Validity of B-Mode Ultrasonographic Findings in Patients Undergoing Carotid Endarterectomy in Comparison With Angiographic and Clinicopathologic Features
Background and Purpose Determining factors for performing carotid endarterectomy (CEA) are the severity of carotid stenosis and the presence of plaque ulcerations. Precise detection of these factors is important. The aim of this study was to assess the applicability of using B-mode ultrasonography for diagnosing carotid lesions in patients undergoing CEA.
Methods B-mode examinations were performed on 64 Japanese patients (68 arteries) who subsequently underwent CEA. In each case, the appearance of plaque on B-mode was classified into one of two types: heterogeneous or homogeneous; plaque echogenicity was expressed as hypoechoic, isoechoic, or hyperechoic. Surface characteristics such as ulceration also were examined, and the degree of carotid artery stenosis was calculated. The B-mode findings were compared with angiographic and pathological features.
Results B-mode accurately visualized macroscopic ulceration and surface irregularity in 93.8% of the lesions examined, which was superior to angiography. Fifty-four lesions (79.4%) were of the heterogeneous type and 14 lesions (20.6%) were of the homogeneous type on B-mode. Microscopically, 57.4% of the heterogeneous-type lesions were a mixture of atheroma and fibrosis, and all homogeneous-type lesions demonstrated fibrous change. The frequency of calcification was higher in the heterogeneous lesions than in the homogeneous lesions.
Conclusions B-mode ultrasonography findings can provide information about macroscopic and microscopic characteristics of carotid lesions.
The main considerations in deciding to perform a CEA are the severity of carotid artery stenosis and the presence of plaque ulceration.1 Whereas a luminal narrowing >70% in symptomatic patients has been the accepted threshold for performing a CEA,2 the importance of plaque ulceration in patients with stenosis <70% has been uncertain.3 Additionally, recent studies showed a much less dramatic difference between medical and surgical management of patients with asymptomatic carotid artery stenosis.4 5 But the latest data on asymptomatic stenosis have shown that CEA is effective in preventing stroke for >60% luminal narrowing in asymptomatic patients.6 A major factor causing this difference could be the inaccurate preoperative detection of the pathological characteristics in the carotid stenotic lesions.1 Although angiography has been the primary examination performed in patients with carotid lesions, pathological characteristics of the lesions cannot always be assessed by this method.7 8 B-mode ultrasonography has been recently recommended to noninvasively obtain accurate information about surface and plaque characteristics of the carotid artery.9 10
In this study, we evaluated the diagnostic value of B-mode in patients with carotid artery stenosis who subsequently underwent CEA by comparing B-mode findings with (1) those of angiography in an assessment of the preoperative location and severity of stenotic lesions, (2) intraoperative macroscopic characteristics of the lesions, and (3) microscopic findings in the lesions.
Subjects and Methods
B-mode examinations were performed in 64 Japanese patients (55 men and 9 women; age, 42 to 77 years [mean, 63.7 years]) who underwent CEA (35 right and 33 left carotid arteries). Of a total of 68 lesions, 13 lesions were present after transient ischemic attacks, one after reversible ischemic neurological deficits, and 28 after minor strokes. The remaining 26 lesions were asymptomatic.
Examination With B-Mode
An 8-MHz probe (Sonobista CARO model MEU 2422, Biosound) was used. The patients were examined in the supine position with their heads slightly rotated to the contralateral side of the carotid artery being studied.11 The carotid arteries were routinely investigated in the longitudinal and axial planes and were observed from multiple directions by shifting the probe. The visualized carotid artery was divided into the following three segments according to the modified Espeland classification12 : ICA, BIF, and CCA (Fig 1⇓). The ICA segment is defined as the portion distal to the flow divider. The BIF segment extends from the flow divider 8 mm proximally to the CCA. The CCA segment extends proximally from the BIF segment. The locations of the carotid lesions were categorized into one of five groups (ICA, ICA-BIF, BIF, BIF-CCA, and CCA) according to the position of the most stenotic part of the lesion. For example, a lesion located only in the ICA segment was placed in the ICA group, a lesion with the most stenotic part extending from the ICA to the BIF was placed in the ICA-BIF group, and a lesion located only in the CCA segment was placed in the CCA group (Fig 1⇓).
An atherosclerotic lesion was defined as an intima-media complex with a thickness of more than 2 mm.13 The appearance of plaque was classified into one of the two types: heterogeneous (type I) or homogeneous (type II).11 14 15 Additionally, the echogenicity of the plaque was expressed as hypoechoic, isoechoic, or hyperechoic. An isoechoic plaque was defined as having the echogenicity of a normal intima-media complex.11 16 A hyperechoic plaque was brighter than an isoechoic plaque, and a hypoechoic plaque was less bright than an isoechoic plaque. Type I lesions were composed of a mixture of hypoechoic, isoechoic, and hyperechoic plaques. Type II lesions consisted of only one of the three types of echogenic plaques. Type I lesions were further divided into four subtypes: type Ia lesions were composed of a mixture of hyper- and isoechoic plaques. Type Ib lesions were composed of a mixture of iso- and hypoechoic plaques. Type Ic lesions were composed of a mixture of hyper- and hypoechoic plaques. Type Id lesions were composed of a mixture of hypo-, iso-, and hyperechoic plaques. Type II lesions were further divided into three subtypes: type IIa, IIb, and IIc lesions were hyper-, iso-, and hypoechoic, respectively. A typical example of a type Id lesion is shown in Fig 2A⇓, and a typical example of a type IIb lesion is shown in Fig 2B⇓.
Surface characteristics of the plaque were classified into three categories: ulcerated, irregular/uncertain ulceration, or smooth/no ulceration.17 An ulcerated plaque was characterized by the absence of visualization of the intimal layer and presence of a crater in the plaque surface.11 An irregular lesion was defined as a lesion lacking surface smoothness and having an intimal layer. A smooth lesion was defined as a lesion without ulceration and surface irregularity.
The degree of carotid stenosis was measured on the longitudinal section demonstrating the most severe stenosis (Fig 3A⇓). We measured the distance from the near media-adventitia interface to the far media-adventitia interface (A) and the distance from the near intimal surface to the far intimal surface (B) at the most stenotic point of the artery.9 13 The degree of carotid stenosis on B-mode was calculated as (A−B)/A×100.
Selective carotid angiography was performed using anteroposterior, lateral, and oblique views.8 Arterial lumen diameters were measured in the most stenotic view at the most stenotic site (Fig 3B⇑),2 and the normal arterial lumen was extrapolated at this site. The appearance of the lesion was classified into one of three categories: ulcerated, irregular/uncertain ulceration, or smooth/no ulceration.8 Briefly, an ulcerated plaque was defined as a lesion seen in profile as a crater penetrating into a stenotic plaque.8 An irregular lesion was defined as wall irregularity or multiple small craters.8
Macroscopic and Microscopic Examination
CEA was performed on patients under general anesthesia, widely exposing the carotid arteries from the ICA to the CCA. An internal shunt system was applied in all cases during cross-clamping of the carotid arteries.19 After arteriotomy, the involved site was macroscopically investigated to clarify the surface characteristics.8 Then the stenotic lesion was carefully removed en bloc. The surface of the excised specimens and the axial sections were macroscopically examined and then photographed. The first 20 CEA specimens were not macroscopically evaluated by the pathologist. Hence, the macroscopic correlations were obtained only from 48 CEA specimens. These surface characteristics of the lesions were classified into three categories: ulcerated, irregular/uncertain ulceration, and smooth/no ulceration. Macroscopic ulcerations were defined by both the surgeon and the pathologist as either disruptions in the surface of the plaque or pits and depression in the plaque, often with a sharply delineated color difference between the base of the pit and the adjacent luminal surface.8 Irregular lesions were defined as plaques with surface pits and plaques without disruptions and color changes.
The specimens were fixed in a 10% formalin solution and stained with hematoxylin-eosin for microscopic examination. Pathological examination was performed in tissue from the main stenotic lesion, including at least four axial sections.10 The existence of intraplaque hemorrhage and calcification was examined in detail. The lesions were microscopically classified into one of five categories: (1) atheromatous lesions composed mainly of lipid material, (2) fibrous lesions composed of fibrotic proliferation over 70% of the lesion, (3) complicated lesions of atheroma and fibrosis, (4) intraplaque hemorrhage within complicated lesions, and (5) miscellaneous lesions.20 21 A calcified plaque was defined as having macroscopic evidence of calcification or more than five calcified deposits of >1 mm in diameter.
Evaluation of Stenotic Lesions
Location of Lesions
B-mode demonstrated all 68 carotid lesions located in the ICA, ICA-BIF, or CCA segments. ICA lesions were seen in 5 (7.4%), ICA-BIF in 58 (85.3%), and CCA in 5 (7.4%) carotid arteries. Furthermore, B-mode showed that 25 of the 58 ICA-BIF lesions extended into the CCA segment. On the other hand, angiography showed stenotic lesions in the ICA segment in 3 (4.4%) and in the ICA-BIF segment in 65 (95.6%) carotid arteries. No main stenotic lesions were visualized in the CCA segment. The sites of the main stenotic lesions demonstrated by B-mode coincided with those observed by angiography in 61 carotid arteries (89.7%). However, B-mode did not demonstrate distal lesions as seen by angiography in the other 7 stenotic lesions because of the limited examining field. Five of these 7 stenotic lesions were diagnosed as CCA lesions with B-mode in contrast to ICA or ICA-BIF lesions with angiography.
B-Mode Findings of Carotid Stenotic Lesions
Echogenicity classification of stenotic lesions is shown in Table 1⇓. Fifty-four (79.4%) of the 68 carotid stenotic lesions were classified as type I and the remaining 14 (20.6%) as type II. The most common type I subtype lesion was type Id (22 lesions, 40.7%), consisting of hypoechogenic, isoechogenic, and hyperechogenic plaques. The most common type II subtype lesion was type IIb (7 lesions, 50.0%), consisting of isoechoic plaques.
Relationship Between Carotid Lesion Location and Echogenicity
The relationship between location of lesions and their echogenicity is shown in Table 2⇓. Type I lesions were observed in 3 (60.0%) of 5 ICA lesions, 47 (81.0%) of 58 ICA-BIF lesions, and 4 (80.0%) of 5 CCA lesions. In the 58 ICA-BIF lesions, type Id and type Ib were seen in 22 (37.9%) and 15 (25.9%) lesions, respectively. Additionally, type Id lesions were detected only in the ICA-BIF. Eleven of 14 type II lesions were observed in the ICA-BIF.
Degree of Stenosis
The degree of carotid stenosis as determined with B-mode ranged from 15% to 97.4% (mean, 49.0%) and that by angiography from 22.6% to 99.5% (mean, 65.4%) (Fig 4⇓). B-mode tended to underestimate the degree of stenosis compared with angiography. In the 68 stenotic carotid lesions, there was no significant linear correlation between the degree of stenosis as determined by B-mode compared with angiography (r=.15).
Comparison of B-Mode, Angiographic, and Macroscopic Findings
Macroscopic examinations were performed on 48 stenotic carotid lesions by both a surgeon and a pathologist. B-mode, angiographic, and macroscopic findings of the surface characteristics are summarized in Table 3⇓.
In these 48 carotid lesions, B-mode diagnosed ulcerations and/or surface irregularities in 38 (79.2%) and angiography in 24 (50.0%). B-mode and angiographic findings of ulcerations and/or surface irregularities correlated with macroscopic features in 45 (93.8%) and 29 (60.4%) of the 48 carotid lesions, respectively. The correlation rate of B-mode was significantly higher than that of angiography (χ2 test, P<.01).
Comparison of B-Mode and Microscopic Examinations
The microscopic features of the stenotic carotid lesions are summarized in Table 4⇓. Mixed lesions with atheroma and fibrosis were most frequently seen (34 [50.0%] of 68 lesions). Fibrosis alone was seen in 15 carotid lesions and atheroma alone in 7. Calcification was observed in 21 carotid lesions and intraplaque hemorrhage in 8. Of the 21 lesions with calcification, 19 (90.5%) showed heterogeneous echogenicity (type I). Thirty-one (57.4%) of the 54 type I lesions were composed of atheroma and fibrosis microscopically. Calcification was found in 12 (54.5%) of the 22 type Id lesions and in 3 (15.8%) of the 19 type Ib lesions. The frequency of calcification was significantly higher in the type Id lesions than in the type Ib lesions (χ2 test, P<.05). Fig 5A⇓ shows a typical example of a type Id lesion with a mixture of hypo-, iso-, and hyperechogenicity at the ICA-BIF. Microscopic examination of the surgical specimen demonstrated an atheromatous lesion with calcification.
Intraplaque hemorrhage was more frequently seen in type Ib and Id lesions. Only one type II lesion indicated intraplaque hemorrhage.
A typical example of a type II lesion is shown in Fig 5B⇑. B-mode demonstrated a homogeneous isoechoic lesion that was macroscopically composed of fibrous thickening of the intimal layer. Microscopically, it was composed of hyalinized fibrosis. Type II lesions did not contain solely atheroma, but all had fibrosis.
Diagnostic Applicability of B-Mode for Detection of Stenotic Carotid Lesions
Advantages and Limitations of B-Mode
B-mode provides a real-time noninvasive assessment of carotid lesions.7 Recently, the quality of B-mode pictures has been dramatically improved; therefore, B-mode findings may be useful not only for identifying stenotic lesions in the carotid arteries but also for providing information about surface and plaque characteristics of the lesions.9 16 22 B-mode, however, is a less reliable technique for visualizing stenotic carotid lesions under the following conditions: (1) when the carotid bifurcation is located above the mandibular edge, (2) when the carotid arteries are markedly tortuous, (3) when lesions are severely calcified, (4) when lesions are echolucent,22 23 and (5) when the carotid artery is completely occluded.7 24 In terms of the carotid bifurcation being higher than the mandibular edge, our Japanese patients may be problematic because the carotid bifurcation is located at the upper part of C3 on the average (R. Kagawa, T. Shima, Y. Okada, A. Toyota, and L. Csiba, unpublished data, 1994).
Location of Stenotic Carotid Lesions and Degree of Stenosis
In this study, the ICA-BIF was the most frequently involved site, which agrees with a previous report.23 The main lesions at the ICA-BIF extended into the CCA, and the border between lesions and normal arteries was not well defined. We therefore thought that careful observation and management of lesions in the CCA would be important.
Agreement of lesion location between B-mode and angiography was observed in 61 of 68 lesions (nearly 90%). Disagreement between the two methods (7 lesions, 10.3%) may have been attributable to the following reasons. First, measurement of the stenotic site tends to be inaccurate with angiography because only the internal wall of the carotid artery is visualized. In contrast, B-mode visualizes changes throughout the carotid artery wall, such as thickening of the intima and thrombus formation. Second, detection of high lesions distal to the mandibular edge by B-mode is difficult, especially in our series of patients, because the probe cannot visualize the distal ICA. Because of this, B-mode could not detect the seven distal lesions depicted by angiography.
B-mode has been suggested to cause underestimation of the degree of stenosis compared with angiography, and no statistically significant correlation was observed between the degree of stenosis as determined by B-mode and angiography. The following difference in measurement between the two methods could account for this result: B-mode accurately shows the outer arterial wall being used for calculation of the stenotic rate. Angiography, however, extrapolates lines from an uninvolved site to obtain the normal arterial lumen at the involved site, which may vary because of poststenotic dilatation in the distal ICA.7
Echogenicity of Stenotic Carotid Lesions
In this study, heterogeneous (type I) lesions were found more frequently than homogeneous (type II) ones. This result was consistent with that of Hennerici et al.25 Type Ib lesions (a mixture of iso- and hypoechogenicities) and type Id lesions (a mixture of hypo-, iso-, and hyperechogenicities) were predominant, which might reflect the multifarious nature of stenotic carotid lesions. It has been suggested that many embolic events coincide with recent intraplaque hemorrhage, which gives heterogeneity to the lesion.7 26 Lipid lesions, demonstrated by low echogenicity, have a high tendency to develop thrombosis, embolism, and ulceration.10 Therefore, accurate information about lesion echogenicity is important in planning therapeutic strategies for stenotic carotid lesions. Type Id lesions were most frequently observed at the ICA-BIF and were not detected at either the ICA or the CCA. This preference suggests that microscopic features of the stenotic lesions are closely related to the site of the lesion.
Relationship Between B-Mode and Angiography in Comparison With Macroscopic Features
B-mode is known to diagnose ulceration in 50% to 95% of patients with carotid artery stenosis7 23 and to have a higher sensitivity than angiography in the detection of ulceration and intraplaque hemorrhage.7 27 False-negative results with B-mode may increase unless lesions are observed from multiple directions.23 Our B-mode examinations demonstrated ulcerations and/or surface irregularities in 38 (79.2%) of 48 macroscopically diagnosed lesions. B-mode examination of surface characteristics was significantly superior to angiography: the B-mode findings were 93.8% accurate, but the angiographic findings were 60.4% accurate.
Relationship Between B-Mode and Microscopic Features
It has been reported that fatty tissue and hemorrhage are hypoechoic, fibrosis is isoechoic or slightly hyperechoic, and calcification is hyperechoic with acoustic shadowing.7 16 26 28 The majority of heterogeneous (type I) lesions in this study were composed of a mixture of atheroma and fibrosis. More than 90% of all calcified lesions were type I lesions. Calcification therefore should be accompanied by a mixture of atheroma and fibrosis. Calcification was significantly more frequent in type Id lesions than in type Ib lesions, which suggests that hyperechoic components were closely related to calcification.
Intraplaque hemorrhages can be seen as echolucent areas with caps of echodense material.15 26 28 Intraplaque hemorrhage was more frequent in type I lesions than in type II lesions. All lesions with intraplaque hemorrhage, except two, contained hypoechoic components. These results suggest that hypoechoic components are closely related to hemorrhage. Differentiation between intraplaque hemorrhage, indicated by hypoechogenicity, and ulceration of echolucent areas has been difficult with B-mode.15
Homogeneous type lesions have been shown to consist most often of fibrosis.7 Hyperechoic and isoechoic homogeneous (type II) lesions were found to have mainly fibrotic changes in our study. This result was consistent with that of Matalanis and Lusby.26
From this study, we suggest that B-mode and angiographic findings have considerable differences in estimations of the degree of stenosis and surface and plaque characteristics of carotid lesions. Furthermore, the echogenicity of carotid lesions is closely related to the macroscopic and microscopic features of the lesions.
Selected Abbreviations and Acronyms
|BIF, Bif||=||carotid bifurcation|
|CCA||=||common carotid artery|
|ICA||=||internal carotid artery|
The authors thank Shigejirou Matsumura, MD, for scientific comments and Toshihiro Nishida, MD, for pathological examinations.
- Received August 3, 1995.
- Revision received December 14, 1995.
- Accepted December 14, 1995.
- Copyright © 1996 by American Heart Association
Feinberg WM, Albers GW, Barnett HJM, Biller J, Caplan LR, Carter LP, Hart RG, Hobson RW II, Kronmal RK, Moore WS, Robertson JT, Adams HP Jr, Mayberg M. Guidelines for the management of transient ischemic attacks. Stroke. 1994;25:1320-1335.
Wechsler LR. Ulceration and carotid artery disease. Stroke. 1988;19:650-653.
Toole JF, Hobson RW II, Howard VJ, Chambless LE. Nearing the finish line? The Asymptomatic Carotid Atherosclerosis Study. Stroke. 1992;23:1054-1055.
The CASANOVA Study Group. Carotid surgery versus medical therapy in asymptomatic carotid stenosis. Stroke. 1991;22:1229-1235.
Streifler JY, Eliasziw M, Fox AJ, Benavente OR, Hachinski VC, Ferguson GG, Barnett HJM. Angiographic detection of carotid plaque ulceration. Stroke. 1994;25:1130-1132.
Handa N, Matsumoto M, Maeda H, Hougaku H, Ogawa S, Fukunaga R, Yoneda S, Kimura K, Kamada T. Ultrasonic evaluation of early carotid atherosclerosis. Stroke. 1990;21:1567-1572.
Urbani MP, Picano E, Parenti G, Mazzarisi A, Fiori L, Paterni M, Pelosi G, Landini L. In vivo radiofrequency-based ultrasonic tissue characterization of the atherosclerotic plaque. Stroke. 1993;24:1507-1512.
Zwiebel WJ. Cerebrovascular Ultrasound, Doppler and B-Mode Techniques. Chicago, Ill: Year Book Medical Publishers Inc; 1986:27-50.
Espeland MA, Hoen H, Byington R, Howard G, Riley WA, Furberg CD. Spatial distribution of carotid intimal-medial thickness as measured by B-mode ultrasonography. Stroke. 1994;25:1812-1819.
Fabris F, Zanocchi M, Bo M, Fonte G, Poli L, Bergoglio I, Ferrario E, Pernigotti L. Carotid plaque, aging, and risk factors. Stroke. 1994;25:1133-1140.
Handa N, Ogawa S, Mieno M, Maeda H, Fukunaga R, Isaka Y, Matsumoto M, Kimura K, Kamada T. Progression and regression of carotid atherosclerosis: an ultrasonic study [in Japanese]. J Jpn Coll Angiol. 1989;29:1237-1241.
Hennerici M, Kleophas W, Gries FA. Regression of carotid plaques during low-density-lipoprotein cholesterol elimination. Stroke. 1991;22:989-992.
Alexandrov AV, Bladin CF, Maggisano R, Norris JW. Measuring carotid stenosis. Stroke. 1993;24:1292-1296.
Ross Russell RW. Vascular Disease of the Central Nervous System. New York, NY: Churchill Livingstone Inc; 1983:51-54.
Harrison MJG, Dyken ML. Neurology 3: Cerebral Vascular Disease. London, UK: Butterworths; 1983:3-26.
Marshall WG Jr, Kouchoukos NT, Murphy SF, Pelate C. Carotid endarterectomy based on duplex scanning without preoperative arteriography. Circulation. 1988;78(suppl I):I-1-I-5.