Prediction of Recurrent Stroke or Transient Ischemic Attack After Noncardiogenic Posterior Circulation Ischemic Stroke
Background and Purpose—Posterior circulation ischemic stroke (IS) is generally considered an illness with a poor prognosis. However, there are no effective rating scales to predict recurrent stroke following it. Therefore, our aim was to identify clinical or radiological measures that could assist in predicting recurrent cerebral ischemic episodes.
Methods—We prospectively enrolled 723 noncardiogenic posterior circulation IS patients with onset of symptoms <7 days. Stroke risk factors, admission symptoms and signs, topographical distribution and responsible cerebral artery of acute infarcts, and any recurrent IS or transient ischemic attack (TIA) within 1 year were assessed. Cox regression was used to identify risk factors associated with recurrent IS or TIA within the year after posterior circulation IS.
Results—A total of 40 patients (5.5%) had recurrent IS or TIA within 1 year of posterior circulation IS. Multivariate Cox regression identified chief complaint with dysphagia (hazard ratio [HR], 4.16; 95% confidence interval [CI], 1.69–10.2; P=0.002), repeated TIAs within 3 months before the stroke (HR, 15.4; 95% CI, 5.55–42.5; P<0.0001), responsible artery stenosis ≥70% (HR, 7.91; 95% CI, 1.00–62.6; P=0.05), multisector infarcts (HR, 5.38; 95% CI, 1.25–23.3; P=0.02), and not on antithrombotics treatment at discharge (HR, 3.06; 95% CI, 1.09–8.58; P=0.03) as independent predictors of recurrent IS or TIA.
Conclusions—Some posterior circulation IS patients are at higher risk for recurrent IS or TIA. Urgent assessment and preventive treatment should be offered to these patients as soon as possible.
The recurrence rate is high in Chinese patients with ischemic stroke (IS).1 Several predictive scores2–5 have been developed to identify patients at higher risk of developing recurrent IS. However, most instruments do not consider stenosis of responsible cerebral artery and imaging features of acute infarcts as prediction factors, although IS patients with stenosis of cerebral artery or multiple infarcts have a higher risk of recurrence or early neurological deterioration.5–7 Besides, posterior circulation IS is generally considered an illness with a poor prognosis because of high rates of recurrence, mortality, and severe disability.8,9 Hence, we included stenosis of responsible cerebral artery, imaging features of acute infarcts, primary symptoms, and signs at admission in our analysis, to identify possible risk factors associated with recurrent IS or transient ischemic attack (TIA) in posterior circulation IS patients.
CICAS study (Chinese Intracranial Atherosclerosis) is a prospective, multicenter, hospital-based study. Clinical and imaging data were prospectively collected from consecutive patients with IS or TIA in 22 Chinese general hospitals. From October 2007 to June 2009, 2864 patients with noncardioembolic ischemic cerebrovascular diseases were enrolled in CICAS. Among them, 723 patients were diagnosed as posterior circulation IS and enrolled into this study. The institutional review board of the participating hospitals approved this study. Each participant signed an informed consent.
Patients enrolled into the study had the onset of symptoms within 7 days and were between 18 and 80 years old. Patients were excluded if they were clinically unstable, required close monitoring, disabled before admission (modified Rankin Scale score of >2), and physically or subjectively unable to comply with magnetic resonance imaging (MRI). We excluded patients with cardioembolic risk factors (atrial fibrillation, valvular heart disease, postcardiac valve replacement, etc) and patients with undetermined causes or other causes. Patients diagnosed as TIA, patients without available magnetic resonance images identifying new cerebral infarct, patients with anterior circulation IS, and patients who underwent angioplasty or stent implantation of intracranial or extracranial artery were also excluded. The final analysis included 723 posterior circulation IS patients (Figure I in the online-only Data Supplement).
Clinical Information Assessment
The clinical information collected included age, sex, hypertension (defined as a history of hypertension or diagnosed at discharge), diabetes mellitus (defined as a history of diabetes mellitus or diagnosed at discharge), hyperlipidemia (defined as low-density lipoprotein cholesterol ≥2.6 mmol/L at the time of admission or a history of hyperlipidemia or received lipid-lowering treatments or diagnosed at discharge), history of IS or intracerebral hemorrhage, history of coronary heart disease (defined as a history of myocardial infarction or angina pectoris), National Institutes of Health Stroke Scale score at admission and discharge, and the modified Rankin Scale before stroke, at discharge and 1 year after stroke onset. Smoking history, current or previous smokers (continuously smoking ≥1 cigarette a day for 6 months), and history of heavy alcohol use (drinking >2 U/d on average for men or >1 U/d on average for women) were also collected. Uses of antithrombotics within 48 hours of admission, at discharge, and in 1 year after stroke onset were recorded. The frequency of primary symptoms and signs at admission was also recorded.
All 723 patients underwent MRI on a 3.0-T magnetic resonance scanner. Imaging sequences obtained included 3-dimensional (3D) time-of-flight magnetic resonance angiography (MRA), axial T2-weighted, T1-weighted imaging, fluid-attenuated inversion recovery sequences, and diffusion-weighted imaging (DWI). All above sequences except MRA had 5 mm slice thickness and 1.5 mm interslice gap. Magnetic resonance images were viewed by using software (RadiAnt DICOM Viewer184.108.40.20639, Medixant Ltd, Poznań, Poland).
The stenotic degree of intracranial and extracranial vessels was examined using 3D time-of-flight MRA and contrast-enhanced MRA or duplex color Doppler ultrasound, respectively. Responsible artery for acute cerebral infarct was defined as intracranial or extracranial artery responsible for acute cerebral infarcts according to the distribution characteristic of infarct lesions and examination results of MRA and contrast-enhanced MRA or color Doppler ultrasound. The degree of intracranial stenosis on MRA was calculated using the published method for the Warfarin–Aspirin Symptomatic Intracranial Disease Study.10 All measurements were made using Wiha DigiMax Digital Calipers 6″ (Germany) with a resolution of 0.01 to 0.03 mm for 0 to 100 mm. The degree of extracranial vertebral artery (VA) stenosis was estimated with ultrasonographic examination according to the published diagnostic criteria or calculated according to North American Symptomatic Carotid Endarterectomy Trial criteria by contrast-enhanced MRA.11,12 We assessed the following arterial segments: VA, basilar artery (BA), and posterior cerebral artery P1/P2. According to the severity of the stenosis, we classified the responsible cerebral vessels into 4 groups: <50% or no stenosis, 50% to 69% stenosis, 70% to 99% stenosis, and occlusion groups.
Topographical distribution of acute infarcts (including single or multiple acute infarcts, single- or multi-sector infarcts, single- or multi-level infarcts, watershed infarcts, small cortical infarct, and territorial infarct) was evaluated (Figure). Multiple acute cerebral infarcts were defined as ≥2 separate lesions that were hyperintense on DWI. Multisector infarcts were defined as multiple discrete infarcts supplied by ≥2 arteries (including bilateral VA, bilateral posterior inferior cerebellar artery, bilateral anterior inferior cerebellar artery, BA, bilateral superior cerebellar artery, and bilateral posterior cerebral artery). Arterial territories of posterior circulation ischemia were categorized according to Tatu et al13 classification of the vascular anatomy. The level of posterior circulation infarction was also categorized as proximal, middle, or distal level according to the previously published classification.14 When infarcts were located in >1 level, they were classified in multilevel infarcts. Watershed infarcts were defined as infarcts occurred at a junction of 2 (or 3) artery territories with arterial collateral circulation.15 With the exception of watershed, small cortical infarct was defined as cortical infarct with a maximum diameter of <2 cm. Territorial infarct was defined as a large ischemic lesion with a maximum diameter of ≥2 cm involving the cerebral cortical and subcortical structure in ≥1 major cerebral artery territories.16 Etiologic subtypes of IS were classified according to the SSS-TOAST (Stop Stroke Study Trial of ORG 10172 in Acute Stroke Treatment) classification criteria.17 Stroke mechanism of large artery atherosclerosis IS was determined as parent artery occluding penetrating artery if isolated acute infarct located in penetrating artery territory and the parent artery had evidence of plaque or any degree of stenosis.18
Two radiologists blinded to the clinical details read all MRI scans. Consensus was reached by them if they had disagreement on interpretations.
Follow-Up and Clinical Outcome Evaluations
At 3, 6, and 12 months after discharge, patients or their relatives were contacted over the telephone by trained research personnel at Beijing Tian Tan Hospital and were asked whether patients had new symptoms or hospitalized again with another stroke. The primary outcome was recurrence of IS or TIA in 1 year.
Recurrence of IS was defined as a new focal neurological deficit of vascular origin lasting >24 hours and without hemorrhage on computed tomography or MRI of the brain.19 A recurrent TIA was defined as a new focal neurological deficit sustained for a duration of <24 hours caused by ischemia in the brain or retina. All recurrence of IS or TIA was verified at the index hospitals based on the presence of new neurological deficits documented in the medical records combining with computed tomography or MRI images. An experienced stroke neurologist reviewed the patients’ medical document to ensure a reliable diagnosis of recurrence of IS. In case of an unclear event that was not hospitalized, the case would be adjudicated by a stroke neurologist and the principle investigator. Any death was verified by examining the hospital medical records or local citizen registry.
The Mann–Whitney U test was used for comparison of continuous variables with non-normal distribution. χ2 test was used for comparison of categorical variables. Univariable Cox regression analysis was used to identify baseline differences in clinical and imaging variables between patients with or without recurrent IS or TIA. The risk of subsequent IS or TIA in relation to each variable was determined by multivariable Cox regression analysis. The regression model included the time from stroke onset to a recurrent stroke or TIA as response, and clinical and imaging predictors of recurrence with a univariate P value <0.05 as independent variables. All parameters that were significant by univariate analysis at P<0.05 level were included in the multivariable Cox regression analysis. All probability values were 2-tailed; P<0.05 was considered to be statistically significant. All analyses were performed by using SAS Version 9.1 (SAS Institute, Cary, NC).
General Patient Characteristics
The studied population consisted of 723 patients (479 men, 244 women) with a mean age of 62.3±10.7 years (range, 25–80 years). As for vascular risk factors, 602 patients (83.3%) had hypertension, 319 (44.1%) had diabetes mellitus, and 560 (77.5%) had hyperlipidemia. In addition, 176 patients (24.4%) had a history of stroke, 66 (9.1%) had a history of coronary heart disease, and 20 (2.8%) patients had repeated TIAs in 3 months before the stroke onset. Admission symptoms and signs of 723 patients were presented in Table 1. Dizziness, diplopia, dysphagia, and decreased alertness were the primary symptoms or signs in 176 (24.3%), 62 (8.6%), 69 (9.5%), and 20 (2.8%) patients. The distribution features of infarcts were presented in Table 2. Multisector infarcts and multilevel infarcts were present in 69 (9.5%) and 79 (11.0%) patients, respectively. Patients with multisector infarcts more often had watershed infarcts and multilevel infarcts and more commonly had decreased alertness at admission (Table I in the online-only Data Supplement). A total of 448 IS patients (62.0%) were because of large artery atherosclerosis, whereas 275 (38.0%) were because of small artery occlusion. Responsible artery stenosis ≥70%, multiple acute infarcts, and watershed infarcts were present in 326 (45.1%), 180 (24.9%), and 56 (7.7%) patients, respectively (Table 3).
Risk Factors Associated With Recurrence of IS or TIA in 1 Year
Forty patients (5.5%) had recurrence of IS or TIA in 1 year, and, of these, 7 patients (1.0%) died, and 35 and 5 patients had recurrent IS or TIA involving the posterior and the anterior circulation, respectively.
Univariate analysis found that admission symptom of dysphagia, visual field loss, or bilateral limb weakness was associated with a higher recurrence risk, whereas patients with admission symptom of unilateral limb weakness or facial palsy had a lower recurrence risk in 1 year (Table 1). History of stroke, higher prestroke modified Rankin Scale score, repeated TIAs before the stroke, responsible artery stenosis ≥70%, multiple acute infarcts, multisector infarcts, multilevel infarcts, and not on antithrombotics treatment at discharge were associated with a higher recurrence risk in 1 year (Table 3).
When adjusted for age, sex, and vascular risk factors, multivariate Cox regression identified admission symptom of dysphagia (hazard ratio [HR], 4.16; 95% confidence interval [CI], 1.69–10.2; P=0.002), repeated TIAs before the stroke (HR, 15.4; 95% CI, 5.55–42.5; P<0.0001), responsible artery stenosis ≥70% (HR, 7.91; 95% CI, 1.00–62.6; P=0.05), multisector infarcts (HR, 5.38; 95% CI, 1.25–23.3; P=0.02), and not on antithrombotics treatment at discharge (HR, 3.06; 95% CI, 1.09–8.58; P=0.03) as independent predictors of recurrent IS or TIA within 1 year, although chief complaint with facial palsy was associated with a lower risk of recurrent stroke or TIA (HR, 0.41; 95% CI, 0.18–0.94; P=0.04; Table 4).
Caplan20 found that >40% posterior circulation IS patients had >50% stenosis or occlusion of VA or BA in the New England Medical Center posterior circulation registry. We also found that 45% patients had >70% stenosis of responsible artery (including VA, BA, and posterior cerebral artery P1/P2). Previous studies found that vertebrobasilar stenosis predicted a high recurrence risk in patients with posterior circulation stroke.21 This study confirmed that the presence of a severe responsible arterial stenosis was predictive of recurrent stroke or TIA. Severe intracranial vertebrobasilar artery stenosis is associated with the detection of circulating emboli distal to the stenosis.22 Intracranial VA is a main source of embolism, whereas BA and posterior cerebral artery are the most common intracranial posterior circulation recipient sites for embolism.20 Severe arterial stenosis may cause hypoperfusion of distal portion of the artery, embolism to the distal artery, or occlusion of small penetrating artery, so it is easy to understand that severe responsible arterial stenosis will increase the risk of recurrent stroke or TIA.
It has been shown that multiple DWI lesions suggest high risk of early future ischemic events.5 Multiple acute infarcts are often caused by small emboli from an unstable source such as rupture of plaque in a large artery atherosclerosis,5 whereas isolated subcortical or deep lesions are often caused by local small artery disease or parent artery plaque at the mouth or along the length of the penetrating artery.23 Multiple and embolic type infarcts suggest the presence of microemboli, and thromboembolism from the site of arterial stenosis seems to be the predominant cause of ischemia.22 Therefore, multisector DWI lesions suggest that the most possible stroke mechanism was thromboembolism. Patients with multisector DWI lesions had a higher risk of recurrence perhaps because of a high risk for second rupture of unstable atherosclerosis plaque or acute occlusion of responsible artery.
Multiple TIAs within 7 days were reported to be associated with an increased subsequent risk of stroke than after a single TIA.24 We also found that repeated TIAs within 3 months before the stroke were associated with a greater subsequent risk of recurrent stroke or TIA. This high risk of early stroke could be explained by repeated embolisms or hemodynamic mechanism. Immediate initiation of preventive treatment might greatly reduce the risk of stroke recurrence.25 Therefore, urgent assessment and treatment is necessary for these patients with repeated TIAs.
Dysphagia was found to be positively correlated with infarcts in the proximal vascular territory of posterior circulation, moderate-to-severe stroke, progressive neurological deficit, and 3-month modified Rankin Scale score of 3 to 6.14,26 Dysphagia may be caused by bilateral pyramidal tract lesions or lateral medullary infarcts. We found that 15 (21.7%), 1 (1.4%), and 35 (50.7%) patients with dysphagia had infarcts located in VA, posterior inferior cerebellar artery, and BA territory, respectively. Patients with dysphagia more often had multisector (12, 17.4%) and multilevel infarcts (15, 21.7%). Patients with dysphagia at admission are at higher risk for recurrent IS or TIA. It is reasonable to pay more attention to these patients. In contrast, patients with chief complaint of facial palsy had a lower risk for recurrent IS or TIA. This is perhaps because more patients (338, 78.1%) with admission symptom of facial palsy had single infarct, whereas less patients (37, 8.5%) had multisector infarct.
Much of the management of a patient with posterior circulation stroke is similar to that for anterior circulation stroke. We found that antithrombotics treatment at discharge can decrease the risk of recurrent stroke or TIA within 1 year. This finding emphasizes the importance of effective secondary prevention after posterior circulation IS.
Our study found that some stroke risk factors, such as hypertension, diabetes mellitus, and hyperlipidemia, were not associated with a higher recurrence risk, this is perhaps because of the short follow-up time (1 year) and the small number of recurrent patients (40) in this study. All patients enrolled in our study were hospitalized, and 706 patients (98.3%) received antithrombotic therapy after admission. Besides, our study excluded cardioembolic stroke. All these may explained the lower recurrence rate in some degree. Our study had several limitations. First, only hospitalized patients were included in our study, besides, patients who were clinically unstable, required close monitoring, disabled before admission, unable to comply with MRI were excluded, these may possibly result in selection bias. Second, some extracranial artery stenosis was calculated by contrast-enhanced MRA, whereas others were estimated with ultrasonographic examination; these may cause some differences when compared. Third, not all patients underwent transesophageal echocardiography and 24-hour Holter monitoring to exclude IS possibly because of cardioembolism, although the possibility is low. Fourth, the size of the recurrent events was limited in this study, although a large number of variables were included in the multivariate Cox regression model. These may result in a limited statistical power. Therefore, we only detected the association between some important potent predictors and stroke recurrence, but did not create any prediction scale. Additional studies with greater sample sizes and longer follow-up periods are required to validate our findings.
In conclusion, it is necessary to pay more attention to those posterior circulation IS patients with responsible artery stenosis ≥70%, multisector infarcts, repeated TIAs within 3 months before the stroke, and admission symptom of dysphagia. Urgent assessment and preventive treatment should be offered to these patients as soon as possible.
Sources of Funding
This study was funded by National Key Technology Research and Development Program of the Ministry of Science and Technology of the People’s Republic of China, (grant no. 2015BAI12B04, 2015BAI12B02, and 2013BAI09B14), and the Department of Organization of Beijing City (grant no. 2016000021469G215).
Guest Editor for this article was Tatjana Rundek, MD, PhD.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.116.016285/-/DC1.
- Received December 4, 2016.
- Revision received April 14, 2017.
- Accepted April 20, 2017.
- © 2017 American Heart Association, Inc.
- Weimar C,
- Diener HC,
- Alberts MJ,
- Steg PG,
- Bhatt DL,
- Wilson PW,
- et al
- Kernan WN,
- Viscoli CM,
- Brass LM,
- Makuch RW,
- Sarrel PM,
- Roberts RS,
- et al
- Schonewille WJ,
- Wijman CA,
- Michel P,
- Rueckert CM,
- Weimar C,
- Mattle HP,
- et al
- Flossmann E,
- Rothwell PM
- Samuels OB,
- Joseph GJ,
- Lynn MJ,
- Smith HA,
- Chimowitz MI
- Szabo K,
- Kern R,
- Gass A,
- Hirsch J,
- Hennerici M
- Ois A,
- Gomis M,
- Rodríguez-Campello A,
- Cuadrado-Godia E,
- Jiménez-Conde J,
- Pont-Sunyer C,
- et al
- Caplan L
- Gulli G,
- Khan S,
- Markus HS
- Hwang J,
- Kim SJ,
- Hong JM,
- Bang OY,
- Chung CS,
- Lee KH,
- et al
- Purroy F,
- Jiménez Caballero PE,
- Gorospe A,
- Torres MJ,
- Alvarez-Sabin J,
- Santamarina E,
- et al