, percutaneous transluminal angioplasty or local endarterectomy) to treat focal disease. If either examination yields accurate results, the referring clinician can confidently inform a patient that angiography can be performed with percutaneous transluminal angioplasty at the same time or, conversely, that the disease is too diffuse and long segmented or an occlusive lesion is present that requires a more extensive or complex procedure such as aortofemoral bypass.

Many patients and some surgeons might be interested in treatment if angioplasty were possible, but they might be less interested if a major reconstruction were required. This approach enables one to avoid performing angiography in patients who are not good candidates for or do not desire major reconstructive surgery. In the current study, contrast-enhanced MR angiography and duplex US were nearly equally reliable in aiding in this selection for treatment of the superficial femoral artery, but MR angiography was more reliable in aiding in this selection for treatment of the aortoiliac arteries.

The contrast-enhanced MR angiography strategy that we used is robust for morphologic evaluation of peripheral arteries. It enables three-dimensional viewing of the peripheral arteries from the infrarenal aorta down to the ankles while allowing the patient to avoid undergoing intraarterial DSA. The extensive anatomic coverage and relatively short examination time are major advantages of the described contrast-enhanced MR angiography technique (16,26). Although duplex US also yields reliable information about peripheral arteries, it has lower accuracy, takes considerably longer to perform, and is subject to greater interobserver variation (19). On the other hand, the use of recently developed US contrast media may increase the diagnostic accuracy of duplex US, although experience with these agents is currently limited (27).

Another more recently developed three-dimensional imaging modality that shows promise for imaging peripheral arteries is multi–detector row computed tomographic (CT) angiography (28). However, with CT angiography, as with intraarterial DSA, concerns about contrast medium nephrotoxicity and radiation exposure remain, as was shown in a study by Cochran et al (29). However, adverse reactions to intravenously injected gadolinium compounds also occur, but these reactions are generally less severe and occur at a much lower frequency than do those to CT contrast media, as was shown by Cochran et al (29).

In view of the results of the current study, we believe that the use of contrast-enhanced MR angiography as a primary imaging work-up examination for PAD in hospitals that have the necessary expertise and equipment is justified. We are currently investigating, in a multicenter trial, whether the use of contrast-enhanced MR angiography is cost-effective, given that it is more expensive than duplex US. In the event that contrast-enhanced MR angiography reveals equivocal findings, targeted duplex US of lesions of questionable severity can then be performed before proceeding to intraarterial DSA. In this context, it should be mentioned that not every patient is a suitable candidate for contrast-enhanced MR angiography because of contraindications such as claustrophobia and/or the presence of a pacemaker or certain other ferromagnetic implants (eg, some intracerebral vascular clips or otologic prostheses).

A technical limitation of contrast-enhanced MR angiography is that it does not yield information about stenoses and obstructions in most types of stents because of artifacts (30). A further drawback of contrast-enhanced MR angiography is that it is currently not as standardized as duplex US: Protocols are hardware and software dependent, and it is not clear if patients with PAD of different degrees of severity can be examined with the same protocol.

In conclusion, the findings of this study indicate that contrast-enhanced MR angiography is highly accurate for the detection and grading of PAD, given that the necessary equipment and expertise are available. Contrast-enhanced MR angiography has significantly and clinically important higher sensitivity compared with duplex US and is slightly more effective for diagnosing disease because of its higher specificity. On the basis of these findings and the potential capability of contrast-enhanced MR angiography to yield inflow and outflow information, we believe that contrast-enhanced MR angiography will have an increasingly important role in the diagnostic work-up of suspected PAD.


   APPENDIX
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
REFERENCES
 
Sampling weights for reconstruction of the total sample can be used to correct point estimates of sensitivity and specificity, and the corresponding CIs, in cases in which only a part of the observations are validated with the reference standard technique. The weighting technique is performed with the assumption that observations are selected through a random process, but different observations may have different probabilities of selection. With logistic regression analysis, we found that patients with more severe PAD (as indicated by a postexercise ABI below the mean) were more likely to be referred for intraarterial DSA compared with the other patients. Surprisingly, the results of neither duplex US nor contrast-enhanced MR angiography—specifically, the presence or absence of greater than 50% stenosis—significantly influenced referrals for intraarterial DSA. Furthermore, we found no other variables to be significantly associated with the referral rate and assumed that the referral for DSA within each stratum, as specified according to the postexercise ABI (above the mean, below the mean, or not measured), was random.

In the group of subjects with a postexercise ABI below the mean and positive duplex US and/or contrast-enhanced MR angiography results, 70% (78 of 111) of patients were referred for intraarterial DSA. In the group of patients with a postexercise ABI above the mean, this percentage was 53% (45 of 85 patients), and in the group in which the ABI measurement was not successfully obtained, this percentage was 55% (29 of 53 patients). By weighting these strata with the inverse of these percentages, the original sample of patients who should have undergone intraarterial DSA can be reconstructed. Consequently, the first stratum should be weighted by 1.43 (1/0.70); the second stratum, by 1.89 (1/0.53); and the third stratum, by 1.82 (1/0.55).

Table A1 is a cross table for findings in the cohort of 249 patients who should have undergone intraarterial DSA after the described weights were applied to the data in Table 4. The data cited in this table were used to calculate sensitivity and specificity; however, this table cannot be used to determine variances because the cell numbers are artificially increased. Therefore, 95% CIs (not shown in Table) were calculated and McNemar testing for significant differences in sensitivity and specificity was performed with robust variances, according to the methods of Huber (31) and White (32,33).


View this table:
[in this window]
[in a new window]

  		TABLE A1. Results of Different Imaging Examination Combinations, with Data Reconstructed after Weighting

 


   FOOTNOTES
 
Abbreviations: ABI = ankle-brachial index, CI = confidence interval, DSA = digital subtraction angiography, PAD = peripheral arterial disease, PSV = peak systolic velocity

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, T.L., A.G.H.K., P.J.N., J.M.A.v.E.; study concepts, T.L., K.Y.J.A.M.H., J.M.A.v.E.; study design, T.L., A.G.H.K., P.J.N., P.E. J.H.M.K., K.Y.J.A.M.H., J.M.A.v.E.; literature research, T.L.; clinical studies, T.L.; data acquisition, T.L., G.B.C.V., M.W.d.H., P.E.J.H.M.K., J.H.M.T.; data analysis/interpretation, T.L., A.G.H.K., P.J.N., J.M.A.v.E.; statistical analysis, T.L., A.G.H.K., P.J.N.; manuscript preparation and editing, T.L.; manuscript definition of intellectual content, revision/review, and final version approval, all authors


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 

  1. Stoffers HE, Kester AD, Kaiser V, Rinkens PE, Knottnerus JA. Diagnostic value of signs and symptoms associated with peripheral arterial occlusive disease seen in general practice: a multivariable approach. Med Decis Making 1997; 17:61-70.[Abstract/Free Full Text]
  2. Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD): TASC Working Group and TransAtlantic Inter-Society Concensus (TASC). J Vasc Surg 2000; 31(1 pt 2):S1-S296.[CrossRef][Medline]
  3. Rutkow IM, Ernst CB. An analysis of vascular surgical manpower requirements and vascular surgical rates in the United States. J Vasc Surg 1986; 3:74-83.[CrossRef][Medline]
  4. Lepantalo M, Salenius JP, Harjola PT. Trends in vascular surgery: an evaluation of operative activity in Finland 1976–1992. Ann Chir Gynaecol 1996; 85:225-229.[Medline]
  5. Loftus IM, Reid A, Thompson MM, London NJ, Bell PR, Naylor AR. The increasing workload required to maintain infrainguinal bypass graft patency. Eur J Vasc Endovasc Surg 1998; 15:337-341.[CrossRef][Medline]
  6. Koelemay MJ, den Hartog D, Prins MH, Kromhout JG, Legemate DA, Jacobs MJ. Diagnosis of arterial disease of the lower extremities with duplex ultrasonography. Br J Surg 1996; 83:404-409.[Medline]
  7. Ho KY, Leiner T, de Haan MW, Kessels AG, Kitslaar PJ, van Engelshoven JM. Peripheral vascular tree stenoses: evaluation with moving-bed infusion-tracking MR angiography. Radiology 1998; 206:683-692.[Abstract/Free Full Text]
  8. Meaney JF, Ridgway JP, Chakraverty S, et al. Stepping-table gadolinium-enhanced digital subtraction MR angiography of the aorta and lower extremity arteries: preliminary experience. Radiology 1999; 211:59-67.[Abstract/Free Full Text]
  9. Ruehm SG, Hany TF, Pfammatter T, Schneider E, Ladd M, Debatin JF. Pelvic and lower extremity arterial imaging: diagnostic performance of three-dimensional contrast-enhanced MR angiography. AJR Am J Roentgenol 2000; 174:1127-1135.[Abstract/Free Full Text]
  10. Nelemans PJ, Leiner T, de Vet HC, van Engelshoven JM. Peripheral arterial disease: meta-analysis of the diagnostic performance of MR angiography. Radiology 2000; 217:105-114.[Abstract/Free Full Text]
  11. Koelemay MJ, Lijmer JG, Stoker J, Legemate DA, Bossuyt PM. Magnetic resonance angiography for the evaluation of lower extremity arterial disease: a meta-analysis. JAMA 2001; 285:1338-1345.[Abstract/Free Full Text]
  12. Becker F. Exploration of arterial function with noninvasive technics: results in chronic arterial occlusive disease of the lower limbs according to Leriche and Fontaine classification. Int Angiol 1985; 4:311-322.[Medline]
  13. Rutherford RB, Baker JD, Ernst C, et al. Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg 1997; 26:517-538.[CrossRef][Medline]
  14. de Smet AA, Ermers EJ, Kitslaar PJ. Duplex velocity characteristics of aortoiliac stenoses. J Vasc Surg 1996; 23:628-636.[CrossRef][Medline]
  15. de Smet AA, Kitslaar PJ. A duplex criterion for aorto-iliac stenosis. Eur J Vasc Surg 1990; 4:275-278.[CrossRef][Medline]
  16. Leiner T, Ho KY, Nelemans PJ, de Haan MW, van Engelshoven JM. Three-dimensional contrast-enhanced moving-bed infusion-tracking (MoBI-track) peripheral MR angiography with flexible choice of imaging parameters for each field of view. J Magn Reson Imaging 2000; 11:368-377.[CrossRef][Medline]
  17. Fleiss JL, Cohen J. The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educ Psychol Meas 1973; 33:613-619.[CrossRef]
  18. Waugh JR, Sacharias N. Arteriographic complications in the DSA era. Radiology 1992; 182:243-246.[Abstract/Free Full Text]
  19. Ubbink DT, Fidler M, Legemate DA. Interobserver variability in aortoiliac and femoropopliteal duplex scanning. J Vasc Surg 2001; 33:540-545.[CrossRef][Medline]
  20. Visser K, Hunink MG. Peripheral arterial disease: gadolinium-enhanced MR angiography versus color-guided duplex US—a meta-analysis. Radiology 2000; 216:67-77.[Abstract/Free Full Text]
  21. Wikstrom J, Holmberg A, Johansson L, et al. Gadolinium-enhanced magnetic resonance angiography, digital subtraction angiography and duplex of the iliac arteries compared with intra-arterial pressure gradient measurements. Eur J Vasc Endovasc Surg 2000; 19:516-523.[CrossRef][Medline]
  22. Lundin P, Svensson A, Henriksen E, et al. Imaging of aortoiliac arterial disease: duplex ultrasound and MR angiography versus digital subtraction angiography. Acta Radiol 2000; 41:125-132.[CrossRef][Medline]
  23. Hany TF, Schmidt M, Davis CP, Gohde SC, Debatin JF. Diagnostic impact of four postprocessing techniques in evaluating contrast-enhanced three-dimensional MR angiography. AJR Am J Roentgenol 1998; 170:907-912.[Abstract/Free Full Text]
  24. Hernandez-Hoyos M, Orkisz M, Puech P, Mansard-Desbleds C, Douek P, Magnin IE. Computer-assisted analysis of three-dimensional MR angiograms. RadioGraphics 2002; 22:421-436.[Abstract/Free Full Text]
  25. Zwolak RM. Arterial duplex scanning. In: Rutherford RB, eds. Vascular surgery. 5th ed. Philadelphia, Pa: Saunders, 2000; 207-208.
  26. Ruehm SG, Goyen M, Barkhausen J, et al. Rapid magnetic resonance angiography for detection of atherosclerosis. Lancet 2001; 357:1086-1091.[CrossRef][Medline]
  27. Ubbink DT, Legemate DA, Llull JB. Color-flow duplex scanning of the leg arteries by use of a new echo-enhancing agent. J Vasc Surg 2002; 35:392-396.[CrossRef][Medline]
  28. Rubin GD, Schmidt AJ, Logan LJ, Sofilos MC. Multi–detector row CT angiography of lower extremity arterial inflow and runoff: initial experience. Radiology 2001; 221:146-158.[Abstract/Free Full Text]
  29. Cochran ST, Bomyea K, Sayre JW. Trends in adverse events after IV administration of contrast media. AJR Am J Roentgenol 2001; 176:1385-1388.[Abstract/Free Full Text]
  30. Hilfiker PR, Quick HH, Debatin JF. Plain and covered stent-grafts: in vitro evaluation of characteristics at three-dimensional MR angiography. Radiology 1999; 211:693-697.[Abstract/Free Full Text]
  31. Huber PJ. The behavior of maximum likelihood estimates under non-standard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, Calif 1967; 221-233.
  32. White H. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 1980; 48:817-830.[CrossRef]
  33. White H. Maximum likelihood estimation of misspecified models. Econometrica 1980; 50:1-15.



This article has been cited by other articles:


Home page
 CirculationHome page
M. Schillinger, S. Sabeti, P. Dick, J. Amighi, W. Mlekusch, O. Schlager, C. Loewe, M. Cejna, J. Lammer, and E. Minar
Sustained Benefit at 2 Years of Primary Femoropopliteal Stenting Compared With Balloon Angioplasty With Optional Stenting
Circulation, May 29, 2007; 115(21): 2745 - 2749.
[Abstract] [Full Text] [PDF]

Home page
 Occup Med (Lond)Home page
A. Thompson and R. House
Hand-arm vibration syndrome with concomitant arterial thrombosis in the hands
Occup. Med., August 1, 2006; 56(5): 317 - 321.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Leiner, T.
Right arrow Articles by van Engelshoven, J. M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Leiner, T.
Right arrow Articles by van Engelshoven,