Stroke mechanisms in symptomatic intracranial atherosclerotic disease: classification and clinical implications

Ph.D.Intracranial atherosclerotic disease (ICAD) is a common cause of ischemic stroke worldwide, with a particularly high prevalence in Asia. Patients with symptomatic ICAD face a high risk of stroke recurrence despite optimal medical treatment. ICAD may cause ischemic stroke through various mechanisms. However, little is known regarding how ICAD lesions cause ischemic strokes via different mechanisms, while the hemodynamic impairment caused by ICAD lesions may play an important role. One reason behind the high risk of recurrence in symptomatic ICAD may be the less-than- ideal recognition of the underlying hemodynamics, other characteristics of the lesion, and the stroke mechanisms, which limited individualized treatment for symptomatic ICAD. Therefore, in this thesis, we recruited patients with symptomatic ICAD (50-99% stenosis) confirmed in CT angiography (CTA), classified their underlying stroke mechanism and built computational cerebral hemodynamics models, to explore the clinical implications of different stroke mechanisms, the associations between cerebral hemodynamics and the stroke mechanisms, and whether such associations would affect outcomes of current medical treatment strategies in symptomatic ICAD patients.We firstly proposed criteria based on infarct topography and magnetic resonance/computed tomography angiography (MRA/CTA) to classify probable stroke mechanisms in symptomatic ICAD patients: parent artery atherosclerosis occluding penetrating artery, artery-to-artery (AA) embolism, hypoperfusion, and mixed mechanisms. Two investigators independently classified stroke mechanisms to assess inter- and intra-rater reproducibilities. To explore whether clinical features would differ in symptomatic ICAD due to different stroke mechanisms, we correlated stroke mechanisms with features of the patients at baseline and recurrent ischemic stroke in the same territory (SIT) or relevant transient ischemic attack within 1 year.Among 153 patients recruited, the most common stroke mechanisms were isolated hypoperfusion (35.3%) and mixed mechanism of AA embolism and hypoperfusion (37.3%) that was associated with a higher incidence of dyslipidemia (P=0.045) and hypertension (P=0.033) than patients with other stroke mechanisms. The proposed criteria showed substantial to excellent intra-rater and inter-rater reproducibilities (κ, 0.791–0.908). Overall, 31 patients received interventional treatment of the diseased intracranial artery; 122 received medical treatment, among whom a mixed mechanism of AA embolism and hypoperfusion at baseline was associated with a higher risk of ischemic stroke in the same territory within 1 year (24.4% versus 7.8%; hazard ratio, 3.40; 95% CI 1.25–9.20; log-rank P=0.010) than other mechanisms combined.Computational fluid dynamics (CFD) models built based on CTA source images can quantify cerebral hemodynamic metrics in ICAD patients; translesional pressure ratio (PR) reflects relative pressure drop across an ICAD lesion, with lower PR representing larger translesional pressure drop and probably impaired downstream perfusion; wall shear stress ratio (WSSR) reflects relative change in WSS across an ICAD lesion, with higher WSSR indicating significantly increased WSS at the stenotic throat. To explore the role of hemodynamics underlying different stroke mechanisms, we associated translesional PR and WSSR assessed in CTA based CFD (CTA-CFD) models with probable stroke mechanisms in symptomatic ICAD patients. Among 99 patients with successful CFD modeling at baseline, 44 had AA embolism as a probable stroke mechanism, 13 with AA embolism alone, and 31 with coexisting hypoperfusion; 18 and 37 respectively had parent artery atherosclerosis occluding penetrating artery and isolated hypoperfusion as the probable stroke mechanisms. High WSSR was independently associated with AA embolism (adjusted OR 4.86; 95% CI 1.50-15.70; P=0.008). The significant, positive relationship between high WSSR and a higher risk of AA embolism remained in those with a low PR (adjusted OR= 4.01; 95% CI 1.03- 15.46; P=0.044), but not in those with a normal PR (P=0.621).There is debate over an optimal systolic blood pressure (SBP) target in secondary stroke prevention of patients with symptomatic ICAD, while factors such as cerebral perfusion may affect the association between SBP and outcomes in such patients. We investigated whether translesional PR across ICAD lesion would alter the relationship between SBP during follow-up (SBPFU) and risk of recurrent ischemic stroke in the same territory within 1 year, in symptomatic ICAD patients. Among 157 eligible patients with successful CFD modeling, multivariate Cox regression revealed significant PR-SBPFU interaction on the primary outcome (P=0.008): in patients with normal PR, the risk of primary outcome significantly decreased with lower SBPFU (HR for 10 mmHg decrement=0.46; 95%CI 0.24-0.88; P=0.018); however, in those with low PR, SBPFU≤130 mmHg was associated with significantly increased risk of the primary outcome, compared with 130< SBPFU <150mmHg (HR 5.08; 95%CI 1.05- 24.49; P=0.043).In conclusion, in this thesis, we proposed reproducible criteria for classifying probable mechanisms of ICAD-related stroke based on routine MRI and MR/CT vascular imaging. AA embolism and hypoperfusion are common mechanisms of stroke attributed to ICAD, which often coexist. Patients with ICAD-related stroke caused by concurrent AA embolism and hypoperfusion possess a higher risk of recurrent ischemic stroke, who may need more stringent medical treatment or other treatment strategies that warrants further investigations. Moreover, we used CTACFD models to simulate cerebral blood flow and quantify cerebral hemodynamic features in symptomatic ICAD patients, which differed in those with different stroke mechanisms. Furthermore, we found that cerebral hemodynamics revealed by CFD models may alter the association between blood pressure and risk of recurrent in such symptomatic ICAD patients. Hemodynamics revealed by CFD models may yield promising indicators to guide more individualized management in symptomatic ICAD patients, which warrants future studies.在全球範圍內, 特別是在亞洲人群中, 顱內動脈粥樣硬化性疾病(ICAD)是引起缺血性卒中的常見病因。ICAD 可能通過多種機制引起缺血性卒中。但是,目前尚未完全清楚ICAD 如何通過不同機制引起缺血性卒中, 而ICAD 潜在的血液動力學改變可能在其中起到重要作用。症狀性ICAD 患者在接受目前最佳的藥物治療後, 仍面臨較高的缺血性卒中複發風險。這可能是由于對ICAD 潛在血液動力學和其他的特点, 及卒中機制的認識尚不足, 限制了症狀性ICAD患者的個體化治療。因此, 我們招募了經CT 血管造影(CTA)確診的症狀性ICAD(狹窄度程度: 50-99%)患者, 对其進行卒中機制分型, 并建立計算機模擬的腦血流動力學模型, 以探討不同卒中機制相关的臨床特点, 腦血流動力學與卒中機制之間的關系, 以及該關聯如何影響症狀性ICAD 患者的临床结局。首先我們提出了基于梗死分布和磁共振血管造影(MRA)/CTA 的卒中機制分類標准載體動脈阻塞穿支, 動脈至動脈(AA)栓塞, 低灌注以及混合機制。並由兩名研究者獨立的對患者基線期可能的卒中機制進行分類, 以驗證該機制的可重複性。爲探討了不同卒中機制導致的症狀性ICAD 患者的臨床特征是否不同,我們將卒中機制與患者的基綫期臨床特點, 以及一年内相關的卒中復發或短暫性腦缺血發作(TIA)相關聯。結果顯示, 該分類標准有较高的研究者內部及研究者間可重複性(κ, 0.791-0.908)。在納入的153 名患者中, 最常見的卒中機制是孤立的低灌注(35.3%)和動脈至動脈栓塞和低灌注的混合機制(37.3%), 且後者有較高的血脂異常(P=0.045)及高血壓(P=0.033)發病率。與其他卒中機制相比, 基線時動脈至動脈栓塞和低灌注共存的混合機制與一年內缺血性卒中的高風險顯著相關(24.4%比7.8%; HR 3.40; 95% CI 1.25–9.20; log-rank P=0.010)。基于CTA 影像建立的计算机流体力学(CFD)模型可以量化ICAD 患者的脑血流动力学指标; 壓力比值(PR)代表ICAD病變遠端與近端相比的壓力變化, 其中较低的PR 代表ICAD 病變遠端壓力下降較大, 遠端灌注可能受損; 管壁剪切力比值(WSSR)反映了ICAD 病变處WSS 的相对变化, 其中较高的WSSR 表示ICAD 狭窄處的WSS 顯著增加。爲了探討症狀性ICAD 患者中血流動力學特征與卒中機制之間的關係, 我們在症狀性ICAD患者中使用CFD測量PR及WSSR,並將其與潛在卒中機制相關聯。在成功建立CFD 模型99 例患者中, 其潛在的基線卒中機制, 44 例存在AA 栓塞 (其中13 例爲孤立的AA 栓塞, 31 例爲AA栓塞合並低灌注); 18 例和37 例分別爲孤立的載體動脈阻塞穿支, 和孤立的低灌注。高WSSR 與AA 栓塞獨立相關(校正OR 4.86; 95% CI 1.50-15.70; P=0.008)。且高WSSR 與較高的AA 栓塞風險之間的顯著正相關關系仅存在於低PR 患者中(校正OR= 4.01; 95% CI 1.03-15.46; P=0.044), 而未存在於正常PR 患者中(P=0.621)。目前在症狀性ICAD 患者的二級預防策略中, 對于收縮壓(SBP)的最佳控制目標尚存在爭論, 一些因素, 比如腦灌注, 可能會影響SBP 與臨床結局之間的關係。本研究探讨了ICAD 病變的PR 是否會影響症狀性ICAD 患者訪期間平均SBP(SBPFU)與1 年內目標ICAD 相關的缺血性卒中複發風險之間的關系。在成功建立CFD 模型的157 例症狀性ICAD 患者中, 多變量Cox 回歸顯示, PRSBPFU對主要終點事件的風險存在顯著的相互作用(P for interaction=0.008): 對于PR 正常的患者, 較低SBPFU 可降低主要終點事件的風險(SBP 每降低10 mmHg 的HR=0.46; 95% CI 0.24-0.88; P=0.018); 然而, 在低PR 患者中, 與130< SBPFU