Original Article
267
Echographic Risk Index and Cerebral Ischemic Brain Lesions in Patients Randomized to Stenting versus Endarterectomy for Symptomatic Carotid Artery Stenosis Die Bedeutung des echografischen Risikoindex für das Risiko zerebraler Ischämien bei Stentbehandlung oder Endarterektomie der symptomatischen Karotisstenose Authors
A. Burow1, P. A. Lyrer1, P. J. Nederkoorn2, M. M. Brown3, R. Sztajzel4, S. T. Engelter1, L. H. Bonati1
Affiliations
1
3 4
Department of Neurology and Stroke Unit, University Hospital Basel Department of Neurology, Academic Medical Center Amsterdam Institute of Neurology, University College London Department of Neurology, University Hospital Geneva
Key words
Zusammenfassung
Abstract
"
!
!
Ziel: Plaqueeigenschaften beeinflussen das Risiko periprozeduraler Embolien während Stent-Angioplastie (CAS) und Endarterektomie (CEA). Es ist unklar, ob sich dieser Einfluss zwischen CAS und CEA unterscheidet. Wir haben untersucht, ob die quantitative Bestimmung der Plaque-Echogenität die Embolierate während CAS oder CEA vorhersagt. Material und Methoden: Wir untersuchten 50 ICSS randomisierte CAS oder CEA Patienten mit semiautomatischen ultraschallgestützten Graustufenmessungen von Karotisplaques. Bestimmt wurden Graustufen-Median (GSM), Prozent der echoluzenten Plaqueanteile, sowie ein echografischer Risikoindex (ERI). Zerebrale Kernspintomografien inklusive DWI-Sequenzen wurden vor und nach Behandlung durchgeführt. Als primärer Endpunkt galt mindestens eine neue hyperintense DWI-Läsion (DWI+). Ergebnisse: Innerhalb der CAS-Gruppe zeigten DWI+ Patienten (n = 18) signifikant höhere ERIWerte (Mittelwert 0,11 ± 0,12) im Vergleich zu Patienten ohne Läsionen (DWI−; n = 8; 0,03 ± 0,01; p = 0,012). GSM (26,7 ± 18,7 versus 34,3 ± 8,0, p = 0,16) und hypoechogene Plaqueanteile (42,8 ± 21,1 versus 31,2 ± 8,2, p = 0,054) zeigten keine signifikanten Unterschiede. In der CEA-Gruppe bestanden keine Unterschiede in der Plaqueechogenität zwischen DWI+ (n = 2) und DWI− Patienten (n = 22). Schlussfolgerung: Patienten mit echoluzenten Plaques und höhergradiger Stenose tragen ein erhöhtes Risiko cerebraler Embolien während CAS. Die quantitative ultraschallgestützte Plaqueanalyse liefert Anhaltspunkte zur Identifizierung von Patienten, die ungeeignet für CAS sind.
Purpose: It remains to be determined whether the impact of plaque characteristics on procedural risks differs between carotid artery stenting (CAS) and endarterectomy (CEA). We studied whether quantitative assessment of carotid plaque echolucency on ultrasound predicts the risk of embolism during CAS or CEA. Materials and Methods: In 50 consecutive patients with symptomatic carotid stenosis randomized to CAS (n = 26) or CEA (n = 24) in the International Carotid Stenting Study (ICSS), semiautomated grayscale measurement of carotid plaques on baseline ultrasound was performed. We determined the grayscale median (GSM), percentage of echolucent plaque area, and a previously defined echographic risk index (ERI) calculated with the echolucent area and degree of stenosis. Brain MRI including diffusion-weighted imaging (DWI) was performed within 7 days before and 3 days after treatment. The primary outcome was the presence of at least 1 new hyperintense DWI lesion (DWI+) after treatment. Results: In the CAS group, DWI+ patients (n = 18) had a significantly higher ERI at baseline (mean 0.11 ± 0.12) than patients without new lesions (n = 8; mean 0.03 ± 0.01; p = 0.012). GSM (mean 26.7 ± 18.7 versus 34.3 ± 8.0, p = 0.16) and echolucent plaque area (mean 42.8 ± 21.1 versus 31.2 ± 8.2, p = 0.054) did not differ significantly. In the CEA group, there were no differences in plaque echogenity measurements between patients with (n = 2) and without DWI lesions (n = 22). Conclusion: Patients with echolucent plaques causing severe narrowing are at increased risk for cerebral embolism during CAS. Quantitative ultrasound plaque analysis, with ERI in particular, may add to clinical variables in identifying patients at risk for procedural stroke with CAS, but larger studies with clinical endpoints are needed.
● carotid stenosis ● plaque echolucency ● stenting ● endarterectomy ● embolism " " " "
received accepted
9.4.2013 3.9.2013
Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1355751 Published online: October 18, 2013 Ultraschall in Med 2014; 35: 267–272 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0172-4614
Correspondence Dr. med. Annika Burow Department of Neurology, University Hospital Basel Petersgraben 4 4031 Basel Switzerland Tel.: ++ 41/6 12 65 25 25 Fax: ++ 41/6 12 65 56 44 [email protected]
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Downloaded by: Universite Laval. Copyrighted material.
2
Original Article
Introduction !
About one in ten ischemic strokes is caused by atherosclerotic stenosis of the internal carotid artery. Carotid endarterectomy (CEA) reduces the risk of stroke in patients with symptomatic carotid stenosis [1, 2]. Carotid artery stenting (CAS) has emerged as a potential alternative to CEA, but is associated with an increased risk of peri-procedural stroke [3]. A pooled analysis of three trials showed that the excess stroke risk with CAS in comparison with CEA was mainly observed in patients above the age of 70 [4]. However, factors other than age may also determine the risk of peri-procedural stroke and aid in the selection of the appropriate treatment strategy. Instability of the atherosclerotic plaque plays an important role in the occurrence of ischemic cerebrovascular events associated with carotid stenosis [5, 6], both spontaneously and during treatment. Reduced plaque echogenicity on duplex ultrasound Bmode images (plaque echolucency) corresponds to histological markers of plaque instability [7, 8] and is associated with an increased risk of stroke [8 – 13]. More recently, an Echographic Risk Index (ERI) based on a combination of degree of stenosis and proportion of echolucent area at the plaque surface has been demonstrated to predict symptom status and presence of ipsilateral ischemic lesions on magnetic resonance imaging (MRI) [14]. Quantitative assessment of plaque ultrasound therefore has the potential to identify “unstable” plaques which are more likely to cause cerebrovascular events or subclinical embolization. However, plaque instability may also influence the risk of peri-procedural stroke during CAS or CEA. So far, the impact of ultrasound plaque morphology on the peri-procedural cerebral ischemia risk has not been compared between CAS and CEA in a randomized study. We investigated whether quantitative assessment of carotid plaque echolucency, alone or in conjunction with degree of stenosis, would predict the risk of cerebral ischemia on MRI in patients randomly assigned to CAS or CEA in the International Carotid Stenting Study (ICSS) [15].
Methods !
ICSS was a large randomized trial comparing the safety and efficacy of CAS versus CEA in 1710 patients with carotid stenosis of ≥ 50 % luminal narrowing and associated symptoms within the past 12 months [15]. In an MRI substudy nested within ICSS, 231 patients (CAS: n = 124, CEA: n = 107) enrolled at 7 ICSS centers received magnetic resonance imaging (MRI) scans of the brain 1 – 7 days before and 1 – 3 days after treatment, in order to assess silent ischemic and hemorrhagic brain lesions occurring peri-procedurally. Details of ICSS and the ICSS-MRI substudy have been reported in previous publications [15, 16]. The primary imaging endpoint of the ICSS-MRI substudy was cerebral ischemia, defined as at least one hyperintense signal alteration on diffusionweighted imaging (DWI) on the post-treatment MRI scan which was not present before treatment [16]. Baseline carotid imaging in ICSS required to identify carotid lesions suitable for both CEA and CAS included the combination of at least 2 noninvasive imaging techniques (magnetic resonance or computer tomography angiography, or duplex ultrasound) or intra-arterial angiography [17]. Degree of stenosis was measured according to the criteria used in the NASCET trial [1]. Two of the seven centers in the ICSS-MRI substudy participated in the
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
present study. We retrieved digitalized color-coded duplex ultrasound images of carotid stenosis on the relevant side, which were obtained before randomization in 53 patients enrolled at these two centers (27 patients in the CAS group, 26 patients in the CEA group). The study population consisted of 50 patients (CAS: n = 26, CEA: n = 24), in whom ultrasound images were suitable for both visual and semi-automated analysis. Ultrasound imaging was performed on a Siemens Acuson Sequoia 512 machine using a 7 – 5 MHz linear transducer and a Philips HDI machine using a 9 – 3 MHz linear transducer at one center; and on two Philips iU22 machines, one using a 8 – 4 MHz and one using a 9 – 3 MHz linear transducer at the other center. Visual analysis of plaque echolucency on B-mode images was performed offline by a single observer using the classification proposed by Gray-Weale [7]: type 1 (echolucent with echogenic fibrous cap), type 2 (predominantly echolucent, with echogenic areas representing less than 25 % of the plaque), type 3 (predominantly echogenic, with echolucent areas representing less than 25 % of the plaque), and type 4 (echogenic and homogenous plaque). Quantitative, semiautomated analysis of carotid plaques was performed offline by two researchers in consensus, as detailed below. Both researchers were experienced sonographers who were unaware of the allocated treatment, the clinical outcome, and whether patients had DWI lesions after treatment or not. The method of computer-aided plaque analysis used in the present study has been described in detail before [14, 18 – 20]. In brief, the base of the plaque was outlined manually at the adventitial border. The boundary between the color flow and the surface of the plaque was delineated automatically by the computer program. The grayscale values of the plaque were normalized by automatic linear scaling after outlining a black region in the perfused vessel lumen (grayscale value = 0) and a bright region in the vessel adventitia (grayscale value = 190). Each pixel of the carotid plaque was then mapped in 3 different colors: red (echolucent), yellow, and green (echogenic), according to its grayscale value. For the present analysis, we defined echolucent (red) plaque area as grayscale < 20, and echogenic (green) area as grayscale " Fig. 1b). > 50 (● The following parameters were then calculated by the computer program for the entire 2 D section of the plaque: (1) the grayscale median (GSM), i. e. the median value of all grayscale values; (2) the proportion of echolucent (red) plaque area as percentage of the whole plaque area; and (3) the Echographic Risk Index (ERI), which is based on echolucent plaque area and degree of stenosis (determined using standardized duplex flow-velocity criteria) according to the following formula [14]: exp (-8.98404 + 0.0458 × degree of stenosis + 0.06018 × proportion of echolucent area) ERI =
1 + exp (-8.98404 + 0.0458 × degree of stenosis + 0.06018 × proportion of echolucent area)
Statistical Analysis !
Statistical analysis was performed using IBM SPSS Version 20.0 software (Chicago, Illinois). Categorical variables were compared using chi-square tests and continuous variables using T-tests. Analyses were performed separately for the CAS and the CEA group. First, baseline plaque parameters were compared between patients with versus without new lesions after treatment. Second, exploratory post-hoc cross-table analyses were performed to study the association of any plaque parameter differing significantly between patients with versus without DWI lesion, sep-
Downloaded by: Universite Laval. Copyrighted material.
268
Original Article
Results !
Among the primary study population, risk factor profile and demographic baseline characteristics as well as plaque ultrasound parameters did not differ significantly between patients allocated " Table 1). 18 of 26 pato CAS and those who underwent CEA (● tients in the CAS group (69 %) and 2 of 24 patients in the CEA group (8 %) had new DWI lesions after treatment (p < 0.001). In the CAS group, there was no difference in whole-plaque GSM between patients with (mean 26.7 ± 18.7) and without lesions " Fig. 2). There was a trend towards a higher (34.3 ± 8.0, p = 0.16) (● proportion of echolucent plaque area in patients with new le-
Fig. 1 a Predominantly echolucent plaque with hyperechogenic areas. b Color mapping of the plaque shows the following proportions for the whole plaque: red 26 % (GSM < 20), yellow 32 % (GSM > 20 and < 50) and green 42 % (GSM > 50); for the surface one-third of the plaque the colors are distributed as follows: red 32 %, yellow 35 % and green 33 %, corresponding to a total grayscale median (GSM) value of 41.
age mean (SD) female gender n (%)
Abb. 1 a Vorwiegend echoluzente Plaque mit hyperechogenen Anteilen. b Das Farbmapping der Plaque ergibt folgende Werte: Rotanteil 26 % (GSM < 20), Gelbanteil 32 % (GSM > 20 and < 50) und Grünanteil 42 % (GSM > 50); für die Plaqueoberfläche (oberstes Drittel) ergeben sich folgende Farbverteilungen: Rot 32 %, gelb 35 % und grün 33 %, einem GSM von 41 entsprechend.
CAS (n = 26)
CEA (n = 24)
73.2 (9.32)
71.7 (8.25)
p-value ns
6 (23 %)
7 (29 %)
ns ns
Table 1
Patient characteristics.
vascular risk factors n (%) current smoking
6 (23 %)
2 (8 %)
ex-smoker
10 (38 %)
11 (46 %)
ns
hypercholersterolemia
14 (54 %)
12 (17 %)
ns
hypertension
20 (77 %)
20 (83 %)
ns
diabetes
6 (24 %)
6 (25 %)
ns
coronary heart disease
6 (23 %)
2 (83 %)
ns
peripheral artery disease
4 (15.4 %)
4 (16.7 %)
ns
145.77 (26.86)
148.71 (19.77)
ns
4.61 (1.06)
5.03 (1.41)
ns
systolic blood pressure at randomization [mmHg] mean (SD) total cholesterol at randomization [mmol/l] mean (SD) most recent ipsilateral event n (%) hemispheric ischemic stroke retinal stroke TIA
15 (57.7 %)
13 (54.2 %)
ns
6 (23.1 %)
3 (12.5 %)
ns
8 (33.3 %)
ns
79.04 (10.49)
81.16 (9.65)
ns
GSM
29.04 (16.38)
33.33 (15.45)
ns
echolucent area 1
39.23 (18.72)
36.25 (18.11)
ns
0.08 (0.10)
0.08 (0.10)
ns
mean degree of ipsilateral stenosis mean (SD)
5 (19.2 %)
quantitative plaque analysis (whole plaque)
ERI Grey-Weale plaque types n (%) type 1
1 (4.2 %)
type 2
13 (54.2 %)
14 (53.8 %)
0
type 3
9 (37.5 %)
12 (46.2 %)
type 4
1 (4.2 %)
0
CAS = carotid artery stenting; CEA = carotid endarterectomy; SD = standard deviation; TIA = transient ischemic attack; GSM = grayscale median; ERI = echographic risk index; ns = not significant. 1 Percentage of plaque area below 20th percentile of grayscale values.
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Downloaded by: Universite Laval. Copyrighted material.
arated at the median of the study population, and the presence of DWI lesions after treatment. Our study population was too small for multivariate adjustment. Instead, we chose to investigate for correlation of any plaque parameter which significantly predicted DWI lesions with age, which is the strongest known clinical predictor of peri-procedural stroke in CAS, using Pearson’s rank test. To account for multiplicity of comparisons of plaque characteristics between patients with and without DWI lesions (GSM, echolucent area, and ERI), we multiplied p-values by three (Bonferroni correction) for those comparisons significant at p < 0.05 without correction. A p-value < 0.05 after correction was considered to indicate statistical significance.
269
Original Article
34.3 (8.0)
80
CAS
80
26.72 (18.70)
DWI+ (n=18)
Echolucent area*
CAS 42.8 (21.1)
p= 0.054
60
DWI(n=22)
60
31.2 (8.2)
DWI+ (n=2)
CEA
80
Echolucent area*
DWI(n=8)
37.3 (18.2)
24.2 (16.8)
40
40
20
20
Fig. 2 Whole plaque GSM, echolucent plaque area and ERI mean (SD) in patients without lesions compared to patients with lesions in the CAS and CEA group. *Percentage of plaque area below 20th percentile of grayscale values. CAS = carotid artery stenting; CEA = carotid endarterectomy; GSM = grayscale median; ERI = echographic risk index; DWI = diffusion-weighted imaging; DWI− = patients without DWI lesion; DWI+ = presence of DWI lesion; SD = standard deviation. P-values are not corrected for multiple comparisons. Abb. 2 Plaqueparameter GSM gesamt, echoluzenter Plaqueanteil und echografischer Risikoindex (ERI) innerhalb der Patientengruppe ohne Läsionen im Vergleich zu Patienten mit Läsionen innerhalb der CAS- und CEA-Gruppen. *Graustufenwerte < 20. Perzentile. GSM = Graustufen Median; ERI = Echografischer Risikoindex; CAS = Stent-Angioplastie; CEA = Endarterektomie; DWI = diffusionsgewichtete Bildgebung; DWI− = Patienten ohne DWILäsion; DWI+ = Vorhandensein einer DWI-Läsion; SD = Standardabweichung.
0
0 DWI(n=8)
DWI+ (n=18)
DWI(n=22)
CAS p=0.012
CEA
80 0.11 (0.12)
60
ERI
60 40 0.03 (0.01)
40
DWI+ (n=2)
0.08 (0.10)
0.05 (0.05)
DWI(n=22)
DWI+ (n=2)
20 0
0 DWI(n=8)
DWI+ (n=18)
sions (mean 42.8 ± 21.1) than patients without new lesions (mean 31.2 ± 8.2) after treatment (p = 0.054). In the CAS group, patients with new DWI lesions after treatment had a significantly higher whole-plaque ERI (0.11 ± 0.12) than patients without new " Fig. 2). lesions (0.03 ± 0.01; p = 0.012) (● In the CEA group, there were no differences in any plaque echogenicity measures between patients with DWI lesions (n = 2) and those without DWI lesions (n = 22). However, due to the small number of CEA patients with new DWI lesions after treatment, no statistical tests were performed within the CEA group. There were no significant differences in degree of carotid stenosis between patients with versus without DWI lesions both in the CAS group (80.6 % ± 11.6 versus 75.6 % ± 6.8, p = 0.19) and in the CEA group (82 %.5 ± 17.7 versus 81.0 % ± 9.3). In order to further explore the relationship between whole-plaque ERI and risk of peri-procedural DWI lesions, we separated patients at the median ERI value of the entire study population " Fig. 3). All 11 CAS patients with ERI above the median (0.04) (● had new ischemic lesions after treatment (p = 0.07), compared with only 1 of 13 CEA patients with ERI above the median (n. s.). We did not find any correlation at all between whole-plaque ERI and patient age across the entire study population (p = 0.9)
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Fig. 3 Number of patients with and without new diffusion-weighted imaging (DWI) lesions after treatment below and above the median of the Echographic Risk Index (ERI). CAS = carotid artery stenting; CEA = carotid endarterectomy; ERI_low = Patients below the median of the ERI; ERI_high = Patients above the median of the ERI; DWI− = no DWI lesion; DWI+ = presence of DWI lesion. Abb. 3 Anzahl der Patienten mit und ohne DWI-Läsionen nach Behandlung unter- sowie oberhalb des Medianwertes des Echografischen Risikoindex (ERI). CAS = Stent-Angioplastie; CEA = Endarterektomie; ERI_low = Patienten unterhalb des Median des ERI; ERI_high = Patienten oberhalb des Median des ERI; DWI = diffusionsgewichtete Bildgebung; DWI− = keine DWI-Läsion; DWI+ = Vorhandensein einer DWI-Läsion.
Downloaded by: Universite Laval. Copyrighted material.
GSM
GSM
0
0
20
39.50 (6.4)
20
20
80
CEA
40
40
80
32.8 (16.0)
60
60
ERI
270
Original Article
Discussion !
In this study, we investigated whether semi-automated quantitative analysis of plaque echolucency, on its own or in combination with degree of stenosis, might predict the risk of peri-procedural cerebral ischemia in patients with symptomatic carotid stenosis randomly allocated to CAS or CEA. Among patients undergoing CAS, an increase in the Echographic Risk Index (ERI) based on echolucent plaque area and degree of stenosis was associated with an increased risk of cerebral ischemia. Echolucent plaque areas, i. e. areas appearing dark on B-mode ultrasound, correspond to necrotic areas with increased lipid content and plaque hemorrhage in histological examination [7, 21]. Computerized analysis of B-mode ultrasound images allows for normalization against reference tissues and therefore a quantitative and objective assessment of plaque echolucency. Plaque echolucency is traditionally expressed by the grayscale median (GSM), i. e. the median grayscale value in the carotid plaque; GSM inversely correlates with the presence of ipsilateral stroke on CT [8, 11] and risk of future stroke [8 – 11]. More recently, an ERI, based on a combination of degree of stenosis and proportion of echolucent area at the plaque surface, has been demonstrated to be superior to conventional parameters such as degree of stenosis and GSM in predicting symptom status and presence of ipsilateral ischemic lesions on MRI [14]. Previous studies have investigated the relationship between plaque morphology on ultrasound and embolic risk in CAS. In the ICAROS study, the risk of stroke during CAS was higher in patients who had plaques with GSM ≤ 25 than in those with GSM values > 25 [22]. We observed lower average GSM values among patients with new DWI lesions after treatment compared with patients without new lesions in the CAS group, but the difference was not statistically significant. However, ICAROS included both patients with symptomatic and asymptomatic carotid stenosis, for which reason the findings might not be directly comparable to our study which included symptomatic patients only. However, due to its larger sample size, the ICAROS study was more powerful to detect such a difference. A recent study prospectively investigated 31 consecutive patients with symptomatic severe carotid stenosis undergoing CAS with dual-frequency transcranial Doppler sonography (TCD). Patients with echolucent plaques defined by a GSM < 50 more often had solid cerebral microemboli on TCD than patients with echogenic plaques [23]. Our results, together with the findings of previous studies, lend support to the hypothesis that plaque instability may pose a greater risk of peri-procedural embolic stroke in CAS than in CEA. During CAS, where blood flow towards the brain is usually maintained, debris of unstable plaques may be released during catheterization or stent insertion, causing embolism to the brain. In contrast, the carotid artery is clamped off during CEA and there is no direct pathway for emboli to the brain. Cerebral protection devices (CPD) have been designed to prevent cerebral embolization during CAS but their efficacy has not been proven in randomized trials. Use of CPD was recommended, but not mandatory in ICSS. Filter-type CPDs were used in all but 2 patients treated with CAS in the present study population.
Other ultrasound techniques may be useful in selecting patients for CEA or CAS, as well. Plaque neovascularization identified by micro-bubble contrast has been shown to increase the risk of stroke in patients with carotid stenosis [24]. More recently, ultrasound elastography has emerged as a method to investigate mechanical tissue properties, in particular the reaction to shear stress [25, 26]. This technique may also potentially be used to characterize the carotid plaque on a mechanical level, which may add to information gained by imaging [27]. CAS and CEA may be complicated by cerebral embolism which often remains clinically silent [28]. In the ICSS-MRI substudy, new DWI lesions occurred in 50 % of patients treated with CAS and 17 % of patients undergoing CEA [16, 29]. Infarction on DWI has been proposed as a sensitive surrogate outcome measure for the risk of peri-procedural stroke [28]. In the present study, we used DWI lesions after treatment as a surrogate outcome measure for peri-procedural embolism. Our findings warrant confirmation in larger randomized trial populations whether quantitative assessment of plaque echolucency in combination with degree of stenosis predicts the risk of peri-procedural stroke.
Strengths and limitations: The major strength and a novelty of our work is that the allocation to the type of revascularization (CAS or CEA) was randomized. This allowed for a comparison of the effect of plaque characteristics on the risk of peri-procedural embolism between the two treatments. However, we are aware of several limitations of our study. Firstly, patients in the present study were investigated on different ultrasound machines using different transducers. Although echolucency was standardized by two reference tissues as described above, we cannot rule out that there were systematic differences in echolucency measured on different machines. Secondly, the duplex images were not initially obtained specifically for the purpose of quantitative analysis, which may have reduced quality in some cases. Thirdly, our study population comprised patients with high-risk symptomatic carotid stenosis enrolled in a randomized intervention trial, and therefore differed from the original study that reported the value of the ERI in differentiating between patients with symptomatic and asymptomatic stenosis [14]. This probably explains why our patients generally had plaques of lower echogenicity (GSM 29 in the CAS group and 33 in the CEA group) than the patients in the previous study (GSM 43). To be able to differentiate the plaque areas among our study population, we therefore decided to use grayscale cut-off values of < 20 and > 50 to define echolucent and echogenic areas, rather than < 60 and > 90 as in the previous publication. Fourthly, our sample size was relatively small. Thus, the risk of false-positive findings (e. g., the association between ERI and new DWI lesion for the CAS group) and the risk of false-negative findings (e. g. the absence of such an association for the CEA group) have to be acknowledged. Therefore, our findings should be interpreted cautiously and confirmation in a larger dataset is required. In conclusion, we found that quantitative analysis of plaque echolucency in combination with degree of stenosis on carotid ultrasound identified patients at increased risk for peri-procedural embolism with CAS. However, patients with a low echographic risk index still had a lower risk of cerebral ischemia with CEA than with CAS. Our findings warrant the application of quantitative ultrasound analysis in large clinical trials to investigate if this
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Downloaded by: Universite Laval. Copyrighted material.
There were no differences with regard to Grey-Weale plaque type comparing patients with versus without DWI lesions, neither in the whole study population nor the CAS and CEA groups separately.
271
Original Article
technology adds to established clinical risk factors such as age, in identifying patients at risk for peri-procedural stroke with CAS.
Abbreviations CAS CEA DWI DWI− DWI+ ERI ERI_low
carotid artery stenting carotid endarterectomy diffusion-weighted imaging no DWI lesion presence of DWI lesion Echographic Risk Index patients below the median of the Echographic Risk Index ERI_high patients above the median of the Echographic Risk Index GSM grayscale median ICSS International Carotid Stenting Study ns not significant SD standard deviation
08
09
10
11
12
13
14
15
Acknowledgement !
ICSS was funded by grants from the Medical Research Council, the Stroke Association, Sanofi-Synthelabo, and the European Union. Funding for MRI scans performed as part of the ICSS-MRI study was provided by grants from the Mach-Gaensslen Foundation, Switzerland and the Stroke Association, United Kingdom. The work of Dr. Burow for this research was supported by the Stroke(Hirnschlag)-Fund Basel and by a grant from the University of Basel. Professor Brown's Chair in Stroke Medicine is supported by the Reta Lila Weston Trust for Medical Research. Dr. Bonati was supported by grants from the Swiss National Science Foundation (PBBSB-116873 and 33CM30-124119) and the University of Basel. This work was partly performed at University College London Hospital and University College London, which received a proportion of funding from the Department of Health National Institute for Health Research Biomedical Research Centres funding scheme.
References 01 Barnett HJ, Taylor DW, Eliasziw M et al. Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med 1998; 339: 1415 – 1425 02 European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: Final results of the MRC European Carotid Surgery Trial (ECST). The Lancet 1998; 351: 1379 – 1387 03 Bonati LH, Lyrer P, Ederle J et al. Percutaneous transluminal balloon angioplasty and stenting for carotid artery stenosis. Cochrane Database Syst Rev 2012, CD000515 9 04 Bonati LH, Dobson J, Algra A et al. Short-term outcome after stenting versus endarterectomy for symptomatic carotid stenosis: a preplanned meta-analysis of individual patient data. Lancet 2010; 376: 1062 – 1073 05 Aburahma AF, Thiele SP, Wulu JT et al. Prospective controlled study of the natural history of asymptomatic 60% to 69% carotid stenosis according to ultrasonic plaque morphology. J Vasc Surg 2002; 36: 437 – 442 06 Eyding J, Geier B, Staub D. Current strategies and possible perspectives of ultrasonic risk stratification of ischemic stroke in internal carotid artery disease. Ultraschall in Med 2011; 32: 267 – 273 07 Gray-Weale AC, Graham JC, Burnett JR et al. Carotid artery atheroma: comparison of preoperative B-mode ultrasound appearance with car-
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
16
17
18
19
20
21
22
23
24 25
26
27
28
29
otid endarterectomy specimen pathology. J Cardiovasc Surg (Torino) 1988; 29: 676 – 681 Biasi GM, Sampaolo A, Mingazzini P et al. Computer analysis of ultrasonic plaque echolucency in identifying high risk carotid bifurcation lesions. Eur J Vasc Endovasc Surg 1999; 17: 476 – 479 Sabetai MM, Tegos TJ, Clifford C et al. Carotid plaque echogenicity and types of silent CT-brain infarcts. Is there an association in patients with asymptomatic carotid stenosis? Int Angiol 2001; 20: 51 – 57 Gronholdt ML, Nordestgaard BG, Schroeder TV et al. Ultrasonic echolucent carotid plaques predict future strokes. Circulation 2001; 104: 68 – 73 el-Barghouty N, Geroulakos G, Nicolaides A et al. Computer-assisted carotid plaque characterisation. Eur J Vasc Endovasc Surg 1995; 9: 389 – 393 Tegos TJ, Sabetai MM, Nicolaides AN et al. Comparability of the ultrasonic tissue characteristics of carotid plaques. J Ultrasound Med 2000; 19: 399 – 407 Mathiesen EB, Bonaa KH, Joakimsen O. Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: the tromso study. Circulation 2001; 103: 2171 – 2175 Momjian-Mayor I. Accuracy of a Novel Risk index Combining Degree of Stenosis of the Carotid Artery and Plaque Surface Echogenicity. Stroke 2012 Ederle J, Dobson J, Featherstone RL et al. Carotid artery stenting compared with endarterectomy in patients with symptomatic carotid stenosis (International Carotid Stenting Study): an interim analysis of a randomised controlled trial. Lancet 2010; 375: 985 – 997 Bonati LH, Jongen LM, Haller S et al. New ischaemic brain lesions on MRI after stenting or endarterectomy for symptomatic carotid stenosis: a substudy of the International Carotid Stenting Study (ICSS). Lancet Neurol 2010; 9: 353 – 362 Featherstone RL, Brown MM, Coward LJ. International carotid stenting study: protocol for a randomised clinical trial comparing carotid stenting with endarterectomy in symptomatic carotid artery stenosis. Cerebrovasc Dis 2004; 18: 69 – 74 Sztajzel R, Momjian S, Momjian-Mayor I et al. Stratified gray-scale median analysis and color mapping of the carotid plaque: correlation with endarterectomy specimen histology of 28 patients. Stroke 2005; 36: 741 – 745 Sztajzel R, Momjian-Mayor I, Comelli M et al. Correlation of cerebrovascular symptoms and microembolic signals with the stratified grayscale median analysis and color mapping of the carotid plaque. Stroke 2006; 37: 824 – 829 Momjian I, Momjian S, Albanese S et al. Visual analysis or semi-automated gray-scale-based color mapping of the carotid plaque: which method correlates the best with the presence of cerebrovascular symptoms and/or lesions on MRI? J Neuroimaging 2009; 19: 119 – 126 Carotid artery plaque composition–relationship to clinical presentation and ultrasound B-mode imaging. European Carotid Plaque Study Group. Eur J Vasc Endovasc Surg 1995; 10: 23 – 30 Biasi GM, Froio A, Diethrich EB et al. Carotid plaque echolucency increases the risk of stroke in carotid stenting: the Imaging in Carotid Angioplasty and Risk of Stroke (ICAROS) study. Circulation 2004; 110: 756 – 762 Rosenkranz M, Russjan A, Goebell E et al. Carotid plaque surface irregularity predicts cerebral embolism during carotid artery stenting. Cerebrovasc Dis 2011; 32: 163 – 169 Staub D, Partovi S, Imfeld S et al. Novel applications of contrast-enhanced ultrasound imaging in vascular medicine. Vasa 2013; 42: 17 – 31 Bamber J, Cosgrove D, Dietrich CF et al. EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 1: Basic principles and technology. Ultraschall in Med 2013; 34: 169 – 184 Cosgrove D, Piscaglia F, Bamber J et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography.Part 2: Clinical Applications. Ultraschall in Med 2013; 34: 238 – 253 Maurice RL, Soulez G, Giroux MF et al. Noninvasive vascular elastography for carotid artery characterization on subjects without previous history of atherosclerosis. Med Phys 2008; 35: 3436 – 3443 Schnaudigel S, Groschel K, Pilgram SM et al. New brain lesions after carotid stenting versus carotid endarterectomy: a systematic review of the literature. Stroke 2008; 39: 1911 – 1919 Gensicke H, Zumbrunn T, Jongen LM et al. Characteristics of ischemic brain lesions after stenting or endarterectomy for symptomatic carotid artery stenosis: results from the international carotid stenting studymagnetic resonance imaging substudy. Stroke 2013; 44: 80 – 86
Downloaded by: Universite Laval. Copyrighted material.
272
267
Echographic Risk Index and Cerebral Ischemic Brain Lesions in Patients Randomized to Stenting versus Endarterectomy for Symptomatic Carotid Artery Stenosis Die Bedeutung des echografischen Risikoindex für das Risiko zerebraler Ischämien bei Stentbehandlung oder Endarterektomie der symptomatischen Karotisstenose Authors
A. Burow1, P. A. Lyrer1, P. J. Nederkoorn2, M. M. Brown3, R. Sztajzel4, S. T. Engelter1, L. H. Bonati1
Affiliations
1
3 4
Department of Neurology and Stroke Unit, University Hospital Basel Department of Neurology, Academic Medical Center Amsterdam Institute of Neurology, University College London Department of Neurology, University Hospital Geneva
Key words
Zusammenfassung
Abstract
"
!
!
Ziel: Plaqueeigenschaften beeinflussen das Risiko periprozeduraler Embolien während Stent-Angioplastie (CAS) und Endarterektomie (CEA). Es ist unklar, ob sich dieser Einfluss zwischen CAS und CEA unterscheidet. Wir haben untersucht, ob die quantitative Bestimmung der Plaque-Echogenität die Embolierate während CAS oder CEA vorhersagt. Material und Methoden: Wir untersuchten 50 ICSS randomisierte CAS oder CEA Patienten mit semiautomatischen ultraschallgestützten Graustufenmessungen von Karotisplaques. Bestimmt wurden Graustufen-Median (GSM), Prozent der echoluzenten Plaqueanteile, sowie ein echografischer Risikoindex (ERI). Zerebrale Kernspintomografien inklusive DWI-Sequenzen wurden vor und nach Behandlung durchgeführt. Als primärer Endpunkt galt mindestens eine neue hyperintense DWI-Läsion (DWI+). Ergebnisse: Innerhalb der CAS-Gruppe zeigten DWI+ Patienten (n = 18) signifikant höhere ERIWerte (Mittelwert 0,11 ± 0,12) im Vergleich zu Patienten ohne Läsionen (DWI−; n = 8; 0,03 ± 0,01; p = 0,012). GSM (26,7 ± 18,7 versus 34,3 ± 8,0, p = 0,16) und hypoechogene Plaqueanteile (42,8 ± 21,1 versus 31,2 ± 8,2, p = 0,054) zeigten keine signifikanten Unterschiede. In der CEA-Gruppe bestanden keine Unterschiede in der Plaqueechogenität zwischen DWI+ (n = 2) und DWI− Patienten (n = 22). Schlussfolgerung: Patienten mit echoluzenten Plaques und höhergradiger Stenose tragen ein erhöhtes Risiko cerebraler Embolien während CAS. Die quantitative ultraschallgestützte Plaqueanalyse liefert Anhaltspunkte zur Identifizierung von Patienten, die ungeeignet für CAS sind.
Purpose: It remains to be determined whether the impact of plaque characteristics on procedural risks differs between carotid artery stenting (CAS) and endarterectomy (CEA). We studied whether quantitative assessment of carotid plaque echolucency on ultrasound predicts the risk of embolism during CAS or CEA. Materials and Methods: In 50 consecutive patients with symptomatic carotid stenosis randomized to CAS (n = 26) or CEA (n = 24) in the International Carotid Stenting Study (ICSS), semiautomated grayscale measurement of carotid plaques on baseline ultrasound was performed. We determined the grayscale median (GSM), percentage of echolucent plaque area, and a previously defined echographic risk index (ERI) calculated with the echolucent area and degree of stenosis. Brain MRI including diffusion-weighted imaging (DWI) was performed within 7 days before and 3 days after treatment. The primary outcome was the presence of at least 1 new hyperintense DWI lesion (DWI+) after treatment. Results: In the CAS group, DWI+ patients (n = 18) had a significantly higher ERI at baseline (mean 0.11 ± 0.12) than patients without new lesions (n = 8; mean 0.03 ± 0.01; p = 0.012). GSM (mean 26.7 ± 18.7 versus 34.3 ± 8.0, p = 0.16) and echolucent plaque area (mean 42.8 ± 21.1 versus 31.2 ± 8.2, p = 0.054) did not differ significantly. In the CEA group, there were no differences in plaque echogenity measurements between patients with (n = 2) and without DWI lesions (n = 22). Conclusion: Patients with echolucent plaques causing severe narrowing are at increased risk for cerebral embolism during CAS. Quantitative ultrasound plaque analysis, with ERI in particular, may add to clinical variables in identifying patients at risk for procedural stroke with CAS, but larger studies with clinical endpoints are needed.
● carotid stenosis ● plaque echolucency ● stenting ● endarterectomy ● embolism " " " "
received accepted
9.4.2013 3.9.2013
Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1355751 Published online: October 18, 2013 Ultraschall in Med 2014; 35: 267–272 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0172-4614
Correspondence Dr. med. Annika Burow Department of Neurology, University Hospital Basel Petersgraben 4 4031 Basel Switzerland Tel.: ++ 41/6 12 65 25 25 Fax: ++ 41/6 12 65 56 44 [email protected]
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Downloaded by: Universite Laval. Copyrighted material.
2
Original Article
Introduction !
About one in ten ischemic strokes is caused by atherosclerotic stenosis of the internal carotid artery. Carotid endarterectomy (CEA) reduces the risk of stroke in patients with symptomatic carotid stenosis [1, 2]. Carotid artery stenting (CAS) has emerged as a potential alternative to CEA, but is associated with an increased risk of peri-procedural stroke [3]. A pooled analysis of three trials showed that the excess stroke risk with CAS in comparison with CEA was mainly observed in patients above the age of 70 [4]. However, factors other than age may also determine the risk of peri-procedural stroke and aid in the selection of the appropriate treatment strategy. Instability of the atherosclerotic plaque plays an important role in the occurrence of ischemic cerebrovascular events associated with carotid stenosis [5, 6], both spontaneously and during treatment. Reduced plaque echogenicity on duplex ultrasound Bmode images (plaque echolucency) corresponds to histological markers of plaque instability [7, 8] and is associated with an increased risk of stroke [8 – 13]. More recently, an Echographic Risk Index (ERI) based on a combination of degree of stenosis and proportion of echolucent area at the plaque surface has been demonstrated to predict symptom status and presence of ipsilateral ischemic lesions on magnetic resonance imaging (MRI) [14]. Quantitative assessment of plaque ultrasound therefore has the potential to identify “unstable” plaques which are more likely to cause cerebrovascular events or subclinical embolization. However, plaque instability may also influence the risk of peri-procedural stroke during CAS or CEA. So far, the impact of ultrasound plaque morphology on the peri-procedural cerebral ischemia risk has not been compared between CAS and CEA in a randomized study. We investigated whether quantitative assessment of carotid plaque echolucency, alone or in conjunction with degree of stenosis, would predict the risk of cerebral ischemia on MRI in patients randomly assigned to CAS or CEA in the International Carotid Stenting Study (ICSS) [15].
Methods !
ICSS was a large randomized trial comparing the safety and efficacy of CAS versus CEA in 1710 patients with carotid stenosis of ≥ 50 % luminal narrowing and associated symptoms within the past 12 months [15]. In an MRI substudy nested within ICSS, 231 patients (CAS: n = 124, CEA: n = 107) enrolled at 7 ICSS centers received magnetic resonance imaging (MRI) scans of the brain 1 – 7 days before and 1 – 3 days after treatment, in order to assess silent ischemic and hemorrhagic brain lesions occurring peri-procedurally. Details of ICSS and the ICSS-MRI substudy have been reported in previous publications [15, 16]. The primary imaging endpoint of the ICSS-MRI substudy was cerebral ischemia, defined as at least one hyperintense signal alteration on diffusionweighted imaging (DWI) on the post-treatment MRI scan which was not present before treatment [16]. Baseline carotid imaging in ICSS required to identify carotid lesions suitable for both CEA and CAS included the combination of at least 2 noninvasive imaging techniques (magnetic resonance or computer tomography angiography, or duplex ultrasound) or intra-arterial angiography [17]. Degree of stenosis was measured according to the criteria used in the NASCET trial [1]. Two of the seven centers in the ICSS-MRI substudy participated in the
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
present study. We retrieved digitalized color-coded duplex ultrasound images of carotid stenosis on the relevant side, which were obtained before randomization in 53 patients enrolled at these two centers (27 patients in the CAS group, 26 patients in the CEA group). The study population consisted of 50 patients (CAS: n = 26, CEA: n = 24), in whom ultrasound images were suitable for both visual and semi-automated analysis. Ultrasound imaging was performed on a Siemens Acuson Sequoia 512 machine using a 7 – 5 MHz linear transducer and a Philips HDI machine using a 9 – 3 MHz linear transducer at one center; and on two Philips iU22 machines, one using a 8 – 4 MHz and one using a 9 – 3 MHz linear transducer at the other center. Visual analysis of plaque echolucency on B-mode images was performed offline by a single observer using the classification proposed by Gray-Weale [7]: type 1 (echolucent with echogenic fibrous cap), type 2 (predominantly echolucent, with echogenic areas representing less than 25 % of the plaque), type 3 (predominantly echogenic, with echolucent areas representing less than 25 % of the plaque), and type 4 (echogenic and homogenous plaque). Quantitative, semiautomated analysis of carotid plaques was performed offline by two researchers in consensus, as detailed below. Both researchers were experienced sonographers who were unaware of the allocated treatment, the clinical outcome, and whether patients had DWI lesions after treatment or not. The method of computer-aided plaque analysis used in the present study has been described in detail before [14, 18 – 20]. In brief, the base of the plaque was outlined manually at the adventitial border. The boundary between the color flow and the surface of the plaque was delineated automatically by the computer program. The grayscale values of the plaque were normalized by automatic linear scaling after outlining a black region in the perfused vessel lumen (grayscale value = 0) and a bright region in the vessel adventitia (grayscale value = 190). Each pixel of the carotid plaque was then mapped in 3 different colors: red (echolucent), yellow, and green (echogenic), according to its grayscale value. For the present analysis, we defined echolucent (red) plaque area as grayscale < 20, and echogenic (green) area as grayscale " Fig. 1b). > 50 (● The following parameters were then calculated by the computer program for the entire 2 D section of the plaque: (1) the grayscale median (GSM), i. e. the median value of all grayscale values; (2) the proportion of echolucent (red) plaque area as percentage of the whole plaque area; and (3) the Echographic Risk Index (ERI), which is based on echolucent plaque area and degree of stenosis (determined using standardized duplex flow-velocity criteria) according to the following formula [14]: exp (-8.98404 + 0.0458 × degree of stenosis + 0.06018 × proportion of echolucent area) ERI =
1 + exp (-8.98404 + 0.0458 × degree of stenosis + 0.06018 × proportion of echolucent area)
Statistical Analysis !
Statistical analysis was performed using IBM SPSS Version 20.0 software (Chicago, Illinois). Categorical variables were compared using chi-square tests and continuous variables using T-tests. Analyses were performed separately for the CAS and the CEA group. First, baseline plaque parameters were compared between patients with versus without new lesions after treatment. Second, exploratory post-hoc cross-table analyses were performed to study the association of any plaque parameter differing significantly between patients with versus without DWI lesion, sep-
Downloaded by: Universite Laval. Copyrighted material.
268
Original Article
Results !
Among the primary study population, risk factor profile and demographic baseline characteristics as well as plaque ultrasound parameters did not differ significantly between patients allocated " Table 1). 18 of 26 pato CAS and those who underwent CEA (● tients in the CAS group (69 %) and 2 of 24 patients in the CEA group (8 %) had new DWI lesions after treatment (p < 0.001). In the CAS group, there was no difference in whole-plaque GSM between patients with (mean 26.7 ± 18.7) and without lesions " Fig. 2). There was a trend towards a higher (34.3 ± 8.0, p = 0.16) (● proportion of echolucent plaque area in patients with new le-
Fig. 1 a Predominantly echolucent plaque with hyperechogenic areas. b Color mapping of the plaque shows the following proportions for the whole plaque: red 26 % (GSM < 20), yellow 32 % (GSM > 20 and < 50) and green 42 % (GSM > 50); for the surface one-third of the plaque the colors are distributed as follows: red 32 %, yellow 35 % and green 33 %, corresponding to a total grayscale median (GSM) value of 41.
age mean (SD) female gender n (%)
Abb. 1 a Vorwiegend echoluzente Plaque mit hyperechogenen Anteilen. b Das Farbmapping der Plaque ergibt folgende Werte: Rotanteil 26 % (GSM < 20), Gelbanteil 32 % (GSM > 20 and < 50) und Grünanteil 42 % (GSM > 50); für die Plaqueoberfläche (oberstes Drittel) ergeben sich folgende Farbverteilungen: Rot 32 %, gelb 35 % und grün 33 %, einem GSM von 41 entsprechend.
CAS (n = 26)
CEA (n = 24)
73.2 (9.32)
71.7 (8.25)
p-value ns
6 (23 %)
7 (29 %)
ns ns
Table 1
Patient characteristics.
vascular risk factors n (%) current smoking
6 (23 %)
2 (8 %)
ex-smoker
10 (38 %)
11 (46 %)
ns
hypercholersterolemia
14 (54 %)
12 (17 %)
ns
hypertension
20 (77 %)
20 (83 %)
ns
diabetes
6 (24 %)
6 (25 %)
ns
coronary heart disease
6 (23 %)
2 (83 %)
ns
peripheral artery disease
4 (15.4 %)
4 (16.7 %)
ns
145.77 (26.86)
148.71 (19.77)
ns
4.61 (1.06)
5.03 (1.41)
ns
systolic blood pressure at randomization [mmHg] mean (SD) total cholesterol at randomization [mmol/l] mean (SD) most recent ipsilateral event n (%) hemispheric ischemic stroke retinal stroke TIA
15 (57.7 %)
13 (54.2 %)
ns
6 (23.1 %)
3 (12.5 %)
ns
8 (33.3 %)
ns
79.04 (10.49)
81.16 (9.65)
ns
GSM
29.04 (16.38)
33.33 (15.45)
ns
echolucent area 1
39.23 (18.72)
36.25 (18.11)
ns
0.08 (0.10)
0.08 (0.10)
ns
mean degree of ipsilateral stenosis mean (SD)
5 (19.2 %)
quantitative plaque analysis (whole plaque)
ERI Grey-Weale plaque types n (%) type 1
1 (4.2 %)
type 2
13 (54.2 %)
14 (53.8 %)
0
type 3
9 (37.5 %)
12 (46.2 %)
type 4
1 (4.2 %)
0
CAS = carotid artery stenting; CEA = carotid endarterectomy; SD = standard deviation; TIA = transient ischemic attack; GSM = grayscale median; ERI = echographic risk index; ns = not significant. 1 Percentage of plaque area below 20th percentile of grayscale values.
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Downloaded by: Universite Laval. Copyrighted material.
arated at the median of the study population, and the presence of DWI lesions after treatment. Our study population was too small for multivariate adjustment. Instead, we chose to investigate for correlation of any plaque parameter which significantly predicted DWI lesions with age, which is the strongest known clinical predictor of peri-procedural stroke in CAS, using Pearson’s rank test. To account for multiplicity of comparisons of plaque characteristics between patients with and without DWI lesions (GSM, echolucent area, and ERI), we multiplied p-values by three (Bonferroni correction) for those comparisons significant at p < 0.05 without correction. A p-value < 0.05 after correction was considered to indicate statistical significance.
269
Original Article
34.3 (8.0)
80
CAS
80
26.72 (18.70)
DWI+ (n=18)
Echolucent area*
CAS 42.8 (21.1)
p= 0.054
60
DWI(n=22)
60
31.2 (8.2)
DWI+ (n=2)
CEA
80
Echolucent area*
DWI(n=8)
37.3 (18.2)
24.2 (16.8)
40
40
20
20
Fig. 2 Whole plaque GSM, echolucent plaque area and ERI mean (SD) in patients without lesions compared to patients with lesions in the CAS and CEA group. *Percentage of plaque area below 20th percentile of grayscale values. CAS = carotid artery stenting; CEA = carotid endarterectomy; GSM = grayscale median; ERI = echographic risk index; DWI = diffusion-weighted imaging; DWI− = patients without DWI lesion; DWI+ = presence of DWI lesion; SD = standard deviation. P-values are not corrected for multiple comparisons. Abb. 2 Plaqueparameter GSM gesamt, echoluzenter Plaqueanteil und echografischer Risikoindex (ERI) innerhalb der Patientengruppe ohne Läsionen im Vergleich zu Patienten mit Läsionen innerhalb der CAS- und CEA-Gruppen. *Graustufenwerte < 20. Perzentile. GSM = Graustufen Median; ERI = Echografischer Risikoindex; CAS = Stent-Angioplastie; CEA = Endarterektomie; DWI = diffusionsgewichtete Bildgebung; DWI− = Patienten ohne DWILäsion; DWI+ = Vorhandensein einer DWI-Läsion; SD = Standardabweichung.
0
0 DWI(n=8)
DWI+ (n=18)
DWI(n=22)
CAS p=0.012
CEA
80 0.11 (0.12)
60
ERI
60 40 0.03 (0.01)
40
DWI+ (n=2)
0.08 (0.10)
0.05 (0.05)
DWI(n=22)
DWI+ (n=2)
20 0
0 DWI(n=8)
DWI+ (n=18)
sions (mean 42.8 ± 21.1) than patients without new lesions (mean 31.2 ± 8.2) after treatment (p = 0.054). In the CAS group, patients with new DWI lesions after treatment had a significantly higher whole-plaque ERI (0.11 ± 0.12) than patients without new " Fig. 2). lesions (0.03 ± 0.01; p = 0.012) (● In the CEA group, there were no differences in any plaque echogenicity measures between patients with DWI lesions (n = 2) and those without DWI lesions (n = 22). However, due to the small number of CEA patients with new DWI lesions after treatment, no statistical tests were performed within the CEA group. There were no significant differences in degree of carotid stenosis between patients with versus without DWI lesions both in the CAS group (80.6 % ± 11.6 versus 75.6 % ± 6.8, p = 0.19) and in the CEA group (82 %.5 ± 17.7 versus 81.0 % ± 9.3). In order to further explore the relationship between whole-plaque ERI and risk of peri-procedural DWI lesions, we separated patients at the median ERI value of the entire study population " Fig. 3). All 11 CAS patients with ERI above the median (0.04) (● had new ischemic lesions after treatment (p = 0.07), compared with only 1 of 13 CEA patients with ERI above the median (n. s.). We did not find any correlation at all between whole-plaque ERI and patient age across the entire study population (p = 0.9)
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Fig. 3 Number of patients with and without new diffusion-weighted imaging (DWI) lesions after treatment below and above the median of the Echographic Risk Index (ERI). CAS = carotid artery stenting; CEA = carotid endarterectomy; ERI_low = Patients below the median of the ERI; ERI_high = Patients above the median of the ERI; DWI− = no DWI lesion; DWI+ = presence of DWI lesion. Abb. 3 Anzahl der Patienten mit und ohne DWI-Läsionen nach Behandlung unter- sowie oberhalb des Medianwertes des Echografischen Risikoindex (ERI). CAS = Stent-Angioplastie; CEA = Endarterektomie; ERI_low = Patienten unterhalb des Median des ERI; ERI_high = Patienten oberhalb des Median des ERI; DWI = diffusionsgewichtete Bildgebung; DWI− = keine DWI-Läsion; DWI+ = Vorhandensein einer DWI-Läsion.
Downloaded by: Universite Laval. Copyrighted material.
GSM
GSM
0
0
20
39.50 (6.4)
20
20
80
CEA
40
40
80
32.8 (16.0)
60
60
ERI
270
Original Article
Discussion !
In this study, we investigated whether semi-automated quantitative analysis of plaque echolucency, on its own or in combination with degree of stenosis, might predict the risk of peri-procedural cerebral ischemia in patients with symptomatic carotid stenosis randomly allocated to CAS or CEA. Among patients undergoing CAS, an increase in the Echographic Risk Index (ERI) based on echolucent plaque area and degree of stenosis was associated with an increased risk of cerebral ischemia. Echolucent plaque areas, i. e. areas appearing dark on B-mode ultrasound, correspond to necrotic areas with increased lipid content and plaque hemorrhage in histological examination [7, 21]. Computerized analysis of B-mode ultrasound images allows for normalization against reference tissues and therefore a quantitative and objective assessment of plaque echolucency. Plaque echolucency is traditionally expressed by the grayscale median (GSM), i. e. the median grayscale value in the carotid plaque; GSM inversely correlates with the presence of ipsilateral stroke on CT [8, 11] and risk of future stroke [8 – 11]. More recently, an ERI, based on a combination of degree of stenosis and proportion of echolucent area at the plaque surface, has been demonstrated to be superior to conventional parameters such as degree of stenosis and GSM in predicting symptom status and presence of ipsilateral ischemic lesions on MRI [14]. Previous studies have investigated the relationship between plaque morphology on ultrasound and embolic risk in CAS. In the ICAROS study, the risk of stroke during CAS was higher in patients who had plaques with GSM ≤ 25 than in those with GSM values > 25 [22]. We observed lower average GSM values among patients with new DWI lesions after treatment compared with patients without new lesions in the CAS group, but the difference was not statistically significant. However, ICAROS included both patients with symptomatic and asymptomatic carotid stenosis, for which reason the findings might not be directly comparable to our study which included symptomatic patients only. However, due to its larger sample size, the ICAROS study was more powerful to detect such a difference. A recent study prospectively investigated 31 consecutive patients with symptomatic severe carotid stenosis undergoing CAS with dual-frequency transcranial Doppler sonography (TCD). Patients with echolucent plaques defined by a GSM < 50 more often had solid cerebral microemboli on TCD than patients with echogenic plaques [23]. Our results, together with the findings of previous studies, lend support to the hypothesis that plaque instability may pose a greater risk of peri-procedural embolic stroke in CAS than in CEA. During CAS, where blood flow towards the brain is usually maintained, debris of unstable plaques may be released during catheterization or stent insertion, causing embolism to the brain. In contrast, the carotid artery is clamped off during CEA and there is no direct pathway for emboli to the brain. Cerebral protection devices (CPD) have been designed to prevent cerebral embolization during CAS but their efficacy has not been proven in randomized trials. Use of CPD was recommended, but not mandatory in ICSS. Filter-type CPDs were used in all but 2 patients treated with CAS in the present study population.
Other ultrasound techniques may be useful in selecting patients for CEA or CAS, as well. Plaque neovascularization identified by micro-bubble contrast has been shown to increase the risk of stroke in patients with carotid stenosis [24]. More recently, ultrasound elastography has emerged as a method to investigate mechanical tissue properties, in particular the reaction to shear stress [25, 26]. This technique may also potentially be used to characterize the carotid plaque on a mechanical level, which may add to information gained by imaging [27]. CAS and CEA may be complicated by cerebral embolism which often remains clinically silent [28]. In the ICSS-MRI substudy, new DWI lesions occurred in 50 % of patients treated with CAS and 17 % of patients undergoing CEA [16, 29]. Infarction on DWI has been proposed as a sensitive surrogate outcome measure for the risk of peri-procedural stroke [28]. In the present study, we used DWI lesions after treatment as a surrogate outcome measure for peri-procedural embolism. Our findings warrant confirmation in larger randomized trial populations whether quantitative assessment of plaque echolucency in combination with degree of stenosis predicts the risk of peri-procedural stroke.
Strengths and limitations: The major strength and a novelty of our work is that the allocation to the type of revascularization (CAS or CEA) was randomized. This allowed for a comparison of the effect of plaque characteristics on the risk of peri-procedural embolism between the two treatments. However, we are aware of several limitations of our study. Firstly, patients in the present study were investigated on different ultrasound machines using different transducers. Although echolucency was standardized by two reference tissues as described above, we cannot rule out that there were systematic differences in echolucency measured on different machines. Secondly, the duplex images were not initially obtained specifically for the purpose of quantitative analysis, which may have reduced quality in some cases. Thirdly, our study population comprised patients with high-risk symptomatic carotid stenosis enrolled in a randomized intervention trial, and therefore differed from the original study that reported the value of the ERI in differentiating between patients with symptomatic and asymptomatic stenosis [14]. This probably explains why our patients generally had plaques of lower echogenicity (GSM 29 in the CAS group and 33 in the CEA group) than the patients in the previous study (GSM 43). To be able to differentiate the plaque areas among our study population, we therefore decided to use grayscale cut-off values of < 20 and > 50 to define echolucent and echogenic areas, rather than < 60 and > 90 as in the previous publication. Fourthly, our sample size was relatively small. Thus, the risk of false-positive findings (e. g., the association between ERI and new DWI lesion for the CAS group) and the risk of false-negative findings (e. g. the absence of such an association for the CEA group) have to be acknowledged. Therefore, our findings should be interpreted cautiously and confirmation in a larger dataset is required. In conclusion, we found that quantitative analysis of plaque echolucency in combination with degree of stenosis on carotid ultrasound identified patients at increased risk for peri-procedural embolism with CAS. However, patients with a low echographic risk index still had a lower risk of cerebral ischemia with CEA than with CAS. Our findings warrant the application of quantitative ultrasound analysis in large clinical trials to investigate if this
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
Downloaded by: Universite Laval. Copyrighted material.
There were no differences with regard to Grey-Weale plaque type comparing patients with versus without DWI lesions, neither in the whole study population nor the CAS and CEA groups separately.
271
Original Article
technology adds to established clinical risk factors such as age, in identifying patients at risk for peri-procedural stroke with CAS.
Abbreviations CAS CEA DWI DWI− DWI+ ERI ERI_low
carotid artery stenting carotid endarterectomy diffusion-weighted imaging no DWI lesion presence of DWI lesion Echographic Risk Index patients below the median of the Echographic Risk Index ERI_high patients above the median of the Echographic Risk Index GSM grayscale median ICSS International Carotid Stenting Study ns not significant SD standard deviation
08
09
10
11
12
13
14
15
Acknowledgement !
ICSS was funded by grants from the Medical Research Council, the Stroke Association, Sanofi-Synthelabo, and the European Union. Funding for MRI scans performed as part of the ICSS-MRI study was provided by grants from the Mach-Gaensslen Foundation, Switzerland and the Stroke Association, United Kingdom. The work of Dr. Burow for this research was supported by the Stroke(Hirnschlag)-Fund Basel and by a grant from the University of Basel. Professor Brown's Chair in Stroke Medicine is supported by the Reta Lila Weston Trust for Medical Research. Dr. Bonati was supported by grants from the Swiss National Science Foundation (PBBSB-116873 and 33CM30-124119) and the University of Basel. This work was partly performed at University College London Hospital and University College London, which received a proportion of funding from the Department of Health National Institute for Health Research Biomedical Research Centres funding scheme.
References 01 Barnett HJ, Taylor DW, Eliasziw M et al. Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med 1998; 339: 1415 – 1425 02 European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: Final results of the MRC European Carotid Surgery Trial (ECST). The Lancet 1998; 351: 1379 – 1387 03 Bonati LH, Lyrer P, Ederle J et al. Percutaneous transluminal balloon angioplasty and stenting for carotid artery stenosis. Cochrane Database Syst Rev 2012, CD000515 9 04 Bonati LH, Dobson J, Algra A et al. Short-term outcome after stenting versus endarterectomy for symptomatic carotid stenosis: a preplanned meta-analysis of individual patient data. Lancet 2010; 376: 1062 – 1073 05 Aburahma AF, Thiele SP, Wulu JT et al. Prospective controlled study of the natural history of asymptomatic 60% to 69% carotid stenosis according to ultrasonic plaque morphology. J Vasc Surg 2002; 36: 437 – 442 06 Eyding J, Geier B, Staub D. Current strategies and possible perspectives of ultrasonic risk stratification of ischemic stroke in internal carotid artery disease. Ultraschall in Med 2011; 32: 267 – 273 07 Gray-Weale AC, Graham JC, Burnett JR et al. Carotid artery atheroma: comparison of preoperative B-mode ultrasound appearance with car-
Burow A et al. Echographic Risk Index … Ultraschall in Med 2014; 35: 267–272
16
17
18
19
20
21
22
23
24 25
26
27
28
29
otid endarterectomy specimen pathology. J Cardiovasc Surg (Torino) 1988; 29: 676 – 681 Biasi GM, Sampaolo A, Mingazzini P et al. Computer analysis of ultrasonic plaque echolucency in identifying high risk carotid bifurcation lesions. Eur J Vasc Endovasc Surg 1999; 17: 476 – 479 Sabetai MM, Tegos TJ, Clifford C et al. Carotid plaque echogenicity and types of silent CT-brain infarcts. Is there an association in patients with asymptomatic carotid stenosis? Int Angiol 2001; 20: 51 – 57 Gronholdt ML, Nordestgaard BG, Schroeder TV et al. Ultrasonic echolucent carotid plaques predict future strokes. Circulation 2001; 104: 68 – 73 el-Barghouty N, Geroulakos G, Nicolaides A et al. Computer-assisted carotid plaque characterisation. Eur J Vasc Endovasc Surg 1995; 9: 389 – 393 Tegos TJ, Sabetai MM, Nicolaides AN et al. Comparability of the ultrasonic tissue characteristics of carotid plaques. J Ultrasound Med 2000; 19: 399 – 407 Mathiesen EB, Bonaa KH, Joakimsen O. Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: the tromso study. Circulation 2001; 103: 2171 – 2175 Momjian-Mayor I. Accuracy of a Novel Risk index Combining Degree of Stenosis of the Carotid Artery and Plaque Surface Echogenicity. Stroke 2012 Ederle J, Dobson J, Featherstone RL et al. Carotid artery stenting compared with endarterectomy in patients with symptomatic carotid stenosis (International Carotid Stenting Study): an interim analysis of a randomised controlled trial. Lancet 2010; 375: 985 – 997 Bonati LH, Jongen LM, Haller S et al. New ischaemic brain lesions on MRI after stenting or endarterectomy for symptomatic carotid stenosis: a substudy of the International Carotid Stenting Study (ICSS). Lancet Neurol 2010; 9: 353 – 362 Featherstone RL, Brown MM, Coward LJ. International carotid stenting study: protocol for a randomised clinical trial comparing carotid stenting with endarterectomy in symptomatic carotid artery stenosis. Cerebrovasc Dis 2004; 18: 69 – 74 Sztajzel R, Momjian S, Momjian-Mayor I et al. Stratified gray-scale median analysis and color mapping of the carotid plaque: correlation with endarterectomy specimen histology of 28 patients. Stroke 2005; 36: 741 – 745 Sztajzel R, Momjian-Mayor I, Comelli M et al. Correlation of cerebrovascular symptoms and microembolic signals with the stratified grayscale median analysis and color mapping of the carotid plaque. Stroke 2006; 37: 824 – 829 Momjian I, Momjian S, Albanese S et al. Visual analysis or semi-automated gray-scale-based color mapping of the carotid plaque: which method correlates the best with the presence of cerebrovascular symptoms and/or lesions on MRI? J Neuroimaging 2009; 19: 119 – 126 Carotid artery plaque composition–relationship to clinical presentation and ultrasound B-mode imaging. European Carotid Plaque Study Group. Eur J Vasc Endovasc Surg 1995; 10: 23 – 30 Biasi GM, Froio A, Diethrich EB et al. Carotid plaque echolucency increases the risk of stroke in carotid stenting: the Imaging in Carotid Angioplasty and Risk of Stroke (ICAROS) study. Circulation 2004; 110: 756 – 762 Rosenkranz M, Russjan A, Goebell E et al. Carotid plaque surface irregularity predicts cerebral embolism during carotid artery stenting. Cerebrovasc Dis 2011; 32: 163 – 169 Staub D, Partovi S, Imfeld S et al. Novel applications of contrast-enhanced ultrasound imaging in vascular medicine. Vasa 2013; 42: 17 – 31 Bamber J, Cosgrove D, Dietrich CF et al. EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 1: Basic principles and technology. Ultraschall in Med 2013; 34: 169 – 184 Cosgrove D, Piscaglia F, Bamber J et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography.Part 2: Clinical Applications. Ultraschall in Med 2013; 34: 238 – 253 Maurice RL, Soulez G, Giroux MF et al. Noninvasive vascular elastography for carotid artery characterization on subjects without previous history of atherosclerosis. Med Phys 2008; 35: 3436 – 3443 Schnaudigel S, Groschel K, Pilgram SM et al. New brain lesions after carotid stenting versus carotid endarterectomy: a systematic review of the literature. Stroke 2008; 39: 1911 – 1919 Gensicke H, Zumbrunn T, Jongen LM et al. Characteristics of ischemic brain lesions after stenting or endarterectomy for symptomatic carotid artery stenosis: results from the international carotid stenting studymagnetic resonance imaging substudy. Stroke 2013; 44: 80 – 86
Downloaded by: Universite Laval. Copyrighted material.
272
