Duplex US of the External Carotid Artery

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Duplex US of the External Carotid Artery M. J. Päivänsalo, T. M. J. Siniluoto, T. A. Tikkakoski, V. Myllylä & I. J. I. Suramo To cite this article: M. J. Päivänsalo, T. M. J. Siniluoto, T. A. Tikkakoski, V. Myllylä & I. J. I. Suramo (1996) Duplex US of the External Carotid Artery, Acta Radiologica, 37:1, To link to this article: Published online: 04 Jan Submit your article to this journal Article views: 2696 Full Terms & Conditions of access and use can be found at

2 Acta Radiologica 37 (1996), 4145 Printed in Denmark. All rights reserved Copyright 0 Acta Radiologica 1996 A C TA RADIO LO GI CA ISSN DUPLEX US OF THE EXTERNAL CAROTID ARTERY M. J. PAIVANSALO', T. M. J. SINILUOTO~, T. A. TIKKAKOSKI~, V. MYLLYLA~ and I. J. I. SURAMO' 'Department of Radiology, University Central Hospital, Oulu; and'department of Radiology, Keski-Pohjanmaa Central Hospital, Kokkola, Finland. Abstract The ratio between the systolic peak velocities of the internal and common carotid arteries (vpicnvpcca), vpica and grey-scale imaging measurement are generally used to evaluate internal carotid stenosis against known flow criteria in order to differentiate non-significant from significant stenosis. The same criteria are also used for evaluating the external carotid artery (ECA). Our data on 707 normal or stenotic ECA nevertheless showed that the systolic peak velocity of the normal ECA (vpeca) and its ratio to the systolic velocity of the CCA (vpecnvpcca) are higher than vpica and vpicnvpcca. vpecnvpcca is about 2 in >0-49% ECA stenosis. Only in severe stenosis are the peak velocities almost comparable. The ratio between the peak end diastolic velocities (edvecnedvcca) and edveca proved to be unreliable, as did grey-scale imaging measurement of the external carotid stenosis. In addition, ipsilateral internal carotid stenosis greatly affects the non-stenotic external carotid flow values, and probably has the same effect on the flow values of a stenotic external artery. Thus, external carotid flow values must be considered carefully. Key words: Carotid arteries, external artery; US; flow. Correspondence: Markku Paivansalo, Diagnostic Radiology, University of Oulu, Kajaanintie 50, FIN Oulu 22. Finland. FAX * Accepted for publication 24 February Carotid US is usually performed to evaluate plaque and stenosis in the carotid bifurcation, especially in the internal carotid artery (ICA). External carotid artery (ECA) stenosis is less frequent and clinically less important than ICA stenosis, but it may be the cause of carotid bruit if there is no stenosis in the ICA. The duplex US findings concerning external carotid stenosis have been handled in the same material with internal carotid stenoses (6, 8, 12, 16), and the same Doppler criteria have been used to evaluate them. However, the flow values seem to be different in the ECA, and we wanted to establish the velocity criteria. Material and Methods A total of 358 patients (234 men, 124 women, mean age 59 years, range years) were examined for neurological symptom(s) first with duplex sonography and then with carotid arteriography within a few days or weeks (maximum 3 months) between April 1988 and March About patients were examined with duplex sonography during this period. Among the 358 patients also examined by angiography, duplex sonography showed at least a 50% diameter stenosis in the ICA in 219 cases. The sonographic examinations were performed with a Toshiba SSA 100 unit with a linear-array transducer with 7.5-MHz frequency for grey-scale imaging and 3.5-MHz frequency for Doppler insonation. The equipment used range-gated pulsed Doppler, a steerable Doppler beam, and simultaneous display of grey-scale imaging and Doppler information, facilitating accurate placement of the sample volume and determination of the Doppler angle. The data gathered from the duplex examination consisted of peak systolic velocity in the common carotid artery (vpcca) and in the ECA (vpeca), the peak systolic velocity ratio vpeca/ vpcca, the ECA end diastolic velocity (edveca), 41

3 M. J. P-ANSALO ET AL. the diastolic velocity ratio edvecnedvcca, and measurement of the diameter stenosis of the ECA lumen from the grey-scale image. The flow was measured about 0.5 to 1 cm distally from the bifurcation, or at the point of highest velocity in cases of stenosis. The diameter of the stenosis was measured on longitudinal and transversal images in 2 projections by comparing the diameter of the residual lumen with the total lumen, and the mean diameter stenosis was calculated. The same data concerning the internal carotid arteries were collected simultaneously (19). Angiographies were performed with an intraarterial injection of contrast medium ( mg Uml). The angiograms were usually performed for visualisation of the ICA, but stenoses of the ECAs were also documented. The DSA technique was used in most cases (conventional aortocervical arteriography with film subtraction in a few cases). At least 3 angiographic projections were used for visualisation of the carotid bifurcations. Selective CCA injections were always performed if visualisation of the stenotic ICA was inadequate. The ECA stenoses were measured by comparing the diameter of the residual lumen with that of the assumed normal lumen. Diameter stenosis was measured in at least 2 projections and the mean was calculated. Each ECA artery was placed into one of the following stenosis categories: 0%, >0-29%, 30-49%, 50-69%, 70-99%, and occlusion (100%). Mean flow values and standard deviations for vpeca, vpecnvpcca, edveca, and edveca/ edvcca in the various angiographic stenosis groups were calculated retrospectively. Receiver operating characteristic (ROC) curves were generated for various values of vpeca, vpeca/vpcca, edveca, edvecnedvcca, and grey-scale imaging measurement of stenosis as discriminators for disease at 70%, 50%, 30%, and >O% stenosis levels. To determine which parameter was the best predictor of each category of stenosis, we calculated the area under each ROC curve and compared the results according to the method of HANLEY & MCNEIL (11) and MCNEIL & ADELSTEIN (17). Multiple regression analysis was performed to determine whether combining the parameters resulted in increased diagnostic accuracy. Since the velocity values increased approximately exponentially with increasing degrees of stenosis, the natural log transform of the flow values was included in the regresr sion. The series of 358 patients provided a total of 707 documented ECAs. Results vpeca and vpecnvpcca were more accurate than edveca, edvecnedvcca, and grey-scale imaging in detecting stenosis. Grey-scale imaging stenosis measurement was poorer for detecting moderate and severe stenosis than the flow parameters. The mean values ( SD) for the flow parameters in each ECA stenosis group are presented in Table 1, and the areas under the ROC curves for the flow parameters and the grey-scale imaging stenosis measurements are compared in Table 2. vpeca correlated with angiography better (r=0.60) than the other flow parameters (vpeca/ vpcca: 0.5 1, edveca: 0.39, edvecnedvcca: 0.3 l), and grey-scale imaging (0.54). The combination of grey-scale imaging and vpeca gave a Pearson s correlation coefficient of 0.65, and the same was obtained when adding vpecnvpcca in combination with grey-scale imaging measurement. The use of grey-scale imaging measurement and all the flow parameters increased the r value only to Ipsilateral ICA stenosis affected the non-stenotic ECA flow values; e.g., vpecnvpcca was about 1 in cases of non-stenotic ICA, but about 2.4 in cases of 100% ipsilateral ICA stenosis. vpeca, vpecn vpcca, and edveca changes in the non-stenotic ECAs in the presence of different grades of ipsilatera1 ICA stenosis are presented in Table 3. Angiography showed 22 occlusions of the ECA with additional CCA occlusion in 4 cases, and collateral filling of an occluded ECA in 8 cases. Table 1 Mean values of vpeca, vpeca/vpcca, edveca, and edveca/edvcca in duplex US examination of stenoses Stenosis, % ECA 0 A e vpeca (ksd) vpecanpcc A edveca (SD) edvecnedvcc A n m/s (SD) m/s (SD) (0.37) 1.2 (0.95) 0.13 (0.08) 0.72 (0.63) (0.48) 1.8 (1.0) 0.14 (0.13) 0.94 (1.0) (0.56) 1.9(1.3) 0.20 (0.16) 1.2 (0.76) (0.49) 2.3 (1.1) 0.24 (0.18) 1.3 (1.0) (0.95) 3.9 (1.8) 0.35 (0.27) 1.6 (1.5) 18 42

4 DUPLEX US OF THE EXTERNAL CAROTID ARTERY Table 2 Comparison of results of ROC curve analysis of vpeca, vpeca/vpcca, edveca, edveca/edvcca, and grey-scale imaging stenosis measurements using the area under the curve as a discriminator ECA: grade Area under ROC curve of stenosis vpeca vpeca/vpcca edveca edveca/edvcc A Grey-scale 70% % % >0% Duplex US showed an occlusion in 19 cases; severe stenosis was suspected in 2 cases and moderate stenosis in 1. The occlusion was probably missed on US because of collateral filling of the artery. US correctly showed collateral external filling in 5 cases, where the collaterals had a downward arterial flow close by the upward flow in the ECA. Discussion Previously reported values (4, 5, 8-10, 12, 16, 20-22, 24) and our own experiences (19) were used as rough guiding values for vpica, vpicnvpcca, vpeca, and vpecnvpcca in various grades of stenosis. The criteria vary depending on whether NASCET (ratio of the residual lumen to the normal distal ICA), ECST (ratio of the residual lumen to the estimated normal bulb diameter), or direct visual estimation of ICA stenosis (our method) is used (2, 14, 18). Although the flow values for different grades of ECA stenosis were not known beforehand, an attempt was made to grade each patient s ECA stenosis at the time of examination, in the knowledge from experience that ECA flow values (vpeca, vpecnvpcca) are higher in normal ves- Table 3 vpeca, vpeca/vpcca, and edveca of the non-stenotic external carotid artery in 485 cases of ipsilateral stenosis of the internal carotid artery Grade of ICA stenosis, %, and number of cases > Total 485 n vpeca, m/s vpeca/ vpcca edveca sels and mild stenoses than in ICA cases. Pearson s correlation coefficient between ECA stenosis at US and arteriographic stenosis was 0.79 (probability O.OOO), being poorer than the correlation (r=0.94) in our ICA work (19). The sensitivity of US with use of grey-scale imaging and flow values for detecting ECA diameter stenosis of 70% or greater was 88%, the specificity 98%, and the accuracy 97%. The sensitivity in our ICA work (19) was 96%. Our ICA flow results (19) corresponded to those earlier reported (8, 10, 16, 20, 21, 24), but the results in this study differed. vpeca was greater than vpica in normal vessels, and it was only in severe stenosis (>70%) that the peak velocities were nearly comparable. vpecnvpcca was already about 2 in 30% to 49% ECA stenosis and 2 in 50% to 69% ICA stenosis (19). In addition, ipsilateral internal carotid stenosis greatly affected nonstenotic external carotid flow values and may affect flow values in cases of stenosis. Contralateral ICA stenosis causes slight change in CCMCA flow (3, 13, 19,23), and perhaps also affects ECA flow. Similarly, a severe ECA stenosis or occlusion may have an effect on ICA flow, although severe ECA stenosis is rarer than ICA stenosis, and the effect is probably slighter due,to the smaller flow effect of a narrower ECA. Thus, there is a wide variation in flow values, resulting in a great overlap of values between the stenosis groups. Thus, carotid flow values must be considered carefully. vpeca proved most accurate in ROC curve analysis, but vpecnvpcca was nearly as good. The ICA ROC curves generally showed lower false-positive rates than the ECA curves. Grey-scale imaging stenosis evaluation was much more accurate in our ICA cases than in the present ECA cases. In our ICA work (19) the combination of grey-scale imaging measurement and systolic velocity ratios for correlation with angiography resulted in an r value of 0.90, and vpica increased it further to Correspondingly, the combination of vpeca, greyscale imaging stenosis measurement, and vpecn 43

5 M. J. PAIVANSALO ET AL. vpcca gave a Pearson s correlation coefficient between US and arteriographic ECA stenosis of edveca and edvecnedvcca proved less reliable in the ECAs than in the ICAs. Grey-scale imaging measurement of external carotid stenosis also proved less reliable than that of internal carotid stenosis (19). This may be due to the higher measuring error in smaller vessels, but it has been demonstrated that B-mode imaging can be highly accurate for assessing minimal disease (15,25) whereas limitations are encountered with more severe degrees of stenosis (7). ZWIEBEL et al. (26) also found external carotid imaging less satisfactory than internal carotid imaging. Calcification produces acoustic shadowing, hindering visualisation of the lumen, and the thrombus can be difficult to visualise due to its generally hypoechoic characteristics (1). The duplex technique has several disadvantages. It requires expertise, and cases with high carotid bifurcations and/or deep, small, tortuous, or acutely angled vessels are difficult to examine. Inaccurate Doppler angle correction can cause significant errors, the angle being difficult to determine in cases of a tortuous vessel, eccentric stenoses, or calcified plaques (21). Colour Doppler was not used in this series due to equipment restrictions. Stenosis is depicted by means of both abnormal colour changes and visible luminal narrowing. Colour Doppler examinations have shown similar degrees of accuracy, sensitivity, and specificity to conventional duplex Doppler examinations (5). In addition, it can clarify confusing situations and shorten the examination time. In mild stenoses, flow values for the ECAs are higher than for the ICAs, and it is only in severe stenosis that the peak systolic velocity and systolic velocity ratios are comparable. edveca and edvecnedvcca are inaccurate and external carotid stenosis is more difficult to measure in the grey-scale image. Ipsilateral internal carotid stenosis affects the flow parameters of the ECA. Thus, measurement of external carotid stenosis is more complicated and less accurate than stenosis in the internal carotid. Peak systolic velocity, systolic velocity ratios, and grey-scale imaging measurement are nevertheless useful predictors of stenosis. REFERENCES 1. ACKER J. D., COLE C. A., MAUNEY K. A. et al.: Duplex carotid ultrasound. Neuroradiology 28 (1986), ALEXANDROV A. V., BLANDIN C. F., MAGGISANO R. et al.: Measuring carotid stenosis. Time for reappraisal. Stroke 24 (1993), BECKETT W. W., DAVIS P. C. & HOFFMAN J. C.: Duplex Dop- pler sonography of the carotid artery. False-positive results in an artery contralateral to an artery with marked stenosis. AJR 155 (1990), BLACKSHEAR W. M., PHILLIPS D. J., CHIKOS P. M. et al.: Carotid artery velocity patterns in normal and stenotic vessels. Stroke 11 (1988) CARROLL B. A.: Carotid sonography. Radiology 178 (1991), CONNOLLY J. E., BROWNELL D. A., LEVINE. F. et al.: Accuracy and indications of diagnostic studies for extracranial carotid disease. Arch. Surg. 120 (1985), DREISBACH J. N., SEIBERT C. E., SMAZAL C. E. et al.: Duplex sonography in the evaluation of carotid artery disease. AJNR 4 (1983), FELL G., PHILLIPS D. J., CHIKOS P. M. et al.: Ultrasonic duplex scanning for disease of the carotid artery. Circulation 64 (1981), FRIEDMAN S. G., HAINLINE B., FEINBERG A. W. et al.: Use of diastolic velocity ratios to predict significant carotid artery stenosis. Stroke 19 (1988), GARTH K. E., CARROLL B. A., SOMMER F. G. et al.: Duplex ultrasound scanning of the carotid arteries with velocity spectrum analysis. Radiology 147 (1983), HANLEY J. A. & MCNE~L B. J.: The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143 (1982), HARTNELL G. G., GIBSON R. N., WILLIAMS R. et al.: Carotid artery stenosis - reassessment of the Doppler criteria. Eur. J. Radiol. 10 (1990), HAYES A. C., JOHNSTON K. W., BAKER W. H. et al.: The effect of contralateral disease on carotid Doppler frequency. Surgery 103 (1988), HUMNK M. G. M., POLAK J. F., BARLAN M. M. et al.: Detection and quantification of carotid artery stenosis. Efficacy of various Doppler velocity parameters. AJR 160 (1993), JACKSON V. P., KUEHN D. S., BENDICK P. J. et al.: Duplex carotid sonography. Correlation with digital subtraction angiography and conventional angiography. J. Ultrasound Med. 4 (1983, JACOBS N. M., GRANT E. G., BYRD M. C. et al.: Duplex carotid sonography. Criteria for stenosis, accuracy, and pitfalls. Radiology 154 (1985), MCNEIL B. J. & ADELSTEIN S. J.: Determining the value of diagnostic and screening tests. TNM 17 (1976), MONETA G. L., EDWARDS J. M., CHITWOOD R. W. et al.: Correlation of North American symptomatic carotid endarterectomy trial (NASCET) angiographic definition of 70% to 99% internal carotid artery stenosis with duplex scanning. J. Vasc. Surg. 17 (1993), PAIVANSALO M., LEINONEN A., TURUNEN J. et al.: Duplex ultrasound compared to arteriography in internal carotid stenosis (in press). 20. POLAK J. F., DOBKIN G. R., O LEARY D. H. et al.: Internal carotid artery stenosis. Accuracy and reproducibility of color- Doppler-assisted duplex imaging. Radiology 173 ( 1989), ROBINSON M. L.: Duplex sonography of the carotid arteries. Semin. Roentgenol. 27 (1992), ROBINSON M. L., SACKS D., PERLMUTER G. S. et al.: Diagnostic criteria for carotid duplex sonography. AJR 151 (1988), SPADONE D. P., BARKMEIER L. D., HODGSON K. J. et al.: Contralateral internal carotid artery stenosis or occlusion. Pitfall of correct ipsilateral classification - a study performed with color-flow imaging. J. Vasc. Surg. 11 (19901,

6 DUPLEX US OF THE EXTERNAL CAROTID ARTERY 24. WITHERS C. E., GOSINK B. B., KEIGHTLEY A. M. et al.: Duplex carotid sonography. Peak systolic velocity in quantifying internal carotid artery stenosis. J. Ultrasound Med. 9 (1990), WOLWERSON M. K., HEIBERG E., SUNLIARAM M. et al.: Carotid atherosclerosis. High-resolution real-time sonography correlated with angiography. AJR 140 (1983), 355 and AJNR 3 (1983), ZWIEBEL W. J., AUSTIN C. W., SACKETT J. F. et al: Correlation of high-resolution, B-mode and continuous-wave Doppler sonography with arteriography in the diagnosis of carotid stenosis. Radiology 149 (1983),