Outcomes of Symptomatic Intracranial Large Artery Stenoses: A Prospective Cohort Study from the Asian Registry of Intracranial Atherosclerosis Open Access

The etiology of stroke is attributed to intracranial atherosclerosis (ICAS) when there is ≥50% diameter reduction and luminal narrowing of the supplying large intracranial artery proximal to the infarct without any other explanatory cause for the stroke [1]. The distribution of cerebral atherosclerosis in ischemic stroke patients is different in different ethnic groups. In Caucasians, ischemic stroke patients are found to have more extracranial carotid atherosclerosis, while in Asians, Africans and Hispanics, the majority of atherosclerotic strokes are caused by ICAS [2, 3]. According to stroke registry studies, ICAS is responsible for 8–16% of ischemic strokes in Caucasians [4‒8], while in Asians, up to 25–65% of ischemic strokes are caused by ICAS [4, 9‒16]. Among South Asian patients with acute ischemic stroke, ICAS is found in 54% of all stroke subtypes [17]. In Asian countries such as China and India where ICAS is known to be prevalent, a disproportionately high rate of stroke burden and mortality has been documented [18]. ICAS is associated with a high risk of recurrent stroke [19]

Most of the published literature on ICAS has been based on single center studies, or multicenter studies within a country. There is a paucity of international, multicenter, collaborative studies evaluating the outcome of acute ischemic stroke due to ICAS, despite its high prevalence in Asian countries. This multicenter international collaborative study aims to study the demographic features, vascular risk factors, 1-year recurrence and all-cause mortality among patients with stroke due to ICAS in Asia.

During the study period 1 July 2014 to 31 June 2019, consecutive patients with ischemic stroke were recruited over a 1-year period from one institution from each of six member countries of the Asian Stroke Advisory Panel (India, Malaysia, Philippines, Singapore, Taiwan, and Thailand). Each institution evaluated patients prospectively who were diagnosed with acute ischemic stroke and had undergone neuroimaging, either using cranial CT scan, magnetic resonance imaging (MRI) or both, to confirm the diagnosis of ischemic stroke. Subsequently, these patients underwent vascular imaging such as magnetic resonance angiography (MRA), CT angiography (CTA), digital subtraction angiography (DSA), or transcranial Doppler (TCD) examination for the diagnosis of intracranial stenosis. Other work-up included full blood count, glucose and lipids, electrocardiogram, echocardiogram and cardiac monitoring.

The presence of intracranial stenosis by MRA, CTA, or DSA was assessed following the method from WASID study [20]. Briefly, stenosis was quantified following the formula: $%stenosis=1‐Dstenosis/Dnormal×100$⁠, where Dstenosis is the diameter of the artery at the site of the segment with severe stenosis, D_normal is the diameter of the proximal normal artery, or if the proximal cannot be used for measurement the distal artery of the feeding artery may be used [20].

The diagnosis of intracranial stenosis by TCD was done using the following criteria: any focal or diffuse increase in the mean flow velocity of the insonated basal arteries with the following cut-off for the various vessels: middle cerebral artery (MCA) >100 cm/s, anterior cerebral artery >80 cm/s, internal carotid artery (ICA)/siphon >70 cm/s, vertebrobasilar artery >60 cm/s. Additionally, an interhemispheric difference of >30% or stenosis-to-pre-stenosis ratio of >2, disturbance of flow such as bruit and waveform changes of the Doppler spectrum were also utilized as supportive for ICAS diagnosis [21].

Study inclusion criteria were: (1) clinical diagnosis of stroke – rapidly developing clinical signs of focal (at times global) disturbance of cerebral function lasting more than 24 h or leading to death with no apparent cause other than that of vascular origin [22]. Stroke was also diagnosed among those with focal or global neurological symptoms lasting <24 h if there was imaging evidence of brain tissue injury due to a vascular cause [23]. (2) Admitted within 7 days of stroke symptom onset. (3) Had undergone computed tomography (CT) or MRI of the brain. (4) Stroke mechanism due to large artery atherosclerosis by TOAST criteria [24] intracranially without significant stenosis (>50%) of the extracranial artery. Patients were excluded if they had co-morbid illnesses other than stroke that would likely lead to death within a year, pre-stroke mRS >2, or if they did not consent to participate.

The demographic data of the recruited patients were collected, such as age, sex and ethnicity. Vascular risk factors (hypertension, diabetes mellitus, hyperlipidemia, ischemic heart disease, prior stroke) were determined based on history of having been previously diagnosed with the factor or on medication for it, and patient medical records. Smokers were those who were smokers at the time of stroke. Data were also collected on location of infarct, National Institute of Health Stroke Scale (NIHSS) on admission. Neuroimaging performed to confirm the presence of infarction. Data on neurovascular imaging were also obtained.

All patients diagnosed to have ICAS were followed up at 3, 6, 9, and 12 months poststroke. Management was at the discretion of the treating stroke neurologist. The outcome of interest was stroke recurrence based on clinical features and neuroimaging, and all-cause mortality at 1 year. The required information was obtained by the treating neurologist from the patients, family, and medical records. In cases of missing data, the last observation was carried forward.

Normally distributed continuous data were described by means and standard deviations, while proportions were used for categorical data. To test for associations with outcome, unpaired T test was used for continuous variables, and χ² for categorical variables. Clinically relevant variables were entered into a logistic regression to determine factors significantly associated with the outcomes of interest of this study – stroke recurrence and mortality at 1 year.

Using an estimated stroke recurrence of 15% at 1 year based on the WASID trial [20] with a confidence level of 90% that the real value is within 14.5%–15.5%, a sample size of 319 would be needed. Assuming a 10% drop-out rate at 1 year, the final study sample size would be 350 subjects.

A total of 356 patients participated in the study. A summary of the demographic profile of patients included in the ARICAS registry is shown in Table 1. Among the 356 patients, majority (96.3%) were those with brain infarction.

Mean age was 62.7 years (standard deviation [SD] 13.8, range 26.6–100), with more males than females. The majority were of Filipino (30.6%), East Asian (28.9%), or South Asian (23.6%) ethnicity. Among the stroke risk factors, the most common were hypertension (68.3%), diabetes mellitus (48.0%), and hyperlipidemia (48.9%).

MRI (55.9%) was the common baseline neuroimaging technique. Mean NIHSS on admission was 9 ± 8, with majority of patients having mild or moderate strokes. Most strokes (96.3%) were in the anterior circulation. 18.76% received intravenous thrombolysis; none received thrombectomy.

The analysis of characteristics and risk factors of patients with and without stroke recurrence anytime within the 12-month follow-up is shown in Table 2. In total, 24 patients (6.7%, 95% CI: 4.4–9.9%) had stroke recurrence. On univariable analysis, stroke recurrence was possibly associated only with increasing age (p = 0.076). On multivariable analysis, statistically significant association was seen with increasing age (odds ratio [OR]: 1.04, 95% CI: 1–1.06, p = 0.05) and hypertension (OR: 3.23, 95% CI: 1.09–9.61, p = 0.035).

A total of 47 (13.2%, 95% CI: 9.9–17.2%) patients had died by 1-year post-stroke. Analysis was performed to determine predictors of death within 12 months (Table 3). On univariable analysis, mortality was associated increasing age, hypertension, chronic smoking, alcohol intake, hyperuricemia, previous stroke, family history of stroke, illicit drug use, ischemic heart disease, and mean NIHSS. On multivariable analysis, only age (OR: 1.05, 95% CI: 1.01–1.08, p = 0.006) and NIHSS (OR: 1.12, 95% CI: 1.07–1.17, p < 0.001) increased the odds of death.

The ARICAS study is the first multinational study from Asia that documents hospital-based data on consecutive patients with symptomatic intracranial atherosclerosis. We found a 1-year stroke recurrence rate of 6.7%, with a trend of higher rates with increasing age and hypertension. We also found a 1-year mortality rate of 13.2%, associated with increasing age and NIHSS.

In our study, the median age of patients was 62.7 years. It is comparable to the median age of patients in the WASID study [20] in which warfarin was compared to aspirin in patients with symptomatic intracranial stenosis, and in the Chinese CICAS study [25].

Although intracranial artery luminal stenosis may be due to a variety of vasculopathies, other types of vasculopathy such as vasculitis, reversible vasoconstriction syndrome, and moyamoya disease can be differentiated clinically in most cases [26] and were excluded from this study. The strong association with atherosclerotic risk factors also suggested that atherosclerosis is the major cause of stenosis in our studied population. Among the traditional risk factors, hypertension was found to be the most prevalent (68.3%), followed by diabetes mellitus (48%) and dyslipidemia (48.9%). This is comparable with the systematic review and study from China [27] where hypertension, diabetes mellitus and dyslipidemia were the 3 most common risk factors. Since ICAS is the etiology of most ischemic stroke cases among Asians, stroke prevention through the management of modifiable factors and individual risk stratification by non-modifiable factors is crucial.

Arterial hypertension has been reported to cause hemodynamic stress and accelerate atherosclerosis by multiple mechanisms. Intracranial arteries are prone to damage induced by arterial hypertension than extracranial arteries. This may be explained by the different architecture of the walls of the intracranial arteries, thinner media, and adventitia, as well as fewer elastic medial fibers [28]. Excessive sodium intake, associated with fluid retention and increased blood pressure, is also considered as a specific characteristic of hypertensive disease in Asia [29]. A risk factor study demonstrated that high salt-sensitivity-associated genetic polymorphisms (e.g., genes coding a-adducin, angiotensinogen, and aldosterone synthase) were more frequent in Asians than that in Caucasians [30]. It has been shown that a 5 gram per day increase in salt intake could increase stroke risk by 23%, which becomes of concern given the high average daily salt intake in Asia [31].

In this study, diagnostic methods used to identify ICAS include TCD ultrasound, MRA, CTA, and DSA. The Stroke Outcomes and Neuroimaging of Intracranial Atherosclerosis (SONIA) trial [32] assessed the accuracy of TCD and MRA compared with DSA and showed that TCD and MRA had high negative predictive values (86–91%) but low positive predictive values (36–59%). These data suggest that TCD and MRA are useful screening tests for exclusion of intracranial arterial stenosis.

The risk of recurrent stroke at 1 year was 6.7%. This is similar to other cohort studies of ≥50% symptomatic ICAS – 5.1% at 14 months reported by Samaniego et al. [33] in the medically treated arm of their observational study, but less than the 56% at 16 months by Kozak et al. [34] and much older study by Wong and Li [35] of 17.1% in the first year. In the medical arm of recent randomized controlled trials of ≥50% symptomatic ICAS stroke or “hard TIA” was 15.1% at 12 months in the VISSIT trial [36], while recurrent stroke was 14% (aspirin arm) or 15% (warfarin arm) at 12 months in the older WASID trial [20]. Thus, our recurrence rates are lower than those in published studies, possibly due better secondary prevention and more aggressive management of vascular risk factors.

We found that increasing age and hypertension were significantly associated with recurrent stroke. The study by Wong and Li [35] also found age to be a predictor of recurrent cerebrovascular events (HR: 1.02; 95% CI: 1.00–1.03; p = 0.02); other predictors they found included atrial fibrillation (such patients were excluded in our study), and previous stroke or TIA (which we did not find), but no association with hypertension (which we had found). The WASID trial [34] found increased recurrent ischemic events among those with basilar artery stenosis, which we did not find. In a clinical trial among Chinese patients with symptomatic MCA stenosis [37], a history of hypertension was associated with high risk of any kind of ipsilateral stroke or transient ischemic attack, or death from any origin during 1-year follow-up, consistent with our study.

We found a 1-year mortality of 13.2%. This is similar to the 8.6% reported by Samaniego et al. [33], and 11.2% reported by Wong and Li [35]. Current guidelines to reduce the risk of stroke or death among patients with symptomatic ICS advise long-term aspirin with the addition of clopidogrel for the first 90 days, high dose statin to achieve a target low-density lipoprotein cholesterol level <70 mg/dL, a long-term blood pressure target of <140/90 mm Hg [38].

Mortality at 1 year in our study was significantly associated with increasing age and stroke severity. In the Oxford Vascular Study [39], patients with 50–99% symptomatic ICAS with recurrent stroke or who died during follow-up were older (p = 0.005), had a greater burden of diabetes (p < 0.001) (not found in our study), and higher 1-month systolic blood pressures than those without. In the WASID trial [20], women were at increased risk, although this was of borderline significance. As mentioned earlier, in a clinical trial among Chinese patients with symptomatic MCA stenosis [37], a history of hypertension was associated with high risk of any kind of ipsilateral stroke or transient ischemic attack, or death from any origin during 1-year follow-up. The much earlier study by Wong and Li [35] showed that risk of death increased with age (HR: 1.08; 95% CI: 1.05–1.10; p < 0.0001) as in our study, and atrial fibrillation (such patients were excluded from our study). Age was also a predictor in another observational study of symptomatic ICS comparing antiplatelets, warfarin and heparin [40]. While we were unable to find studies reporting the effect of stroke severity on mortality in patients with symptomatic ICAS, a review of 28 studies showed that age and high NIHSS score (as in our study), and body mass index (not found in our study) were identified as important predictors for mortality after stroke [41].

Atherogenesis in intracranial arteries at older ages was proposed to be secondary to declining antioxidant enzyme activities in the implicated vessels [42]. The frequent presence of intracranial atherosclerosis in younger ischemic stroke patients in Asia was consistent with previous studies [43‒45]. One postulated mechanism on why Asians develop intracranial atherosclerosis at a younger age compared to Caucasians was the differential presence of antioxidants in intracranial and extracranial sites [42]. Other contributing factors for atherogenesis such as the presence of multiple conventional risk factors affected our entire study cohort equally without impact on ischemic stroke recurrence [42, 46]. Hypertension and the presence of intracranial atherosclerosis may also be indicators for generalized vascular disease which may lead to greater mortality from stroke and nonstroke related deaths.

Our study has important clinical implications. First, the risk of stroke recurrence and mortality in ICAS is not insignificant – patients and families need to be informed, and advised to be compliant with medications. The association of stroke recurrence with hypertension suggests that blood pressure control after stroke remains an important therapeutic target [47]. The finding that mortality after stroke is associated with stroke severity underscores the need to better manage severely disabled stroke survivors who would be more likely to succumb to the complications of stroke, such as infection – support systems and care-giver training may be needed to provide the necessary assistance and skill-sets to better look after severe stroke patients.

Our study has some limitations. While the study centers were from different parts of Asia, only 6 institutions were involved. Sample size was modest. Compliance to treatment was not evaluated. As this is an Asian-based study, the results may not be extrapolated to non-Asian populations. Still, our study has some strengths. This is a multicenter study, a standardized protocol was followed for subject recruitment and evaluation and follow-up, well-accepted techniques were employed to diagnose ICAS, a broad range of factors known to affect outcome were analyzed.

ICAS is an important cause of acute ischemic stroke. Our multicenter Asian study has shown that stroke recurrence is associated with increasing age and hypertension, while mortality is associated with increasing age and stroke severity. More multicenter studies on ICAS in Asia are needed.

This study protocol was reviewed and approved by Institutional Review Board Committee, Jose R. Reyes Medical Center and Institutional Review Board Committee, Stroke Data Bank, St Lukes Medical Center, Approval No. CT 171138; Institutional Ethics Committee, Christian Medical College Ludhiana, Approval No. IEC/CMCL/145; Institutional Review Board, Faculty of Medicine, Chulalongkorn University, approval No. 595/61; Institutional Review Board, Chang Gung Medical Foundation, Approval No. 202001745B0; Medical Research Ethics Committee, University of Malaya Medical Centre, Approval No. 2018476207; and Institutional Review Board, Raffles Hospital, approval code ARICAS. Written informed consent to participate in the study was obtained from all adult participants and all vulnerable participants’ parent/legal guardian/next of kin.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Jose C Navarro led the study, conceptualized the study, wrote the paper, provided critical feedback, and gave final approval for journal submission. Jeyaraj D. Pandian, Tsong-Hai Lee, Kay Sin Tan, Narayanaswamy Venketasubramanian, and Nijasri C. Suwanwela conceptualized the study, wrote the paper, provided critical feedback, and gave final approval for journal submission.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author (N.V.) upon reasonable request.