Helical CT angiography in the preoperative evaluation of carotid artery stenosis

Transcription

1 Helical CT angiography in the preoperative evaluation of carotid artery stenosis Marianne Cinat, MD, Christopher T. Lane, MD, Hanh Pham, MD, Andrew Lee, MD, Samuel Eric Wilson, MD, and Ian Gordon, MD, Orange, Calif. Purpose: To determine the utility and accuracy of helical CT angiography (CTA) in the evaluation of carotid artery stenosis. Methods: A comparison of CTA and conventional arteriogram was performed in 53 patients undergoing evaluation for carotid artery stenosis. Ninety-six carotid systems were evaluable. CTA stenosis was determined by the percent of area reduction seen on axial images through the level of greatest narrowing. MIP images were used to identify the point of maximal stenosis and to visualize overall vascular anatomy. The percent diameter stenosis was measured on conventional arteriograms using strict North American Symptomatic Carotid Endarterectomy Trial (NASCET) and European Carotid Surgery Trial (ECST) criteria. Results: Significant correlation was found between CTA and arteriography (NASCET method R = 0.87, ECST method R = 0.87, p < 0.001). Using NASCET >60% as an indicator for disease, CTA had a sensitivity of 87%, specificity of 90%, accuracy of 89%, negative predictive value of 88%, and positive predictive value of 89%. CTA identified plaque characteristics such as ulcerations (8), occlusion (10), fatty plaques (22), calcifications (48), and fibrosis (2). CTA underestimated 2 cases of short segment stenoses because of volume averaging, but this discrepancy was detected by duplex scan. No complications or renal dysfunction occurred with CTA; 1 patient became symptomatic during arteriography, necessitating termination of the procedure. Conclusion: CTA is a safe, non-invasive technique that precisely measures carotid artery area reduction and highly correlates to conventional arteriography. With this new technology, the current standards for carotid artery imaging may need to be reevaluated, and the precise role for helical CTA more clearly defined. (J Vasc Surg 1998;28: ) The guidelines established by the recent cooperative studies on carotid endarterectomy in the treatment of patients with severe carotid artery stenosis depend on accurate determination of the degree of stenosis. 1 4 This has prompted a renewed interest in From the Departments of Surgery, and the Department of Radiology (Drs Pham and Lee), The University of California Irvine Medical Center and the Long Beach Veterans Administration Medical Center. Presented at the Twelfth Annual Meeting of the Western Vascular Society, Lana i, Hawaii, Sep 27 Oct 1, Reprint requests: Marianne Cinat, MD, University of California Irvine Medical Center, Department of Surgery, 101 The City Drive, Bldg. 53, Rte. 81, Orange, CA Copyright 1998 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter /98/$ /6/91147 the best method of imaging the carotid bifurcation. Traditionally, catheter arteriography has been considered the standard for measuring carotid artery stenosis. However, significant discrepancies in the angiographic criteria used in the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and the European Carotid Surgery Trial (ECST) has resulted in continued controversy about the most accurate method of measuring carotid artery stenosis The difficulty with both the NASCET and ECST methods lies in establishing the true diameter of the normal internal carotid artery distal to the stenosis (NASCET), or the true diameter of the vessel at the point of maximal stenosis (ECST). Both methods are subject to error, because measurements are extrapolated based on an estimation of what is thought to be normal anatomy. 290

2 Volume 28, Number 2 Cinat et al 291 Fig. 1 Severe stenosis of the left internal carotid artery shown by standard arteriography A, and an MIP image on CTA, B. In addition, because conventional arteriography provides only static planar views, eccentricities of the vascular lumen may not be identified. Helical CT angiography (CTA) overcomes these drawbacks. Helical CTA is a new, minimally invasive technique that allows the rapid acquisition of data that can be reconstructed and displayed in two- or three-dimensional images. 11,12 Shadedsurface display (SSD) or maximal intensity projection (MIP) reconstructions can be used to create images similar to conventional arteriograms. These images can then be rotated 360 to more accurately determine the point of maximal stenosis. Axial images can then be magnified and examined from original scan data to give a cross-sectional view of the arterial lumen and plaque morphology. Thus, the extent of disease along the length of the vessel can be clearly identified. The true cross-sectional area of the normal vessel lumen and the true area of the residual lumen can be definitively visualized and measured. Early studies using CTA in carotid imaging produced varied results Castillo et al. 13 showed only a 50% agreement between CTA and arteriography. As the imaging techniques and protocols improved, however, so did the results of CTA. Recent studies have shown agreement between CTA and arteriography in the range of 80% to 90% These studies, however, involve only a small number of patients, and their measurement of carotid artery stenosis relies on SSD and MIP images. Because MIP projections produce images similar to conventional arteriograms, measurements of stenosis are subject to the same difficulties. Only recently has the utility and improved precision of axial images in carotid artery imaging been recognized In addition, SSD images have been shown to be less accurate than MIP images in measuring carotid artery stenosis. 21,26 This study was undertaken to determine the utility and accuracy of CTA with both axial and MIP projections. Comparisons were made with conventional arteriography using both NASCET and ECST criteria. In addition, information regarding plaque morphology was obtained. Our aim was to clarify the role of CTA in the preoperative evaluation of carotid artery stenosis.

3 292 Cinat et al August 1998 Fig. 2. Percentage of area reduction was calculated by electronically outlining the residual lumen and the normal vessel lumen at the level of greatest stenosis. In this example, the area of the residual lumen is 11 mm 2, and the area of the normal lumen is 71 mm 2, resulting in an 86% area reduction. Calcifications are also seen within the plaque. Fig. 3. The expected curvilinear relationship between diameter reduction (as measured by arteriography) and area reduction (as measured by CTA). PATIENTS AND METHODS Fifty-three patients undergoing both helical CTA and catheter arteriography for the evaluation of carotid artery stenosis between January 1996 and June 1997 were studied. There were 52 men and 1 woman; the mean age of the patients was 68.6 years. Diagnostic studies were ordered at the discretion of the attending vascular surgeon or primary care physician. Helical CT angiography was performed using a Picker PQ-5000 scanner. Patients were asked to remain still, without swallowing, throughout the scan process. Scout images were obtained, and the table was positioned so that scanning would begin at the pedicles of the sixth cervical vertebra. Iodinated contrast material (ISO-370) was injected via a peripheral vein at a rate of 2.5 to 3 ml/s (the total volume was approximately 125 ml). After a 12 second delay, spiral scans were performed using a pitch of 1.25 mm/s and a 2 mm beam collimation. An index of 2 mm was used for both the initial scan and for the reconstructed images. The entire scan time was 30 to 40 seconds, and approximately 80 axial images were obtained for each artery. Data were then transferred to a work station, where bone and venous structures were electronically removed from axial images. Maximum intensity projection technique was used to reconstruct axial images into an arteriogram that could be projected from any angle (Fig. 1). CTA stenosis was determined based on the percentage of area reduction seen on axial images through the level of greatest narrowing. Area reduction was calculated by electronically outlining the residual lumen and the normal arterial lumen at the level of greatest stenosis (Fig. 2). The true percent area reduction could then be determined. In addition, CTA images were used to identify plaque characteristics such as ulceration, calcification, and fat content. Pixel density in Hounsfield units was used to distinguish fat from fibrosis and calcification from contrast. Any additional vessel abnormalities were noted. Catheter arteriograms were performed using a standard femoral approach. Digital subtraction images were obtained from multiple projections (anteroposterior, lateral, and oblique). The typical contrast dose (ISO-370) was 150 to 200 ml. Images were read using strict NASCET and ECST criteria 1,3 by physicians who were unaware of the results of CTA. In 10 patients, only 1 carotid vessel was included in the study because the contralateral vessel was known to be occluded on previous studies, or the carotid artery was read as normal, with no absolute value assigned. Overall, 96 carotid systems were able to be evaluated by both CTA and arteriogram. Reports were reviewed and recorded. Spearman correlation was used for all comparisons. In addition, the sensitivity, specificity, accuracy, negative predictive value, and positive predictive value were calculated for CTA, using conventional arteriogram as a standard.

4 Volume 28, Number 2 Cinat et al 293 Fig. 4. XY plots and correlation of CTA to arteriogram using NASCET criteria (R = 0.87, p < 0.001) A, and CTA to arteriogram using ECST criteria (R = 0.87, p < 0.001), B RESULTS With CTA, area reduction, as opposed to diameter reduction, is used to measure carotid artery stenosis. Fig. 3 illustrates the expected curvilinear relationship between diameter reduction (as measured by arteriography) and area reduction (as measured by CTA). Fig. 4 graphs the association of CTA to arteriogram (NASCET and ECST). Correlation was significant for each: NASCET R = 0.87, ECST R = 0.87(p < ). The sensitivity, specificity, accuracy, negative predictive value, and positive predictive value were calculated for CTA using arteriogram as a standard (Table 1). A CTA area reduction of 80% was considered a significant stenosis. This parameter was compared with a NASCET criteria stenosis of greater than 60%, based on the recent Asymptomatic Carotid Artery Stenosis trial, 4 and an ECST criteria stenosis of greater than 70%. 3 Plaque morphology and characteristics were also evaluated. Eight ulcerations were identified with CTA: 7 corresponded to arteriogram, 6 were in cases in which no arteriogram was performed, and 1 was identified by means of CTA but not noted on arteriogram. Nine ulcerations seen on arteriogram were not noted on CTA. Twenty-two patients were identified as having soft (fatty) plaques by means of CTA (Fig. 5). Calcifications were common, occurring in Table I. Parameters of accuracy for CTA in identifying carotid artery stenosis measured on arteriogram NASCET > 60% ECST > 70% CTA > 80 CTA > 80 Sensitivity 87% 79% Specificity 90% 88% Accuracy 89% 83% Negative predictive value 88% 81% Positive predictive value 89% 86% 48 of the 53 patients undergoing CTA; calcifications were only identified in 6 arteriograms (Figs. 2 and 5). Total occlusion was correctly identified by means of CTA in 10 of 11 cases. In one patient, a severe (95%) stenosis of the internal carotid artery was reported with CTA; arteriography performed several days later showed complete occlusion. Other vascular characteristics were also examined. An occluded common carotid artery with distal refilling by collaterals and a carotid artery aneurysm were correctly identified by means of CTA. However, a subclavian steal in 1 patient that was noted by means of arteriography was not identified with CTA. Cerebral cross-filling in 6 patients and intracranial aneurysm in 1 patient was identified by means of arteriography but not with CTA. In

5 294 Cinat et al August patient after injection for arteriography. Dissection was suggested with follow-up magnetic resonance angiography (MRA). No episodes of renal dysfunction were noted because of CTA or arteriography. Fig. 5. MIP images show severe narrowing of the right internal carotid artery, A. Calcifications are seen within the plaque. B, Magnified axial images with measurement of pixel density illustrate a fatty plaque (arrow) within the internal carotid artery (b). The internal jugular vein (a) and the external carotid artery are also shown (c). addition, 1 vessel was stenosed after endarterectomy. Arteriography showed marked narrowing, as did the MIP images by CTA; intimal hyperplasia and vessel wall thickening was confirmed by means of axial images on CTA (Fig. 6). No complications occurred because of CTA. Acute but transient neurologic deficits developed in DISCUSSION The accuracy and precision of CTA in the evaluation of carotid artery stenosis was confirmed. Unlike previous studies, our technique fully integrates both MIP and axial images for carotid evaluation. MIP reconstructions are used to identify the point of greatest narrowing and to visualize overall vascular anatomy. Axial images provide a cross-sectional view of the carotid vessels and atherosclerotic plaque. Eccentricities of the vascular lumen can also be identified. Thus, the percentage of area reduction can be precisely measured on magnified axial images by electronically outlining the normal vessel lumen and the residual lumen at the level of greatest narrowing at any point along the entire vessel length. No extrapolation or estimation is necessary, as is required by both NASCET and ECST methods. This eliminates the need to estimate the normal vessel anatomy, as is required by both NASCET and ECST methods. Although clinicians are accustomed to vascular data being expressed as the percentage of diameter reduction of a vascular lumen, the hemodynamic significance of a vascular stenosis is directly proportional to the percentage of area reduction of the lumen. Laboratory studies have shown that an area reduction of 80% results in a significant drop of pressure across the stenosis. 27 This effect may be more pronounced in low resistance systems, such as the carotid circulation. Thus, in this study, a CTA area reduction of 80% was considered significant. This parameter was then evaluated for its precision in identifying carotid artery stenosis with arteriogram. Using a NASCET criteria stenosis of 60% as an indicator for disease, CTA was found to have a sensitivity rate of 87%, specificity rate of 90%, accuracy rate of 89%, negative predictive value rate of 88%, and a positive predictive value rate of 89%. The corresponding values for ECST methods were slightly lower (Table 1). One explanation for this may be that when using ECST criteria there is greater difficulty in estimating the true diameter of the carotid bulb, resulting in more variability and a higher potential for error. 28,29 Similar difficulties occur in identifying the normal distal internal carotid artery when using NASCET criteria. In both instances, some degree of approximation is necessary. Thus, CTA measurements may be more accurate than arteriogram, because axial images allow a

6 Volume 28, Number 2 Cinat et al 295 Fig. 6. A, MIP images show severe narrowing of the right internal carotid artery. B and C, Axial images confirm intimal hyperplasia and fibrosis after a carotid endarterectomy. The residual lumen measures 3 mm 2, and the normal vessel measures 27 mm 2, resulting in an area reduction of 89%. definitive measure of the normal vessel lumen and the residual lumen. Plaque morphology may also play an important role in determining the risk for cerebral vascular accidents. 30,31 CTA has several advantages over conventional arteriography in identifying plaque characteristics. Axial imaging by CTA allows a cross-sectional evaluation of the carotid plaque. Differentiation of fat, fibrosis, calcium, and contrast using pixel density permits visualization of plaque composition, intimal hyperplasia, intraplaque hemorrhage, and dissection. 32,33 Ulcerations were seen best on MIP images and were identified in 8 patients by means of CTA. However, 9 ulcerations reported by means of arteriogram were not identified by means of CTA. To clarify this discrepancy, further correlation with surgical findings and pathologic specimens in a prospective study is necessary. Ten of 11 total occlusions of the internal carotid artery were identified by means of CTA. CTA showed severe stenosis in 1 patient, but arteriogram performed several days later showed total occlusion. Because this patient did not undergo surgical exploration, we do not know whether CTA or arteriogram misread the occlusion, or whether the vessel completely occluded during the interval between tests. We also find it encouraging that several unusual vascular abnormalities, including a carotid artery aneurysm (1), an anastomotic stenosis (1), and extracranial tandem lesions (2), were identified by means of CTA. An occluded common carotid artery with reconstitution of the internal carotid artery distally via external carotid collaterals was correctly detected by means of CTA. Thus, it appears that we are able to accurately identify unusual vascular abnormalities by means of CTA. Close examination of the false negative and false positive results reported by means of CTA illustrates some deficiencies in CTA technique. Perhaps of most concern was that 2 cases of short segment

7 296 Cinat et al August 1998 stenoses because of volume averaging and another because of motion artifact were underestimated with CTA. These lesions, however, were all identified by means of duplex scan. The discrepancy prompted further evaluation, and duplex findings were confirmed by means of arteriogram. In addition, lesions within the carotid bulb tended to be overestimated with CTA. This was especially true when using NASCET criteria, which notoriously underestimate carotid bulb lesions. The accuracy of measurement of stenosis by axial images depends on the scan plane, which has to be perpendicular to the carotid artery and residual lumen. For areas where the vessel may be angled, such as the carotid bulb, the cross-sectional area may be falsely elongated. In these instances, MIP images can be used to estimate diameter stenosis. Disadvantages to CTA as compared with arteriogram include inability to detect intracranial vascular pathology (intracranial aneurysms) and flow dynamics (cross-filling). The significance of this information, however, is controversial because these characteristics rarely change therapeutic plans. Better visualization of the intracranial circulation may be obtained if the scan technique is modified by extending the scan distance, the scan time, and the amount of contrast injected. Although we did not use them, protocols are currently available that allow CTA to image the intracranial circulation and the circle of Willis. 34,35 In addition, CTA combined with other imaging modalities, such as transcranial Doppler and MRA, can provide substantial information regarding flow dynamics of the intracranial circulation. Other disadvantages include the excessive artifact that is created by metallic objects, such as vascular clips, which makes accurate imaging of the carotid vessels impossible. Data reconstruction for MIP images requires approximately 20 to 30 minutes at a workstation for removal of bone and venous structures. However, if this is added to the short scan time (30 to 40 seconds), the overall time for CTA is less than the time required for conventional arteriogram. Finally, CTA quality depends on technique. Poor results in early studies using CTA were likely caused by errors in technique. 13 We found our protocol provided quality images in the most time-effective manner. Having the patient refrain from swallowing during the examination was also important to image quality. Swallowing frequently created motion artifact, which appears as waviness in the walls of the vessel on MIP images. When screening duplex scan results are equivo- cal, a test is required before carotid endarterectomy for confirmation. Both CTA and MRA have an advantage over conventional arteriography: an arterial puncture, and the associated risk of stroke and carotid dissection, can be avoided. CTA also has several additional advantages over MRA. Perhaps most important, severely stenotic lesions with low flow that may be missed and read as total occlusion with both MRA and duplex can be detected with CTA. Also, CTA requires a scan time of only 3 minutes, which is much less than that required for MRA; thus, claustrophobic patients can tolerate CTA without difficulty. There is no contraindication to CTA for patients with pacemakers or implanted devices. However, CTA does require the use of intravenous contrast; therefore, in patients with renal insufficiency, MRA should be considered for confirmation before CTA. In conclusion, helical CTA appears to be a promising new technique for carotid artery imaging that overcomes several of the drawbacks of conventional arteriography. An arterial puncture can be avoided, and axial images allow direct visualization of the arterial lumen and plaque morphology. A precise measurement of carotid artery area reduction can then be calculated at any point along the length of the vessel. Sensitivity and specificity approach 90%. Thus, when equivocal results are obtained by means of duplex scan, helical CTA can be used as confirmation, before conventional arteriography. With this new technology, the current standards for carotid artery imaging may need to be reevaluated, and the precise role for helical CTA more clearly defined. REFERENCES 1. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991;325: Mayberg MR, Wilson SE, Yatsu F, Weiss DG, Messina L, Hershey LA, et al. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. JAMA 1991;266: European Carotid Surgery Trialists Collaborative Group. MRC European Carotid Surgery Trial: Interim results for symptomatic patients with severe (70 99%) or with mild (0 29%) carotid stenosis. Lancet 1991;337: Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995;273: Fox AJ. How to measure carotid stenosis. Radiology 1993;186: Hobson RW, Strandness DE. Carotid artery stenosis: What s in the measurement? J Vasc Surg 1993;18: Bladin CF, Alexandrov AV, Murphy J, Maggisano R,

8 Volume 28, Number 2 Cinat et al 297 Norris JW. Carotid stenosis index: A new method for measuring internal carotid artery stenosis. Stroke 1995; 26: Rothwell PM, Gibson RJ, Slattery J, Sellar RJ, Warlow CP. Equivalence of measurements of carotid stenosis: a comparison of three methods on 1001 angiograms. Stroke 1994;25: Barnett HJM, Warlow CP. Carotid endarterectomy and the measurement of stenosis. Stroke 1993;24: Eliasziw M, Smith RF, Singh N, Holdsworth DW, Fox AJ, Barnett HJM. Further comments on the measurement of carotid stenosis from angiograms. Stroke 1994;25: Schwartz RB. Neuroradiological applications of spiral CT. Seminars in Ultrasound, CT, and MRI 1994:15: Dillon EH, van Leeuwen MS, Fernandez MA, Mali WPTM. Spiral CT Angiography. AJR Am J Roentgenol 1993:160: Castillo M. Diagnosis of disease of the common carotid artery bifurcation: CT angiography vs. catheter angiography. AJR Am J Roentgenol 1993: Castillo M, Wilson JD. CT angiography of the common carotid artery bifurcation: Comparison between two techniques and conventional angiography. Neuroradiology 1994;36: Schwartz RB, Jones KM, Chernoff DM, Mukherji SK, Khorasani R, Tice HM, et al. Common carotid artery bifurcation: Evaluation with spiral CT. Radiology 1992;185: Dillon EH, van Leeuwen MS, Fernandez MA, Eikelboom BC, Mali WPTM. CT angiography: Application to the evaluation of carotid artery stenosis. Radiology 1993; 189: Marks MP, Napel S, Jordan JE, Enzmann DR. Diagnosis of carotid artery disease: Preliminary experience with maximum intensity projection spiral CT angiography. AJR Am J Roentgenol 1993;160: Cumming MJ, Morrow IM. Carotid artery stenosis: a prospective comparison of CT angiography and conventional angiography. AJR Am J Roentgenol 1994;163: Link J, Brossmann J, Grabener M, Mueller-Huelsbeck S, Steffens JC, Brinkmann G, et al. Spiral CT angiography and selective digital subtraction angiography of internal carotid artery stenosis. Am J Neuroradiol 1996;17: Link J, Brossman J, Penselin V, Gluer CC, Heller M. Common carotid artery bifurcation: Preliminary results of CT angiography and color-coded duplex sonography compared with digital subtraction angiography. AJR Am J Roentgenol 1997;168: Leclerc X, Godefroy O, Pruvo JP, Leys D. Computed tomographic angiography for the evaluation of carotid artery stenosis. Stroke 1995;26: Papp Z, Patel M, Ashtari M, Takahashi M, Goldstein J, Maguire W, et al. Carotid artery stenosis: Optimization of CT angiography with a combination of shaded surface display and source images. Am J Neuroradiol 1997;18: Dix JE, Evans AJ, Kallmes DF, Sobel AH, Phillips CD. Accuracy and precision of CT angiography in a model of carotid artery bifurcation stenosis. Am J Neuroradiol 1997;18: Cinat ME, Lane CT, Pham H, Lee A, Wilson SE, Gordon I. Spiral CT angiography definitively measures carotid artery stenosis. Presented at the Twenty-third World Congress of the International Society for Cardiovascular Surgery, 1997, London, England. 25. Calzolari F. Imaging carotid artery stenosis: The role of CT angiography [letter]. Am J Neuroradiol 1997;18: Takahashi M, Ashtari M, Papp Z, Patel M, Goldstein J, Maguire WM, et al. CT angiography of the carotid bifurcation: Artifacts and pitfalls in shaded-surface display. AJR Am J Roentgenol 1997;168: Sumner DS. Essential Hemodynamic Principles. In: Rutherford RB, editor. Vascular Surgery. 4th edition. Philadelphia: WB Saunders Company; p Alexandrov AV, Bladin CF, Maggisano R, Norris JW. Measuring carotid stenosis: Time for a reappraisal. Stroke 1993;24: Rothwell PM, Gibson RJ, Slattery J, Warlow CP. Prognostic value and reproducibility of measurements of carotid stenosis: A comparison of three methods on 1001 angiograms. Stroke 1994;25: Bornstein NM, Norris JW. The unstable carotid plaque. Stroke 1989;20: Markus HS, Droste DW, Brown MM. Detection of asymptomatic cerebral embolic signals with Doppler ultrasound. Lancet 1994;343: Leclerc X, Godefroy O, Salhi A, Lucas C, Leys D, Pruvo JP. Helical CT for the diagnosis of extracranial internal carotid artery dissection. Stroke 1996;27: Hopper KD, Gouldy CA, Kasales CJ, TenHave TR, Fischer AL. The effect of helical CT on X-ray attenuation. J Comput Assisted Tomogr 1997;21(1): Napel S, Marks MP, Rubin GD, et al. CT angiography with spiral CT and maximum intensity projection. Radiology 1992;185: Gorzer H, Heimberger K, Schindler E. Spiral CT angiography with digital subtraction of extra- and intracranial vessels. J Comput Assisted Tomogr 1994;18: Submitted Oct 8, 1997; accepted Apr 17, DISCUSSION Dr. Linda M. Reilly (San Francisco, Calif.). The authors have presented a study that represents a further attempt to identify the optimal method for preoperative imaging of extracranial cerebrovascular disease. I have several questions for the authors that pertain to study design, methodology, and the overall significance of the results. You report on 102 patients who were evaluated for carotid disease in an 18-month interval. Was this a consecutive series? If not, how many patients were eliminated and what were the reasons? Because it appears that no protocol was used to select the method of preoperative imaging, is it possible that there was some inadvertent bias in the selection of the patients who underwent conventional angiogram? Only 56 of the 102 patients had conventional angiograms. How many patients actually underwent all three studies duplex scan, computed tomography angiography (CTA), and conventional angiogram? Should that group be your study group?

9 298 Cinat et al August 1998 Throughout the manuscript, you compare the percent area reduction as measured by CTA with the percent diameter reduction as measured by conventional angiography or duplex ultrasound scan. This is a potential source of confusion for the reader because a 250% diameter reduction corresponds to a 75% area reduction. You acknowledge this curvilinear relationship in the manuscript, but it is unclear whether the data displayed in the tables allows for this difference. In other words, when I look at your data, should I be correlating your 70% to 99% area stenosis group to the 30% to 69% diameter stenosis group, or not? There is a simple way around this problem. You should just back-calculate from percent area reduction to percent diameter reduction. For clarity of data correlation in our trial that compared MRA, duplex scan, and angiography, in which we also measured area reduction with the magnetic resonance angiography (MRA), we back-calculated what we called the effective diameter at the point of maximal stenosis and at the point of no disease. Then, we used those to calculate the percent diameter reduction for the MRA. You can use this same method for your CTA data, and this will eliminate this point of confusion. You have stated that axial imaging by CTA allows a unique evaluation of the carotid plaque not seen by any other technique. This is not entirely true because duplex carotid imaging has been used to characterize plaque structure for at least 15 years. Did any of these patients undergo endarterectomy? If so, were the specimens used to validate the imaging results, particularly the plaque characterization? If so, what were the results of the correlation with the specimens? How were the specimens handled? You have reported that CTA correlates well with conventional angiography and less well with ultrasound scan. Is this really good news? In our study that compared MRA, duplex scan, and angiography with the resected specimen as the gold standard, duplex scan and MRA correlated well with each other and with the harvested specimen. However, angiography correlated poorly with duplex scan, with MRA, and with the harvested specimen. Similar results were obtained in two other studies by Fontenelle and Weintraub, with correlation coefficients of around 0.8 for duplex scan and the specimen, and MRA and the specimen, but only 0.57 for angiography and the specimen. I am concerned that you have demonstrated good correlation with the least accurate imaging method. Could you comment please? Although the submitted abstract includes MRA imaging data, the manuscript excludes it. Why did you not include results of imaging with MRA in your comparative analysis? It is, in fact, your strongest competitor. I do not have any quarrel with your conclusion that CTA is an acceptable method for imaging the carotid artery. I do question the clinical significance of this observation. Your paper is entitled Helical CTA in the preoperative evaluation of carotid artery stenosis, but, in fact, arteriography has already been replaced by either duplex scan alone or duplex scan plus MRA, the two most common imaging paradigms currently in use. That brings me to my last, and most basic, question. Why would I, as a clinician, use CTA to assess a patient before surgery as opposed to using duplex ultrasound scan alone or duplex scan plus MRA? CTA clearly is more expensive than duplex imaging and has the additional disadvantage of contrast administration. CTA may be equivalent in cost to MRA, but again has the disadvantage of contrast administration. So, if I am accustomed to performing carotid thromboendarterectomy on the basis of duplex ultrasound scan with or without confirmatory MRA, why would I substitute or add CTA? I enjoyed reading your manuscript. I thank the Society for inviting me to discuss this work. Dr. Marianne Cinat. Thank you, Dr. Reilly. I will start from the top and try to work my way down through your questions. The first question concerns methodology. This was not a prospective study, and we did not intend it to be a prospective study. We began using CTA about a year and a half ago. The initial images were not good. We slowly changed altered the protocol because we were impressed with the axial images and how we could see plaque morphology and various changes in the vessels. We slowly developed a larger series and eventually moved to just doing a screening duplex scan followed by a CTA. This is why there is a discrepancy in the numbers and why some patients did not have arteriography. I thought about comparing only the patients who had all three studies. However, because I also wanted to compare helical CTA with duplex scan and I felt those were the two tests we may use clinically in our evaluation of the patients, it gave me larger numbers for that comparison. As far as the conversion from area reduction to diameter reduction, I think, if anything, it should be the other way. One problem with a standard arteriogram with respect to diameter reduction is that it is a static planar view. If you have any obliqueness or eccentricity in the lumen and you do not hit that directly, you will have an inaccurate reading by standard arteriography. By measuring area stenosis and outlining the residual lumen and the vessel, you have a more precise degree of hemodynamic significance of the lesion. Now, obviously hemodynamic significance is not the only important part of cerebrovascular disease. There is also embolic phenomena, but as we learn more about plaque morphology and its relationship to embolic phenomena, CTA may provide a means for doing this. You also stated that duplex scan does give us information regarding plaque morphology. I think that the information provided by duplex scan is much more imprecise than what we can obtain with the axial imaging. Thick calcified plaques sometimes obscure our vision with duplex scan, and the axial imaging will obviate that difficulty. You brought up a good point, which is actually what I would like to embark on next, and that was about correlating the results of CTA to surgical specimens. Again, this was a retrospective review, and as we improve our protocol, my hope is that over the next several months we can perform a prospective study preserving those specimens

10 Volume 28, Number 2 Cinat et al 299 and actually looking at the percent area reduction in the surgical specimens after they are appropriately prepared. We dropped MRA from our analysis because it has not proved to be useful at our institution. We have about 20 patients in the series that did have MRA, and the data were scattered. Some patients had an MRA months or years before the CTA and duplex scan results, and I felt that it detracted from the study data. Also, in the abstract you saw the correlation was poor, a To conclude, I want to address your question on why would a clinician use CTA. If you would feel comfortable operating on a patient with only a duplex scan, then that is appropriate. CTA is not meant to replace that strategy mainly because there is a contrast load. I think that if you can do a study without putting a patient at the risks associated with contrast, you should. Still, I think there are patients in whom you need confirmation of the duplex study or for whom you want a hard copy of the data. In those patients, I think CTA may be better than MRA. It is a more rapid test, done in less than 40 seconds. So, a patient who is claustrophic or has trouble being in an MRA machine for 30 or 40 minutes, will not have to deal with that problem. MRA also tends to have an increased false positive rate. It can indicate a greater degree of stenosis than arteriography or duplex scan, and I think that the CTA gives more information regarding plaque morphology. Also, low-flow conditions with very tight lesions may be missed on MRA. Advantages of CTA over duplex scan include obtaining a hard copy of the data. Again, we do not have to rely on a technologist in the lab. We can bring these arteriograms/angiograms to the operating room. I think that as we investigate plaque morphology more closely over the next several years, CTA may be the best mechanism to define whether plaque is calcified, fatty, or fibrotic. CTA may help us choose which patients could undergo endovascular intervention or should not because of a risk of embolization. Dr. Eugene Strandness (Seattle, Wash.). I agree that helical CT will work in this area. There is no question about that, and I would not argue with that for a minute. The problem with this field right now is it is in turmoil. The Veteran Affairs trial said you could operate on a 50% diameter stenosis, ACAS was 60%, NASCET was 70%, and the European carotid trial was 70%. However, it is not the NASCET 70%. The issue boils down to what we are really discussing. Dr. Reilly mentioned the point about cross-sectional area. I think that as easy as it sounds, most people will not pay any attention to it. They always want to have diameter reduction. In fact, Dr. Barnett states that you have to have a 70% diameter reduction and not a 69%. This is how bad this whole field has gotten. In addition to that, now the Europeans and the Americans are arguing about the proper site to make your measurements. The Europeans have suggested that the common carotid artery should be the index artery and not the internal carotid artery. The North Americans have rejected that. So, vascular surgeons are confronted with all of these dilemmas. I enjoyed your presentation. I do not disagree with anything you said. I just do not think we know at the moment exactly where we are going. I do not think that it makes a difference whether you use cross-sectional area or diameter reduction. You will still have to go back and look at the patient and estimate the degree of diameter reduction. I would appreciate your comments in that area. Dr. Cinat. I agree. The whole issue is in turmoil. There are countless papers trying to determine the best way to image the carotid artery and the best way to measure diameter reduction. If people are arguing about something for years and years, that probably means nobody is right. Right now, this is a new way to image the vessel. You can see a clear cross-sectional area. Where it will fall exactly in the future, I do not know. I know that our technology in using it for a year has advanced. I think it gives us a unique perspective of the vessel. It helps us to look at things that we have not been able to see before, and it may continually play a more significant role. Obviously if we make a change to percent area reduction, which again I feel is most accurate, I think that we would have a precise measurement rather than ranges of measurements, which are frequently given with duplex scan. Then, there will not be any room for technical error or variations amongst labs. It will provide a hard copy. It will be a precise measurement, and perhaps we need to make a change to percent area reduction. I still feel that that is a very valid contribution of CTA. Dr. Gregory Moneta (Portland, Ore.). That was a nice presentation. I have just a couple questions and a comment. One problem that people have with the concept of using duplex scan as the only imaging method before operating on the carotid artery is whether you can image other sites of interest with your CT imaging. For example, could you look at the aortic arch or the intercranial vessels? Secondly, the concept of confirming a duplex study with MRA does not make sense. I think it is wrong to use one test, which in most cases is not good, to confirm another test, which is quite excellent. We ought to move away from this. In most of our institutions, MRA is not useful for confirming high-grade stenosis. If you have to confirm the results of the duplex examination, you ought to do a conventional angiogram. Dr. Cinat. Thank you. The first question had to do with whether we imaged other anatomical areas. No, we did not, but we are able to do this. The protocol for CTA can be adjusted to image any area. We have been using CTA for imaging abdominal aortas and thoracic aortas. All it requires is proper timing so that the contrast bolus will be maximal in the area that you want to scan. It would require a slightly longer time to visualize the aortic arch and slightly longer to image the intercranial vessels. However, there are several papers that have used it to image intracranial circulation, and they give good

11 300 Cinat et al August 1998 images of the intercranial circulation. So, it is just a matter of working out the details of the protocol. The second question is whether or not we need to use a confirmatory test. I think that that is really up to the individual surgeon. Duplex scan is not perfect. I think that CTA offers advantages over arteriography. You do have to use contrast, but again, it is quicker and less invasive, and it gives much more information than is obtained on a standard arteriogram. I think that with a duplex scan and a CTA you will not miss any lesions. If you feel that a confirmatory test is necessary, I think CTA will certainly play a significant role in the future. THE WILLIAM J. VON LIEBIG FOUNDATION AWARD FOR VASCULAR SURGERY RESEARCH, th Annual William J. von Liebig Foundation Award for Residents and Fellows $5000 Award Eligibility requirements: Research performed by a Resident or Fellow on staff at an institution in the United States, Canada, or Mexico with senior collaborators acting in a consultative capacity. Manuscript accompanied by a signed letter from the author s superior attesting that the author performed all the essential parts of the experimental work reported. A full curriculum vitae must be attached. General requirements for the award: The research may be experimental or clinical in nature dealing with some fundamental or clinical aspect of vascular surgery. Clinical research papers are especially encouraged. Research performed by an individual on staff at an institution in the United States, Canada, or Mexico. Must be an original, unpublished work (not submitted elsewhere for publication, except to the ACS Surgical Forum). Submitted in English (10 copies of the typed manuscript and 10 copies of glossy prints of illustrations), complying with the Instructions to Authors of the Journal of Vascular Surgery and including an abstract of 250 words or less. A cover sheet indicating that the manuscript is to be considered for: The 18th Annual William J. von Liebig Foundation Award for Residents and Fellows. The manuscripts submitted will be reviewed by a select committee of vascular surgeons. The award will be presented at the annual meeting of the Southern Association for Vascular Surgery. The von Liebig Foundation reserves the right to withhold the grant of the award at the sole discretion of the Award Committee, whose judgement with respect thereto shall be final and conclusive. Further inquiries may be directed to the same address. Manuscripts must be postmarked no later than September 1, Jean A. Goggins, PhD, Award Committee Secretary, The William J. von Liebig Foundation, 281 Broad Ave. South, Naples, FL 34102;