Keywords Ischemic Stroke - Ischemic Attack Transient - Intracranial Arterial Diseases - Ultrasonography
Doppler Transcranial
Palavras-chave AVC Isquêmico - Ataque Isquêmico Transitório - Doenças Arteriais Intracranianas -
Ultrassonografia Doppler Transcraniana
INTRODUCTION
Intracranial arterial stenosis (ICAS) is still considered the most significant cause
of acute ischemic stroke (AIS) worldwide.[1 ] However, its prevalence in Latin America is largely unknown and variable in different
study methodologies and scenarios.[2 ] In Brazil, data about ICAS frequency within the AIS and TIA populations is insufficient.
Patients with ICAS, particularly those with symptomatic stenosis, have a higher risk
of recurrent stroke, especially during the first year after the initial stroke.[3 ] Therefore, it is essential to identify these symptoms promptly in AIS or TIA patients.
Furthermore, patients with more than 50% luminal stenosis should be followed closely
by aggressive secondary stroke prevention strategies.[4 ]
Techniques such as digital subtraction (DSA), computed tomographic (CTA), and magnetic
resonance (MRA) angiographies are commonly used in developed countries for the diagnosis
of ICAS in AIS/TIA patients, but they have limitations such as invasiveness, high
cost, contrast toxicity, and high-magnetic field restrictions. Moreover, the low availability
and high cost of these modalities in developing countries may also limit their use
as routine workup of ICAS as part of the diagnostic assessment of AIS patients, contributing
to underestimating the burden of this condition in Brazil.[5 ] Alternatively, among neuroimaging methods, transcranial color-coded duplex sonography
(TCCS) study is a noninvasive, broadly available, and inexpensive method for assessing
intracranial blood flow and hemodynamic changes.[6 ]
[7 ] It is mainly used for the detection of hemodynamically significant ICAS in acute
stroke patients. This study aimed to evaluate the frequency and severity of ICAS in
patients with AIS and TIA admitted to an academic tertiary stroke center in Brazil
using TCCS.
METHODS
Population and inclusion criteria
The study population consisted of all consecutive patients above 18-years-old who
were admitted with AIS or TIA and evaluated by TCCS as part of their routine stroke
etiologic investigation, selected prospectively from February 2014 to December 2014.
During the study period, investigation of the intracranial vasculature using an imaging
technique (TCCS or CTA) as part of the routine investigation of all patients suspected
of having a stroke. The ethical committee of the Hospital das Clínicas from the Ribeirão
Preto Medical School approved this study, and written informed consent was required
from all participants. Patients without TCCS in their routine investigation, or those
with bilateral inadequate transtemporal acoustic bone window, defined as the absence
of visualization of the midbrain and sphenoid wing bone by TCCS on B-mode, as well
as those in whom no vascular structure could be identified, were excluded from the
study.[8 ]
[9 ]
Study protocol
The diagnosis of AIS or TIA was performed by an experienced stroke neurologist based
on the patients' clinical history, neurological examination, and neuroimaging results.
Symptomatic ICAS was defined as the narrowed artery being responsible for new symptoms
or ischemia in the corresponding brain region.[10 ]
The data collected included demographic information, clinical history, and diagnostic
workup. On admission, the national institutes of health stroke scale (NIHSS) was used
to assess stroke severity, and the modified Rankin scale (mRS) was used to assess
functional outcome at discharge and 90 days after the cerebrovascular event.[11 ]
[12 ]
[13 ]
The personal history of main risk factors for stroke was collected: systemic arterial
hypertension, diabetes mellitus, smoking, alcoholism, dyslipidemia, elevated body
mass index, atrial fibrillation, and prior occurrence of myocardial infarction, cerebral
infarction, and TIA. Blood pressure, capillary glycemia, glycosylated hemoglobin,
hemoglobin, hematocrit, and cholesterol levels were also recorded on admission. The
definition of the presence of each risk factor was based on the guidelines in force
at the time of data collection, as well as on the results of laboratory and imaging
tests in the study protocol.
Complementary exams were performed to determine the AIS subtype according to the Trial
of Org 10172 in Acute Stroke Treatment (TOAST).[14 ]
Imaging evaluation
Transcranial color-coded duplex sonography (TCCS) study
A TCCS study was conducted to investigate the presence and severity of ICAS in each
target vessel, indicated below. All exams were performed by just one experienced neurosonographer
within seven days of hospital admission using a Xario SSA-660A (Toshiba Corp. Tokio,
Japan) 2MHz probe.
The study measured the maximum mean flow velocities (MFV) of the following arteries
bilaterally: terminal internal carotid artery (TICA), proximal segments (M1 and M2)
of the middle cerebral artery (MCA), the anterior cerebral artery (ACA), and the P1
and P2 segments of the posterior cerebral artery (PCA). The vertebral (VA) and basilar
(BA) arteries were assessed through suboccipital window.[15 ]
According to the institutional protocol data were recorded at 0° and corrected by
aligning the insonation angle parallel to the blood flow vector whenever necessary.
Ultrasound investigation of cervical carotid and vertebral arteries was also performed
through TCCS. If significant stenosis (70%) was found, the patient was excluded from
the study as it could interfere with the analysis.
Intracranial arterial stenosis (ICAS) criteria
The following MFV cutoffs on TCCS were used for identification of hemodynamically
significant stenosis (≥50% luminal stenosis) according to the previous criteria, established
through TCCS ultrasound examination:[16 ]
[17 ] MCA-MFV > 100 cm/s, TICA MFV > 90 cm/s, VA and BA MFV > 80 cm/s, ACA > 80 cm/s, and PCA >50 cm/s. When the absolute value of velocity did not achieve the
criterion, the degree of stenosis was estimated as (1–[MV pre- or post-stenotic /
MV intrastenotic] x 100).
Having met the prerequisite of adequate quality of the ultrasound window, we defined
occlusion by absent or minimal signals and sub occlusion by blunted signals. We checked
all vessels every 2mm of their path, as the possibility of tortuosity could make the
examination difficult and simulate occlusion. We also excluded patients with significant
cervical ICAS.
In order to confirm the degree of stenosis by TCCS, we compared their findings with
CTA, if this test was available, with exams being performed using a Sensation 64 machine
(Siemens AG, Erlangen, Germany) and analyzed by experienced radiologist, blinded to
the TCCS results, using the Osirix (Bernex, Switzerland) v.6.0.2 imaging software.
Statistical analysis
All analyses were conducted using the Statistical Package Social Sciences (SPSS, IBM
Corp., Armonk, NY, USA) version 20.0. Univariate analysis was performed using the
Student t, Mann-Whitney U, Chi-squared, or Fisher exact test, as appropriate. Variables
with a p -value of less than 0.1 were included in the multivariate analysis. Multivariate logistic
regression (backward selection method) was then used to identify independent predictors
of stenosis. A p -value of less than 0.05 (2-sided) was considered statistically significant. We evaluated
the Cohen Kappa, sensitivity, specificity, and positive and negative predictive values
to compare the performance of TCCS with CTA.
RESULTS
We examined 170 subjects, of which 27 (15.9%) were excluded from the study because
of inadequate acoustic transtemporal bone windows for TCCS. These patients were similar
in baseline characteristics to those with adequate transcranial windows. Of the 143
remaining patients, 89 (62.2%) were men, and 62.9% were White. The mean age was 63.6 ± 11.1
years. Furthermore, 124 (86.7%) participants had AIS, and 19 (13.3%) had TIA. The
sample demographic and clinical details are displayed in [Table 1 ].
Table 1
Baseline characteristics of patients according to TCCS findings
Characteristics
Total n = 143
ICAS ≥ 50%
p- value*
No = 88 (61.5%)
Yes = 55 (38.5%)
Age, median ± SD*
63.57 ± 11.13
63.15 ± 11.14
64.25 ± 11.18
0.46
Male sex (%)
89 (62.2)
50 (56.8)
39 (70.9)
0.09
Ethnicity#
White
90 (62.9)
52 (59.1)
38 (69.1)
0.44
Risk Factors
Hypertension (%)
110 (76.9)
68 (77.3)
42 (76.4)
0.90
Diabetes (%)
58 (40.6)
38 (43.2)
20 (36.4)
0.41
Current smoker (%)
47 (32.9)
29 (32.9)
18 (32.7)
0.97
Prior stroke or TIA (%)
63 (44.1)
39 (44.3)
24 (43.6)
0.93
Cardiomyopathy (%)
32 (22.4)
19 (21.6)
13 (23.6)
0.77
Atrial fibrillation (%)
27 (18.9)
21 (23.9)
10 (18.2)
0.42
Chagas disease (%)
13 (9.1)
10 (11.4)
3 (5.5)
0.23
Dyslipidemia (%)
56 (39.2)
37 (42.1)
19 (34.5)
0.37
Alcoholism
45 (31.5)
24 (27.3)
21 (38.2)
0.17
Prior aspirin use (%)
59 (41.3)
33 (37.5)
26 (47.3)
0.24
Prior statins use in dyslipidemia (%)
24 (16.8)
15 (17.0)
9 (16.4)
0.62
NIHSS, median (IQR)*
7 (3–15)
6 (3–13)
10 (4–19)
0.004
Glasgow, median (IQR)*
15 (12–15)
15 (13.5–15)
14 (11–15)
0.08
Event type
AIS (%)
124 (86.7)
77 (87.5)
47 (85.5)
TIA (%)
19 (13.3)
11 (12.5)
8 (14.5)
0.55
Non-lacunar#
105 (74.5)
61 (70.1)
44 (81.5)
0.13
BMI median (IQR)*
26.7 (23.9–30.5)
26.4 (23.4–29.5)
27.1 (24.5–31.2)
0.27
Baseline systolic BP, median (IQR)*
150 (132–164)
140 (130–155)
160 (145–170)
< 0.001
Baseline diastolic BP, median (IQR)*
90 (70.5–93.5)
90 (77.5–91.0)
90 (80–95)
0.29
NIHSS 90 days, median (IQR)*#
3 (0–6)
3 (0–5)
3 (0–8)
0.87
mRS 90 days ≥ 3 (%)#
59 (47.2)
35 (45.5)
24 (50.0)
0.62
Mortality < 3 months (%)#
14 (11.2)
6 (7.8)
8 (16.7)
0.57
TOAST
Undetermined
40 (27.9)
34 (38.6)
6 (10.9)
Cardioembolism
41 (28.7)
27 (30.7)
14 (25.5)
Large artery atherosclerosis
41 (28.7)
13 (14.7)
28 (50.9)
< 0.0001
Small artery occlusion
17 (11.9)
11 (12.5)
6 (10.9)
Other determined causes
4 (2.8)
3 (3.4)
1 (1.8)
Abbreviations: AIS, acute ischemic stroke; BMI, body mass index; BP, blood pressure;
GCS, Glasgow Coma Scale; IQR, interquartile range; mRS, modified Rankin scale; NIHSS,
National Institutes of Health Stroke Scale; SD, standard deviation; TIA, transient
ischemic attack; TOAST, Trial of Org 10172 in Acute Stroke Treatment. Notes: *Data
expressed as median and interquartile range; # missing data were removed from the
analysis.
The TCCS detected ICAS in 55 (38.5%) patients, and 36 (65.5%) were considered symptomatic.
Male sex (p = 0.09), NIHSS (p = 0.004), systolic blood pressure on admission (p < 0.001), TOAST classification (p < 0.0001), and high-density lipoprotein (HDL) cholesterol (p = 0.02) had an association ([Tables 1 ] and [2 ]). Increased systolic blood pressure on admission was the only factor independently
associated with ICAS in the final multivariate logistic regression models (OR: 1.02;
95% confidence interval [CI]: 1.01–1.05; p = 0.008), as shown in [Table 3 ].
Table 2
Laboratorial baseline characteristics of patients according to TCCS findings
Characteristics; N (IQR)
Total N = 143
ICAS ≥50%
p- value
No = 88 (61.5%)
Yes = 55 (38.5%)
HbA1C (%)
6.2 (5.6–7.5)
6.1 (5.6–7.2)
6.4 (5.6–8)
0.30
HbA1C ≥6.5 (%)*
40 (44.9)
24 (42.8)
16 (48.5)
0.60
Hemoglobin (g/dL)
14.2 (12.8–15.3)
14.1 (12.9–15.5)
14.4 (12.6–15.1)
0.95
Hematocrit (%)
43 (39–46)
42 (39–46)
43 (38–45)
0.66
Total cholesterol (mg/dL)
173 (144–208)
176 (45–206)
173 (138–208)
0.54
LDL cholesterol (mg/dL)
110 (83–144)
111 (84–144)
102 (83–136)
0.46
HDL cholesterol (mg/dL)
35 (30–43)
36 (30–45)
32 (27–39)
0.02
Triglycerides (mg/dL)
120 (87–165)
119.5 (85–159)
127 (100–168)
0.57
Abbreviations: HbA1C, glycosylated hemoglobin; HDL, high-density lipoprotein; ICA,
intracranial arterial stenosis; IQR, interquartile range LDL, low-density lipoprotein.
Note: *values show the number and percentage of patients, missing values were excluded.
Table 3
Association with ICAS in the multivariate analysis
Stenosis ≥ 50%
OR
(95% CI)
p -value
No = 88 (61.5%)
Yes = 55 (38.5%)
SBP
(IQR)
140
(130–155)
160
(145–170)
1.02
(1.01–1.05)
0.008
Abbreviations: CI, confidence interval; ICAS, intracranial arterial stenosis; IQR,
interquartile range; OR, odds ratio adjusted; SBP, systolic blood pressure. Note:
The variables considered in the model are male sex, NIHSS, Glasgow, systolic blood
pressure, and HDL cholesterol.
The frequency of ICAS according to the different locations in proximal arteries is
given in [Table 4 ]. The number of ICAS vessels was one in 25 patients (45.5%), two in 17 (30.9%), three
in 5 (9.1%), four in 6 (10.9%), and five in 2 patients (3.6%). Overall, ICAS was most
commonly located in the anterior circulation (n = 61, 55.5%), especially in the left
M1 segment (15.5%) and in the right M1 segment (12.7%).
Table 4
Site of ICAS ≥ 50%
Location
≥ 50%
(N = 54)
Subocclusion/occlusion (N = 56)
Total
Right – TICA
7
2
9 (8.2%)
Right – M1CA
6
8
14 (12.7%)
Right – M2CA
0
5
5 (4.5%)
Right – ACA
4
1
5 (4.5%)
Right – P1CA
7
2
9 (8.2%)
Right – P2CA
2
1
3 (2.7%)
Right – VA
0
9
9 (8.2%)
BA
2
7
9 (8.2%)
Left – TICA
3
1
4 (3.6%)
Left – M1MCA
9
8
17 (15.5%)
Left – M2MCA
3
2
5 (4.5%)
Left – ACA
1
1
2 (1.8%)
Left – P1CA
5
1
6 (5.5%)
Left – P2CA
4
1
5 (4.5%)
Left – VA
1
7
8 (7.3%)
Abbreviations: ACA, anterior cerebral artery; BA, basilar artery; ICAS, intracranial
arterial stenosis; M1CA, M1 segment of the middle cerebral artery; M2CA, M2 segment
of the middle cerebral artery; P1CA, P1 segment of the posterior cerebral artery;
P2CA, P2 segment of the posterior cerebral artery; TICA, terminal internal carotid
artery; VA, vertebral artery.
The TCCS showed a good performance in predicting ICAS compared with CTA (Coehn kappa:
0.58; area under the curve [AUC]: 0.88; 95% CI 0.69–0.92; p < 0.0001) with 74% of sensitivity, 87% of specificity, 65% of predictive positive
value, and 90% of predictive negative value ([Figure 1 ] and [Table 5 ]).
Figure 1 (A ) Intracranial stenosis in the M1 segment of the left middle cerebral artery detected
by TCCS (mean flow velocity of 169.1 cm/s). (B ) Confirmation by computed tomography angiography (white arrow).
Table 5
Performance of TCCS compared with CTA in MCA ≥ 50% of 100 patients
MCA
95%CI
AUC
0.88
0.69–0.92
Sensitivity
0.74
0.53–0.87
Specificity
0.87
0.76–0.93
PPV
0.65
0.49–0.78
NPV
0.90
0.83–0.95
LR+
5.58
2.9–0.75
LR-
0.30
0.15–0.60
Abbreviations: AUC, area under the curve; CI, confidence interval; CTA, computed tomographic
angiography; LR + , positive likelihood ratio; LR-, negative likelihood ratio; MCA,
middle cerebral artery; NPV, negative predictive value; PPV, positive predictive value;
TCCS, transcranial color-coded duplex sonography.
DISCUSSION
This study performed at a public tertiary academic stroke center in Brazil found a
high frequency (38.5%) of ICAS compared with other Western studies, which indicated
8 to 10% of occurrence.[18 ]
[19 ] To the best of our knowledge, no previous study evaluating the frequency and severity
of ICAS assessed by TCCS has been provided in Brazil, and limited studies have investigated
the prevalence of this condition in Brazilian patients by other methods. Despite 15.9%
of inadequate transtemporal echo-window, we were able to show high rates of ICAS with
consecutive enrollment of acute stroke patients evaluated by the same methodology.
We believe that one of the reasons for the high rates found in our study may be the
systematic investigation of the intracranial vasculature.
It is important to mention that none of our patients met clinical and/or radiological
suspicion criteria for vasculitis, inflammatory arteriopathies, intracranial dissection,
or reversible cerebral vasoconstriction syndrome. We also excluded from the sample
patients who presented clinical and laboratory criteria for anemia, which allows us
to assume that the sample analyzed possibly corresponds to the intracranial atherosclerotic
disease.
Stroke is a heterogeneous disease, and intracranial atherosclerosis is considered
one of the most common causes of AIS worldwide.[20 ]
[21 ] Its incidence and prevalence vary between geographical regions and ethnicities,
even in Asian countries, where it accounts for about 50% of cases.[22 ]
[23 ] In Western countries, ICAS has historically been poorly studied and may be underestimated
compared to extracranial lesions.[18 ]
[24 ]
[25 ] In 1995, Sacco et al. found 8% of intracranial atherosclerosis in a stroke population
of 438 patients (35% Black, 46% Hispanic, and 19% White). The rate of extracranial
atherosclerosis was similar between ethnic groups; however, intracranial lesions were
more frequent in Blacks and Hispanics (OR: 7.8; 95% CI: 1.04–57.7).[19 ] In the present study, there wasn't a significant difference between ethnic groups
(p = 0.44). However, 69.1% of our patients with ICA were White.
Another reason for the high rates of intracranial atherosclerosis is the poor control
of classic risk factors for cerebrovascular diseases found in our sample. Systemic
arterial hypertension was the most frequent risk factor in our patients, with a prevalence
rate of 76.9%, followed by prior stroke or TIA (44.1%), diabetes (40.6%), and dyslipidemia
(39.2%). A study conducted on stroke patients detected that those with MCA stenosis
had a higher prevalence of hypertension, hypercholesterolemia, and diabetes.[26 ] Several other studies have also indicated a correlation between high blood pressure
and ICAS.[27 ]
[28 ]
[29 ]
In our study, systolic blood pressure on admission was significantly higher in patients
with ICAS (p < 0.001) and it was the only independently associated risk factor found in the multivariate
analysis. One of the pathophysiological explanations for this might be a preexistent
history of systemic arterial hypertension.
This data is consistent with other studies, including stroke-free populations with
intracranial atherosclerosis.[30 ]
[31 ]
[32 ] A Chinese study performed between 2009 and 2013 also showed a strong correlation
between high blood pressure and symptomatic atherosclerosis of the internal carotid
artery (OR: 5.98; 95% CI: 1.79–19.98) and other asymptomatic intracranial atherosclerosis
(OR: 2.56; 95% CI: 1.22–5.37).[33 ]
Of the 56 (39.2%) previously dyslipidemic patients, only 24 (42.8%) were taking statins.
The correct use of statins was found in 15 (17.0%) patients without ICAS and in only
9 (16.4%) of those with diagnosis, demonstrating a failure to control one of the most
critical risk factors for atherosclerosis and stroke.
Concerning stroke severity, patients with ICAS had higher NIHSS scores (p = 0.004) on admission, which may demonstrate a trend to more severe strokes in these
circumstances, similar to previous studies.[34 ] Nevertheless, this data was not statistically significant, nor was associated with
functional outcomes and mortality at 3 months. A study conducted by Lau et al. with
39% of ICAS prevalence checked by CTA, the patients had more severe strokes (median
NIHSS: 9 vs. 3; p < 0.001), worse outcomes at 6 months (mRS: 0–2; 57 vs. 73%; p < 0.001), and higher mortality (18 vs. 8%; p = 0.001).[35 ]
Many previous studies reported an association between low levels of HDL and the development
of intracranial atherosclerosis.[36 ] In an analysis of the CICAS study, low HDL levels are strongly associated with the
development of ICAS with an inverse relationship between both (p = 0.001).[37 ] In our study, the level of HDL was significantly lower in patients without ICAS
than in those with (p = 0.02).
Our study has some limitations, mainly because the TCCS is an operator-dependent technique
and can sometimes result in incomplete assessments due to unsatisfactory transtemporal
bone windows. However, it offers a noninvasive option that can be easily and quickly
repeatable at the bedside, allowing accurate evaluation of cerebral blood flow and
providing information about anatomical structures.[6 ]
[15 ]
[38 ] Furthermore, we performed the study in a single center, and the results were limited
to patients with AIS or TIA. Despite previously mentioned limitations, it is essential
to note that all selected patients were consecutively evaluated by TCCS, and the characteristics
of patients with and without transcranial windows were similar, reducing the possibility
of a relevant selection bias. Another possible limitation was that we could not perform
CTA on all patients due to contraindications. Nevertheless, our results support the
application of TCCS as a diagnostic strategy in Brazilian stroke centers, particularly
those lacking more modern imaging techniques.
In conclusion, our study shows that ICAS is a common condition in our population,
identified in about one third of patients admitted with AIS or TIA. Most of these
ICAS were symptomatic and considered to be culprits for the stroke episodes. Due to
the high prevalence of significant ICAS in the Brazilian population, it is believed
that inclusion of TCCS is essential as a bedside examination to evaluate stroke etiology,
particularly in patients with high blood pressure on admission. These preliminary
findings support further collaborative initiatives among stroke physicians to increase
ICAS detection in Brazilian patients with AIS or TIA.
Bibliographical Record
