| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full paper |
Departments of 1Diagnostic Radiology and 2Surgery, University of Würzburg, Josef-Schneider-Strasse 2, D-97080 Würzburg, Germany
Correspondence: Dr H-P Dinkel, Institut für Diagnostische Radiologie, Inselspital, CH 3010 Bern, Switzerland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Selection of appropriate candidates for CEA relied on angiography in the NASCET study and angiography and/or Doppler ultrasound in the ECST study. However, the Ad Hoc Committee to the Joint Council of the Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery has included pre-operative angiography in its practice guidelines [5]. Supra-aortic catheter angiography itself carries a risk of stroke [6, 7]. The ACAS trial has reported an angiography stroke risk of more than 2% [2]. It has therefore been proposed that a non-invasive test may be sufficient for assessment of the extracranial carotid arteries prior to CEA [8].
The purpose of the present study was to assess the accuracy of colour flow Doppler ultrasound (CFD) evaluation of the carotid arteries and to evaluate its ability to replace angiography in the pre-operative work-up prior to CEA.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Both CFD and digital subtraction angiography (DSA) were available for comparison in 116 patients, most of whom had symptomatic cerebrovascular disease. The CFD examinations were performed on an Angiodynograph Quantum 2000 (Siemens, Erlangen, Germany). The diagnosis of stenosis was based on spectral analysis, and the degree of stenosis was estimated by the Doppler flow velocity in the carotid bifurcation. In the presence of cardiac arrhythmia, flow velocity was averaged over at least six cardiac cycles. One patient with a severe tandem lesion and pseudo-occlusion (which was prospectively diagnosed as such by CFD) was excluded from the evaluation. Indications for CEA depended on the patient's symptoms and the degree of stenosis. The degree of stenosis was classified with cut-offs of equal to or greater than 50%, 70% and 90%. At our institution, we use a cut-off of 2.3 m s-1 systolic peak flow in the internal carotid artery to describe 70% NASCET type stenosis and a 1.8 m s-1 cut-off for a 50% diameter stenosis, as described by Hunink et al [9]. A 90% stenosis was assumed when peak flow was 3.5 m s-1 or greater. The degree of stenosis used in this study was derived from the written report, which also stated the Doppler flow velocities. The CFD examination was performed before angiography in 111 of 116 patients by three radiologists with 3–8 years of experience with the method.
Angiography was performed as intraarterial (IA) DSA on a Siemens Polydoros (Siemens, Erlangen, Germany) by a fellow or consultant radiologist via a femoral approach. It included aortic arch angiography with a 5 F pigtail catheter and imaging in at least two planes (30° right anterior oblique and 45° left anterior oblique with slight contralateral rotation of the neck) with power injection of 30 ml of 300 mg ml-1 iodinated non-ionic contrast at 15 ml s-1. Selective carotid angiography (using a headhunter H1 or sidewinder S1 catheter) was performed in 81 of 116 cases where the presence of a stenosis was not adequately demonstrated by the arch aortogram. Two to three angiographic projections were obtained per bifurcation during manual injection of 6–10 ml of a mixture of 50% contrast medium diluted with 50% isotonic saline. 11 selective carotid angiograms were performed by the Department of Neuroradiology by three board-certified radiologists on a Philips DVI-S digital subtraction angiography unit (Philips, Eindhoven, The Netherlands). The degree of stenosis was assessed by two independent observers blinded to clinical data, who reviewed the radiographs, taking compass measurement by means of overhead projection, tabulating diameters at the point of the most severe narrowing (A) and at the distal non-stenotic segment (B), according to the NASCET criteria. The NASCET criteria define the degree of stenosis as 1 minus the ratio of the diameter in the point of maximal stenosis (A) to the distal vessel diameter of the affected internal carotid artery (B) on the projection demonstrating maximal stenosis multiplied by 100% [1]. Biplanar IA-DSA was used as the gold standard. We evaluated the feasibility and potential of CFD to diagnose and quantify a haemodynamically significant stenosis [8]. The consequences for patient management as well as diagnostic accuracy and complications were also assessed. The contralateral bifurcation was included in the evaluation whenever it was sufficiently depicted by angiography in two planes, which yielded a total of 225 carotid bifurcations for evaluation in this study.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
shows the performance of CFD in detecting carotid stenosis in the 116 patients where biplanar carotid angiography was available as the gold standard.
|
One patient with cerebrovascular insufficiency (CVI) suffered sudden onset of hemiparesis and dysphasia during selective angiography performed by an experienced neuroradiologist. The neurological deficit was only partially reversible. This resulted in a procedure-related stroke rate for carotid angiography of 0.8%. The surgical stroke rate was 1.6% and the mortality was 0.8%.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The potential limitations of CFD may arise if the patient has a short neck, a high bifurcation or heavily calcified plaques, which can impede depiction of the bifurcation. However, it is still possible to estimate the presence of stenosis in many such cases since small sonic windows sufficient to obtain Doppler spectral signals often still exist in the calcified areas. Moreover, each of these potential problems can be identified during the examination and the decision for further imaging can be taken at that time [8]. When interpreting the results given in Table 1
it must be borne in mind that the data do not distinguish between equivocal and unequivocal CFD results. This means that even examinations where the examiner had limited vision counted as a false positive or false negative examination when they deviated from the angiographic findings. If these examinations (six "false positives" and one "false negative" for the 50% cut-off) were disregarded, the diagnostic performance of the test would be far higher, approaching 99% sensitivity and specificity.
Our data show that the accuracy of CFD is quite independent of the selected cut-off for the definition of significant stenosis. The accuracy, specificity, sensitivity and predictive values, even when including inconclusive CFD, are typically around or above 90%, which is characteristic of an excellent diagnostic test. This is compatible with data reported by other authors [12, 13] and means that CFD can be used as the sole pre-operative imaging test [14].
The indications for DSA of the carotid arteries are relative, and in our experience DSA is not indicated in cases with an unequivocal CFD diagnosis of significant stenosis. DSA is also unnecessary when CFD has ruled out stenosis. However, DSA is indicated in cases with equivocal CFD results if therapeutic consequences exist, and may be indicated in patients with symptomatic CVI and intermediate grade stenosis by CFD. 9 (7.8%) of our 116 carotid angiograms were indicated because of equivocal CFD results, leaving 106 (91.4%) of the 116 IA carotid DSAs unnecessary. In only one case did DSA yield relevant results not previously suspected after CFD.
Similar experiences were reported by Dawson et al [8] who in a series of 103 patients found only a single case where angiography contributed relevant information. Economical factors gain added importance in times of cost containment. The German interhospital feeing system charges the equivalent of 700 Euro for an IA-DSA. Our institute could save approximately 50 000 Euro per annum if it employed DSA selectively before CEA [15]. Charges for carotid DSA can be as high as $5000 per examination in other countries, for instance the USA, creating an even higher potential for savings. Neither magnetic resonance angiography (MRA) nor DSA is cost effective. In a hypothetical model study, Kent et al [16] found MRA resulted in an additional cost of $22 400 and DSA in an additional cost of $99 200 per quality-adjusted life-year gained. A model calculation by Kraiss et al [17] showed that dispensing with carotid angiography before carotid surgery reduced effective costs by about $5000 per patient without affecting the quality of treatment. In a two-armed study by Garrard [18], one arm with and one without carotid DSA, per patient costs were $6000 higher in the DSA group while the clinical outcome was equal. A number of similarly encouraging reports have been published [19–22].
Even more important than the financial aspects is the effect angiography has on total patient benefit. Catheter angiography of the supra-aortic vessels bears a recognized risk of morbidity. The reported (major) stroke rates in large studies covering over 13 500 examinations varied between 0.4% and 3.7% depending on the degree of training of the examiner [1, 4, 6, 7]. One of the patient's with CVI in the present study suffered an incompletely reversible hemiparesis during the procedure, giving an overall stroke rate of 0.8%. This must be balanced against the peri-operative stroke rates and the expected benefit. Peri-operative stroke rates in excellent surgical centres lie between 2.1% and 5.8% [1]. Thus, the angiography-related morbidity might amount to as much as one-quarter to one-half of all peri-operative morbidity.
One limitation, which has led to criticism of the use of pre-operative Doppler ultrasound alone, is its inability to evaluate intracranial lesions. Recent improvements in Doppler techniques have enabled increasingly accurate depiction of intracranial circulation using transcranial Doppler ultrasound [23]. Tandem lesions may constitute a confounding factor by affecting velocity measurements at the level of the bifurcation. It must be noted, however, that the clinical significance of intracranial lesions is not yet clear and that tandem lesions seem to have little relevance in clinical practice [21]. There are little hard data on the prognosis of patients with intracranial stenosis because, for example, the NASCET study excluded patients with intracranial stenosis more severe than the extracranial stenosis. Roederer et al [24] found a prevalence of 9% of tandem lesions, which, however, did not alter surgical indications, outcome or severity of pre-operative symptoms; nor was there a correlation between intracranial stenosis and recurrent symptoms of CVI after CEA. Analogously, Schuler et al [25] found that clinical improvement after CEA occurred regardless of the presence of intracranial stenosis. More importantly, intracranial disease did not increase the peri-operative risk [25]. Mackey et al [26] found no difference in outcome and peri-operative risk in 463 patients with or without concomitant intracranial stenosis. There is even evidence that excluding surgery on bifurcation stenosis in patients with tandem lesions as a pre-operative criterion may lead to an increase in fatality [8].
Several recent clinical studies have shown favourable results from surgery without the routine use of pre-operative DSA [18, 19]. David et al [21] reported more than 400 patients predominantly operated upon after ultrasound diagnosis of stenosis; morbidity and mortality were exceptionally low (0.5% and 0.25%, respectively), comparing favourably with the respective NASCET and ECST rates. In a series of 380 carotid disobliterations reported by Melissano [20] the angiography rate was only 14%. This series had an excellent low rate of mortality (0.25%) and neurological morbidity (1.6%), the authors concluding that the selective use of arteriography and the intensive care unit enabled dramatic cost reduction for carotid surgery without sacrificing quality [20].
In conclusion, CFD is an accurate and reproducible test providing not merely morphological but also functional information. DSA is unnecessary for the detection and quantification of stenosis if CFD is unequivocal, which applies to the vast majority of cases. When CFD is inconclusive, this becomes evident during the examination and DSA may then be considered. Avoiding unnecessary carotid angiography before CEA not only offers great financial savings but also minimizes the risk of complications.
Received for publication January 17, 2001. Revision received April 12, 2001. Accepted for publication April 19, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Sabeti, M. Schillinger, W. Mlekusch, A. Willfort, M. Haumer, T. Nachtmann, M. Mullner, W. Lang, R. Ahmadi, and E. Minar Quantification of Internal Carotid Artery Stenosis with Duplex US: Comparative Analysis of Different Flow Velocity Criteria Radiology, August 1, 2004; 232(2): 431 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Kriegshauser, M. D. Patel, and K. D. Nelson Carotid Pseudostring Sign From Vasa Vasorum Collaterals J. Ultrasound Med., September 1, 2003; 22(9): 959 - 963. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
