国际神经病学神经外科学杂志  2017, Vol. 44 Issue (3): 270-274

扩展功能

文章信息

陈海恋, 何超明, 庞明武, 林康, 符明昌
CHEN Hai-Lian, HE Chao-Ming, PANG Ming-Wu, LIN Kang, FU Ming-Chang
高龄急性脑梗死患者血清同型半胱氨酸和载脂蛋白B/A1与颈动脉内膜中层厚度的相关性研究
Correlation of serum homocysteine level and apolipoprotein B/A1 ratio with carotid intima-media thickness in elderly patients with acute cerebral infarction
国际神经病学神经外科学杂志, 2017, 44(3): 270-274
Journal of International Neurology and Neurosurgery, 2017, 44(3): 270-274

文章历史

收稿日期: 2016-11-15
修回日期: 2017-05-27
高龄急性脑梗死患者血清同型半胱氨酸和载脂蛋白B/A1与颈动脉内膜中层厚度的相关性研究
陈海恋, 何超明, 庞明武, 林康, 符明昌     
海南省第三人民医院神经内科, 海南省三亚市 572000
摘要: 目的 探讨血清同型半胱氨酸(Hcy)及载脂蛋白B/A1(ApoB/ApoA1)水平在高龄急性脑梗死(ACI)患者中的变化及与颈动脉内膜中层厚度的相关性。方法 收集252例高龄ACI患者作为病例组,根据颈动脉IMT检测结果分为IMT正常组65例、IMT增厚组87例和颈动脉斑块形成组100例。依据其血清Hcy水平分为低Hcy组(Hcy≤10 μmol/L)85例和高Hcy组(Hcy>10 μmol/L)167例。另选取同期健康体检者190例作为对照组。比较各组血清Hcy、ApoB/ApoA-l、血脂及颈动脉IMT情况,多因素logistic回归分析颈动脉IMT增厚的影响因素,血清Hcy、ApoB/ApoA1与颈动脉IMT的相关性采用Pearson相关。结果 病例组血清Hcy、ApoB/ApoA-l及颈动脉IMT明显高于对照组(P < 0.05)。与IMT正常组比较,IMT增厚组血清Hcy、ApoB/ApoA-l、LDL水平明显升高(P < 0.05),且斑块形成组升高更为明显(P < 0.05)。高Hcy组血清Hcy、ApoB/ApoA-l及颈动脉IMT明显高于低Hcy组(P < 0.05)。多因素logistic回归分析显示Hcy及ApoB/ApoA-l是影响颈动脉IMT增厚的危险因素,其OR及95%CI分别5.138(3.862~8.825)和4.742(3.571~7.283)。相关分析显示,高龄ACI患者血清Hcy、ApoB/ApoA1与颈动脉IMT均呈正相关(r=0.672,P < 0.001;r=0.538,P=0.012)。结论 高龄ACI患者血清Hcy、ApoB/ApoA1及颈动脉IMT水平与动脉粥样硬化发生相关,且Hcy水平及ApoB/ApoA1比值增高是颈动脉IMT增厚的危险因素。
关键词高龄     急性脑梗死     同型半胱氨酸     载脂蛋白B/A1     颈动脉内膜中层厚度    
Correlation of serum homocysteine level and apolipoprotein B/A1 ratio with carotid intima-media thickness in elderly patients with acute cerebral infarction
CHEN Hai-Lian, HE Chao-Ming, PANG Ming-Wu, LIN Kang, FU Ming-Chang     
Department of Neurology, Third People's Hospital of Hainan Province, Sanya, Hainan 572000, China
Corresponding author: CHEN Hai-Lian, Email: yongxinzuoshi99@163.com.
Abstract: Objective To investigate the changes in serum homocysteine (Hcy) level and apolipoprotein B/A1 (ApoB/ApoA1) ratio in elderly patients with acute cerebral infarction (ACI) and their correlation with carotid intima-media thickness (IMT). Methods A total of 252 elderly patients with ACI were enrolled as case group, and according to carotid IMT, these patients were divided into normal IMT group with 65 patients, IMT thickening group with 87 patients, and carotid plaque formation group with 100 patients. According to serum Hcy level, the patients were divided into low Hcy group (Hcy ≤10 μmol/L) with 85 patients and high Hcy group (Hcy >10 μmol/L) with 167 patients. A total of 190 subjects who underwent physical examination were enrolled as control group. The groups were compared in terms of serum Hcy level, ApoB/ApoAl ratio, blood lipids, and carotid IMT. The multivariate logistic regression analysis was performed to analyze the influencing factors for carotid IMT thickening, and Pearson correlation was used to analyze the correlation of serum Hcy level and ApoB/ApoAl ratio with carotid IMT. Results The case group had significantly higher serum Hcy level, ApoB/ApoAl ratio, and carotid IMT than the control group (P < 0.05). Compared with the normal IMT group, the IMT thickening group had significant increases in serum Hcy level, ApoB/ApoAl ratio, and low-density lipoprotein level (P < 0.05), and the carotid plaque formation group had significantly greater increases (P < 0.05). The high Hcy group had significantly higher serum Hcy level, ApoB/ApoAl ratio, and carotid IMT than the low Hcy group (P < 0.05). The multivariate logistic regression analysis showed that Hcy (odds ratio [OR]=5.138, 95% confidence interval [CI] 3.862-8.825) and ApoB/ApoAl ratio (OR=4.742, 95% CI 3.571-7.283) were the risk factors for carotid IMT thickening. The correlation analysis showed that serum Hcy level and ApoB/ApoA1 ratio were positively correlated with carotid IMT in elderly patients with ACI (r=0.672, P < 0.001; r=0.538, P=0.012). Conclusions Serum Hcy level, ApoB/ApoA1 ratio, and carotid IMT are associated with the development of atherosclerosis in elderly patients with ACI, and Hcy level and ApoB/ApoA1 ratio are risk factors for carotid IMT thickening.
Key words: elderly     acute cerebral infarction     homocysteine     apolipoprotein B/A1 ratio     carotid intima-media thickness    

急性脑梗死(aute cerebral infarction, ACI)是高龄人群常见疾病,能够加速动脉粥样硬化的发生和发展,是导致患者发生死亡的主要危险因素之一[1, 2]。颈动脉内膜中层厚度(intima media thickness, IMT)具有操作简便,无创伤,易于重复等特点,可作为反映冠状动脉及全身动脉早期粥样硬化的重要指标[3]。研究表明,同型半胱氨酸(homocysteine, Hcy)水平升高与心脑血管事件的发生显著相关[4]。载脂蛋白B/A1(Apolipoprotein B/A1, ApoB/ApoA1) 是反映体内动脉粥样硬化与抗动脉硬化血脂平衡的理想指标[5]。但关于高龄急性脑梗死患者血清Hcy及ApoB/ApoA1水平变化与颈动脉IMT的关系仍存在争议。本研究拟通过观察高龄ACI患者血清Hcy及ApoB/ApoA1水平变化,探讨血清Hcy及ApoB/ApoA1水平与颈动脉IMT的关系,以期为临床防治提供依据。

1 资料与方法 1.1 一般资料

选取2014年1月至2016年12月海南省第三人民医院收治的高龄急性脑梗死患者252例作为病例组,男134例,女118例,年龄80~91岁,平均年龄(85.17±9.24) 岁。急性脑梗死诊断标准《中国急性缺血性脑卒中诊治指南2010》,并通过头颅磁共振成像(MRI)/计算机断层扫描(CT)及扩散加权(DW1) 确诊。排除标准:高血脂症、急性心力衰竭、心肌梗死、急性冠状动脉综合征、风湿性心脏病、严重心律失常、外周血管病、脑出血、恶性肿瘤、恶性贫血、急慢性感染性疾病、慢性阻塞性肺疾病、严重肝肾功能不全和自身免疫性疾病等。

另选取我院体检中心健康体检者190例作为对照组,男104例,女86例,年龄80~87岁,平均年龄(84.25±9.13) 岁。

1.2 方法 1.2.1 研究方法

将252例ACI患者根据颈动脉IMT检测结果分为IMT正常组65例、IMT增厚组87例和颈动脉斑块形成组100例。依据其血清Hcy水平分为低Hcy组(Hcy≤10 μmol/L)85例和高Hcy组(Hcy>10 μmol/L)167例。比较各组血清Hcy、ApoB/ApoAl、血脂及颈动脉IMT情况,并应用多因素logistic回归分析颈动脉IMT增厚的影响因素。

1.2.2 观察指标

收集所有研究对象的一般资料,包括年龄、性别、体质指数(BMI)、饮酒史、吸烟史、基础疾病、心率等。于清晨空腹采集静脉血5 ml,采用全自动生化分析仪分别检测Hcy、ApoB/ApoAl、高密度脂蛋白(HDL)、低密度脂蛋白(LDL)、总胆固醇(TC)及甘油三酯(TG)等。

1.2.3 颈动脉IMT检测

采用德国西门子Acuson Sequoia 512型彩色多普勒超声仪,扫查受检者颈总动脉全程、颈内外动脉长轴及短轴切面,并在右侧颈动脉分叉近端下方1 cm处,于心室舒张期(心电图R波处)测量血管长轴切面最厚处的IMT(管腔内膜交界面至中膜与外膜交界面之间的回声距离即为IMT)。重复测3次,取双侧平均值为受检者的IMT值。依次探查双侧颈总动脉、颈外动脉、颈内动脉有无斑块与钙化,以及斑块的数量。IMT < 1.0 mm为厚度正常,IMT≥1.0且 < 1.5 mm为增厚,IMT≥1.55 mm为斑块形成。

1.3 统计学方法

采用SPSS 17.0统计软件分析,所有连续变量均通过正态性检验和方差齐性检验,符合正态分布的计量资料以均数±标准差(x±s)表示,多组间比较采用单因素方差分析,组内两两比较采用独立样本t检验。计数资料的比较采用χ2检验。应用多因素logistic回归分析颈动脉IMT增厚的影响因素,相关性分析采用Pearson相关分析。P < 0.05为差异有统计学意义。

2 结果 2.1 两组临床特征基线资料比较

病例组与对照组的性别、年龄、BMI、饮酒史、吸烟史、高血压史、心率等比较差异无统计学意义(P>0.05)。见表 1

表 1 两组临床特征基线资料比较[n(%); x±s]
组别例数男性年龄(岁)BMI(kg/m2)饮酒史吸烟史高血压史心率(次/min)
对照组190104(54.7)84.25±9.1322.58±2.5446(24.2)52(27.4)54(28.4)73.86±7.34
病例组252134(53.2)86.12±9.3122.98±2.7656(22.2)77(30.6)86(34.1)75.31±7.84
χ2/t0.1060.5180.6420.2410.5321.6300.472
P0.7440.6720.5790.6230.4660.2020.695
2.2 两组血清Hcy、ApoB/ApoAl、血脂及颈动脉IMT比较

病例组血清Hcy、ApoB/ApoAl及颈动脉IMT明显高于对照组,差异有统计学意义(P < 0.05);而两组HDL、LDL、TC及TG水平比较,差异无统计学意义(P>0.05)。见表 2

表 2 两组血清Hcy、ApoB/ApoAl、血脂及颈动脉IMT比较(x±s)
组别例数Hcy(μmol/L)ApoB/ApoAlIMT(mm)HDL(mmol/L)LDL(mmol/L)TC(mmol/L)TG(mmol/L)
对照组1907.85±1.560.67±0.130.75±0.161.43±0.372.61±1.244.42±1.151.50±0.24
病例组25216.38±2.470.94±0.281.44±0.181.33±0.282.68±1.314.58±1.321.59±0.34
t10.2588.7533.620-1.4801.0961.0311.568
P< 0.001< 0.0010.0190.2050.3140.3430.201
2.3 各组血清Hcy、ApoB/ApoAl及血脂水平比较

斑块形成组和IMT增厚组血清Hcy、ApoB/ApoAl和LDL水平明显高于IMT正常组,差异有统计学意义(P < 0.05);且斑块形成组血清Hcy、ApoB/ApoAl和LDL水平明显高于IMT增厚组,差异有统计学意义(P < 0.05)。IMT正常组HDL水平明显高于斑块形成组和增厚组,差异有统计学意义(P < 0.05)。各组TC、TG水平比较,差异无统计学意义(P>0.05)。见表 3

表 3 各组血清Hcy、ApoB/ApoAl及血脂水平比较(x±s)
组别例数Hcy(μmol/L)ApoB/ApoAlHDL(mmol/L)LDL(mmol/L)TC(mmol/L)TG(mmol/L)
IMT正常组6511.85±1.960.71±0.171.51±0.341.96±0.354.43±1.161.53±0.27
IMT增厚组8718.15±2.53a1.02±0.32a1.38±0.27a3.11±0.54a4.68±1.341.61±0.32
斑块形成组10023.12±2.92ab1.22±0.41ab1.17±0.24ab3.77±0.68ab4.71±1.341.66±0.40
F9.31610.542-8.7365.9040.9920.755
P< 0.001< 0.001< 0.0010.0070.4050.497
注:a为与IMT正常组比较,P < 0.05;b为与IMT增厚组比较,P < 0.05。
2.4 低Hcy组和高Hcy组血清Hcy、ApoB/ApoAl、血脂及颈动脉IMT比较

高Hcy组血清Hcy、ApoB/ApoA-l及颈动脉IMT明显高于低Hcy组,差异有统计学意义(P < 0.05);而两组HDL、LDL、TC及TG水平比较,差异无统计学意义(P>0.05)。见表 4

表 4 低Hcy组和高Hcy组血清Hcy、ApoB/ApoAl、血脂及颈动脉IMT比较(x±s)
组别例数Hcy(μmol/L)ApoB/ApoAlIMT(mm)HDL(mmol/L)LDL(mmol/L)TC(mmol/L)TG(mmol/L)
低Hcy组858.76±1.630.77±0.201.30±0.131.32±0.262.78±1.284.52±1.281.54±0.25
高Hcy组16724.40±3.271.18±0.341.68±0.211.38±0.422.86±1.344.67±1.351.64±0.30
t15.50310.8343.4460.5320.7251.0221.453
P< 0.001< 0.0010.0340.6680.5020.3470.269
2.5 多因素logistic回归分析颈动脉IMT增厚的影响因素

以有或无颈动脉IMT增厚为因变量,以性别(女性=0,男性=1)、年龄、BMI、心率、吸烟史(无=0,有=1)、饮酒史(无=0,有=1)、高血压史(无=0,有=1)、Hcy、ApoB/ApoAl、HDL、LDL、TC及TG为自变量,按α=0.05水平进行多因素logistic回归分析,发现年龄、高血压史、Hcy、ApoB/ApoAl、HDL和LDL进人模型,差异均有统计学意义(P < 0.05)。见表 5

表 5 多因素logistic回归分析颈动脉IMT增厚的影响因素
因素偏回归系数Wald值OR95%CIP
年龄0.2253.2801.5421.311~1.9300.040
高血压史0.6393.1952.1652.065~3.1520.046
Hcy1.7345.9415.1383.862~8.8250.004
ApoB/ApoAl1.4905.6164.7423.571~7.2830.012
HDL0.7874.2632.3852.174~3.5440.035
LDL0.0963.6381.3351.172~2.2280.037
2.6 Hcy、ApoB/ApoA1、HDL和LDL与颈动脉IMT的相关性

Pearson相关分析显示,高龄ACI患者血清Hcy与颈动脉IMT呈正相关(r=0.672, P < 0.001);ApoB/ApoA1与颈动脉IMT呈正相关(r=0.538, P=0.012)。而HDL和LDL与颈动脉IMT均无明显相关性(r=-0.041和r=0.024,均P>0.05)。

3 讨论

随着我国人口老龄化的日趋严重,近年来高龄ACI发病率逐年增加。目前,颈动脉IMT作为早期预测动脉粥样硬化发生的指标,已广泛应用于判定动脉粥样硬化的进展程度[6, 7]。研究表明,颈动脉IMT增厚是全身动脉粥样硬化的早期征象,对心脑血管事件的发生有较高的预测性[8],也可作为脑梗死复发的预测指标[9]。Hcy是半胱氨酸和蛋氨酸的重要中间代谢产物,其水平升高可促进动脉粥样硬化的发生和血栓形成[10]。Dietrich等[11]研究发现,Hcy水平与动脉粥样硬化的形成具有较高的相关性,Hcy水平升高可预示动脉IMT增厚以及斑块的产生。血脂代谢异常也是影响动脉粥样硬化进程的主要原因,其中ApoB/ApoAl是预测ACI发生的较强因素[12]。ApoB/ApoAl比值反映了体内胆固醇转运的平衡,ApoB/ApoAl比值越高,说明有更多的胆固醇在血液循环中和更多的胆固醇沉积在血管壁中,这可加速动脉粥样硬化进一步发展为脑梗死。

本研究显示,病例组血清Hcy、ApoB/ApoAl及颈动脉IMT明显高于对照组,提示血清Hcy水平升高、ApoB/ApoAl比值增高及颈动脉IMT增厚可能参与动脉粥样硬化的发生发展。既往研究也表明,脑梗死患者的血清Hcy水平明显高于正常者[13]。颈动脉IMT检查结果显示,斑块形成组和增厚组血清Hcy、ApoB/ApoAl和LDL水平明显高于正常组,且斑块形成组血清Hcy、ApoB/ApoAl和LDL水平明显高于增厚组。提示Hcy、ApoB/ApoAl和LDL水平与颈动脉IMT增厚存在一定的相关性,Hcy水平及ApoB/ApoAl比值越高,其发生动脉粥样硬化风险越高。相关分析显示,高龄ACI患者血清Hcy水平和ApoB/ApoA1比值与颈动脉IMT均呈明显正相关,进一步支持上述研究结果。有研究表明,血清Hcy水平与颈动脉IMT之间可能存在协同作用,能共同加速动脉粥样硬化的过程[14]。Jain等[15]研究认为,ApoB/ApoAl与ACI的颈动脉IMT密切相关,是比血脂及脂蛋白能更好的预测动脉粥样硬化发生的指标。不同Hcy水平对动脉粥样硬化的影响程度不同,高水平的Hcy通过氧化应激损伤内皮细胞,破坏抗凝和纤溶系统,促进血小板聚集,进而加速动脉粥样硬化的进程。本研究显示,高Hcy组血清Hcy、ApoB/ApoAl及颈动脉IMT明显高于低Hcy组,与Wang等[16]研究结果相符合。Fu等[17]的临床荟萃分析显示,血清Hcy水平可作为ACI早期诊断及预后评估的重要生物学指标。另有研究也表明,ApoB/ApoAl比值不仅与颈动脉IMT水平相关,还与颈动脉不稳定斑块形成有关[18]

笔者进一步分析影响高龄ACI患者颈动脉IMT增厚的相关因素,发现年龄、高血压史、Hcy、ApoB/ApoAl、HDL和LDL是影响颈动脉IMT增厚的危险因素。由此可见Hcy水平升高及ApoB/ApoAl比值增高是导致动脉粥样硬化的重要危险因素。Rundek等[19]研究也发现,Hcy水平升高与动脉粥样硬化的发生发展有关,是ACI早期动脉粥样硬化的危险因素之一。另外ApoB/ApoAl比值增高也可刺激巨噬细胞内胆固醇酯化,促进泡沫细胞形成,从而导致动脉粥样硬化的发生。与HDL和LDL相比,ApoB/ApoAl比值受生物学标记的影响不大,能更准确地反映潜在的致动脉粥样硬化和抗动脉粥样硬化颗粒之间的平衡,可作为动脉粥样硬化及IMT增厚的预测因子[20]

综上所述,血清Hcy、ApoB/ApoA1水平增高可能参与了高龄ACI患者颈动脉硬化的发生发展,是颈动脉IMT增厚的危险因素,应采取措施积极干预,以降低脑梗死的发生率。

参考文献
[1]
Romano JG, Sacco RL. Decade in review-stroke:progress in acute ischaemic stroke treatment and prevention[J]. Nat Rev Neurol, 2015, 11(11): 619-621. DOI:10.1038/nrneurol.2015.199
[2]
Tanahashi N. Hypertension associated with cerebrovascular disease[J]. Nihon Rinsho, 2015, 73(11): 1864-1870.
[3]
Nezu T, Hosomi N, Aoki S, et al. Carotid Intima-Media Thickness for Atherosclerosis[J]. J Atheroscler Thromb, 2016, 23(1): 18-31. DOI:10.5551/jat.31989
[4]
Catena C, Colussi G, Nait F, et al. Elevated Homocysteine Levels Are Associated With the Metabolic Syndrome and Cardiovascular Events in Hypertensive Patients[J]. Am J Hypertens, 2015, 28(7): 943-950. DOI:10.1093/ajh/hpu248
[5]
Tani S, Nagao K, Anazawa T, et al. Relation of change in apolipoprotein B/apolipoprotein A-I ratio to coronary plaque regression after Pravastatin treatment in patients with coronary artery disease[J]. Am J Cardiol, 2010, 105(2): 144-148. DOI:10.1016/j.amjcard.2009.08.670
[6]
Mirza W, Arain M, Ali A, et al. Carotid intima media thickness evaluation by ultrasound comparison amongst healthy, diabetic and hypertensive Pakistani patients[J]. J Pak Med Assoc, 2016, 66(11): 1396-1400.
[7]
Tran LT, Park HJ, Kim HD. Is the carotid intima-media thickness really a good surrogate marker of atherosclerosis?[J]. J Atheroscler Thromb, 2012, 19(7): 680-690. DOI:10.5551/jat.11767
[8]
Lee S, Cho GY, Kim HS, et al. Common carotid intima-media thickness as a risk factor for outcomes in Asian patients with acute ST-elevation myocardial infarction[J]. Can J Cardiol, 2014, 30(12): 1620-1626. DOI:10.1016/j.cjca.2014.06.026
[9]
Suanprasert N, Tantirithisak T. Impact of risk factors for recurrent ischemic stroke in Prasat Neurological Institute[J]. J Med Assoc Thai, 2011, 94(9): 1035-1043.
[10]
McCully KS. Homocysteine and the pathogenesis of atherosclerosis[J]. Expert Rev Clin Pharmacol, 2015, 8(2): 211-219. DOI:10.1586/17512433.2015.1010516
[11]
Dietrich M, Jacques PF, Polak JF, et al. Segment-specific association between plasma homocysteine level and carotid artery intima-media thickness in the Framingham Offspring Study[J]. J Stroke Cerebrovasc Dis, 2011, 20(2): 155-161. DOI:10.1016/j.jstrokecerebrovasdis.2009.10.012
[12]
O'Donnell MJ, Xavier D, Liu L, et al. Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study):a case-control study[J]. Lancet, 2010, 376(9735): 112-123. DOI:10.1016/S0140-6736(10)60834-3
[13]
朱勇冬. 血清胱抑素C、基质金属蛋白酶-9及同型半胱氨酸与脑梗死关系的研究[J]. 国际神经病学神经外科学杂志, 2016, 43(3): 233-236.
[14]
Bogdanski P, Miller-Kasprzak E, Pupek-Musialik D, et al. Plasma total homocysteine is a determinant of carotid intima-media thickness and circulating endothelial progenitor cells in patients with newly diagnosed hypertension[J]. Clin Chem Lab Med, 2012, 50(6): 1107-1113.
[15]
Jain J, Lathia T, Gupta OP, et al. Carotid intima-media thickness and apolipoproteins in patients of ischemic stroke in a rural hospital setting in central India:A cross-sectional study[J]. J Neurosci Rural Pract, 2012, 3(1): 21-27. DOI:10.4103/0976-3147.91926
[16]
Wang C, Han L, Wu Q, et al. Association between homocysteine and incidence of ischemic stroke in subjects with essential hypertension:a matched case-control study[J]. Clin Exp Hypertens, 2015, 37(7): 557-562. DOI:10.3109/10641963.2015.1026039
[17]
Fu HJ, Zhao LB, Xue JJ, et al. Elevated Serum Homocysteine (Hcy) Levels May Contribute to the Pathogenesis of Cerebral Infarction[J]. J Mol Neurosci, 2015, 56(3): 553-561. DOI:10.1007/s12031-015-0497-6
[18]
Panayiotou A, Griffin M, Georgiou N, et al. ApoB/ApoA1 ratio and subclinical atherosclerosis[J]. Int Angiol, 2008, 27(1): 74-80.
[19]
Rundek T, Blanton SH, Bartels S, et al. Traditional risk factors are not major contributors to the variance in carotid intima-media thickness[J]. Stroke, 2013, 44(8): 2101-2108. DOI:10.1161/STROKEAHA.111.000745
[20]
Kostapanos MS, Christogiannis LG, Bika E, et al. Apolipoprotein B-to-A1 ratio as a predictor of acute ischemic nonembolic stroke in elderly subjects[J]. J Stroke Cerebrovasc Dis, 2010, 19(6): 497-502. DOI:10.1016/j.jstrokecerebrovasdis.2009.09.010