后循环缺血性卒中的危险因素和影像学特征：回顾性病例系列研究

Risk factors for posterior circulation stroke and imaging features： a retrospective case series study

目的探讨后循环缺血性卒中的主要危险因素以及合并糖尿病的后循环缺血性卒中患者的临床和影像学特征。方法收集急性缺血性卒中患者的临床资料，并对后循环缺血性卒中组与前循环缺血性卒中组进行比较；后循环卒中患者进一步分为糖尿病组和非糖尿病组，比较两组血管危险因素和影像学特征；将后循环缺血性卒中患者按病变血管分布分为近段组、中段组、远段组和混合组，分析糖尿病与各组之间的相关性和影像学特征。结果共纳入328例后循环缺血性卒中病例，其中男性194例，糖尿病组108例；前循环缺血性卒中336例，其中男性214例，糖尿病组59例。后循环缺血性卒中组糖尿病（32．9％对21．7％；Xz：10．501，P=0．001）、高脂血症（60．1％对47．9％；X^2=9．852，P=0．002）、既往卒中或短暂性脑缺血发作史（29．0％对22．0％ X^2=4．213，P=0．040）患者构成比均显著性高于前循环缺血性卒中组（P均〈0．05），而吸烟患者构成比显著性低于前循环缺血性卒中组（18．3％对26．2％；X^2=5．977，P=0．014）；总胆固醇[（4．72±1．07）mmol／L对（4．56±0．98）mmol／L；t=2．079，P=0．038]、三酰甘油[（1．54±1．07）mmol／L对（1．33±0．71）mmol／L；t=3．085，P=0．002]和低密度脂蛋白胆固醇[（2．91±0．90）mmol／L对（2．75±0．80）mmol／L；t=2．373，P=0．018]均显著性高于前循环缺血性卒中组，而高密度脂蛋白胆固醇显著性低于前循环缺血性卒中组[（1．13±0．31）mmol／L对（1．18±0．32）mmol／L；t=2．045，P=0．041]。多变量logistic回归分析显示，糖尿病[优势比（oddsratio，OR）1．560，95％可信区间（confidence interval，C1）1．086～2．239；P=0．016]和既往卒中或短暂性脑缺血发作史（OR1．455，95％CI1．013～2．090；P=0．042）是后循环缺血性卒中的独立危险因素。在后循环缺血性卒中患者中，糖尿病组（n=108）高脂血症（66．7％对55．5％；Ⅳ。=5．069，P=0．024）和饮酒（13．0％对4．5％；X^2=7．568，P=0．006）患者构成比显著性高于非糖尿病组（n=220），心房颤动患者的构成比显著性低于非糖尿病组（3．7％对11．4％；X^2=5．274，P=0．022）；三酰甘油[（1．70±0．93）mmol／L对（1．45±1．11）mmol／L；t=1．989，P=0．048]、空腹血糖[（8．46±2．96）mmol／L对（5．30±0．96）mmol／L；t=10．706，P=0．000]和糖基化血红蛋白[（8．36±1．94）％对（6．07±0．55）％；t=10．576，P=0．000]显著性高于非糖尿病组；大动脉粥样硬化性卒中患者构成比显著性高于非糖尿病组（73．1％对60．0％X^2=5．457，P=0．019），而心源性脑栓塞显著性低于非糖尿病组（2．8％对9．1％X^2=4．428，P=0．035）；后循环中段梗死患者构成比显著性高于非糖尿组（49．1％对31．4％；X^2=9．726，P=0．002）；脑干梗死（60．2％对48．2％ X^2=4．182，P=0．041）和单发性脑干梗死（55．6％对30．5％；X^2=19．235，P=0．000）患者构成比均显著性高于非糖尿病组。在单发性脑干梗死患者中，糖尿病组脑桥梗死（43．5％对25．9％；X^2=10．374，P=0．001）和延髓梗死（7．4％对1．8％；P=0．023）患者构成比均显著性高于非糖尿病组。结论糖尿病和既往卒中或短暂性脑缺血发作史是后循环缺血性卒中的独立危险因素。糖尿病与脑干梗死关系密切，更易导致脑桥梗死。

Objective To investigate the major risk factors for posterior circulation stroke and the clinical and imaging features of posterior circulation stroke patients with diabetes. Methods The patients with acute cerebral infarction were enrolled. The clinical data of patients with posterior circulation and anterior circulation stroke were compared. The patients with posterior circulation stroke were further divided into either a diabetic group or a non-diabetic group, and the vascular risk factors and imaging features of both groups were cor-npared. The patients with posterior circulation stroke were divided into proximal segnent, middle segnent and distal segment and mixed groups according to the distribution of vascular lesions. The correlations between diabetes and each group and the imaging features were analyzed. Results A total of 328 patients with posterior circulation stroke （male 194, the diabetic group 108） and 336 patients with anterior circulation stroke （male 214, the diabetes group 59） were enrolled. The proportions of patients with diabetes （32. 9% vs. 21.7% ; X^2 = 10. 501, P = 0. 001）, hyperlipidemia （60. 1% vs. 47. 9% ;X^2 = 9. 852, P = 0. 002）, previous stroke or transient ischemic attack （TIA） （29. 0% vs. 22. 0% ;X^2 =4. 213, P =0. 040） in the posterior circulation ischemic stroke group were significantly higher than those in the anterior circulation ischemic stroke group, and the proportion of smoking patients was significantly stroke group （18.3% vs. 26. 2% ; X^2 = 5. 977, P = lower than that in the anterior circulation ischemic 0. 014）. The levels of total cholesterol （4. 72 ±1.07 mmol/L vs. 4. 56 ± 0. 98 mmol/L; t = 2. 079, P = 0. 038）, triglycerides （1.54 ± 1.07 mmol/L vs. 1.33 ± 0. 71 mmol/L; t= 3. 085, P= 0. 002） and low-density lipoprotein cholesterol （2. 91± 0. 90 mmol/L vs. 2. 75 ±0. 80 mmol/L; t =2. 373, P =0. 018） were significantly higher than those in the anterior circulation ischemic stroke group, and the level of high-density lipoprotein cholesterol was sigtificantly lower than that in the anterior circulation ischemic stroke group （1.13 -＋ 0. 31 rnmol/L vs. 1.18 ＋-0. 32 mmol/L; t = 2. 045, P = 0. 041）. Multivariate logistic regression analysis showed that diabetes （odds ratio [ OR] 1. 560, 95% confidence interval [ CII 1. 086 - 2. 239; P = 0. 016） and previous stroke or TIA history （OR 1. 455, 95% CI 1. 013 -2. 090; P =0. 042） were the independent risk factors for posterior circulation ischemic stroke. In patients with posterior circulation ischemic stroke, the patient＇s proportions of hyperlipidemia （66. 7% vs. 55.5% ;X^2 = 5. 069, P = 0. 024） and drinking （13.0% vs. 4. 5% ;X： =7. 568, P =0. 006） in the diabetic group （n = 108） were significantly higher than those in the non-diabetic group （n =220）; the proportion of atrial fibrillation patients was significantly lower than that in the non-diabetic group （3.7% vs. 11.4% ;X^2 =5. 274, P =0. 022）. The levels of triglycerides （1.70 ＋ 0. 93 mmol/L vs. 1.45-＋ 1.11 lnn^l/L; t = 1. 989, P= 0. 048）, fasting glucose （8.46 ~ 2. 96） mmol/L vs. 5.30 -＋ 0. 96 mmol/L; t = 10. 706, P = 0. 000） and glycosylated hemoglobin （8. 36% -＋ 1.94% vs. 6. 07% -＋ 0. 55% ; t = 10. 576, P = 0. 000） in the diabetic group were significantly higher than those in the non-diabetic group. The proportion of patients with large artery atherosclerosis stroke in the diabetic group was significantly higher than that in the non-diabetic group （73.1% vs. 60.0%; X^2= 5.457, P= 0.019）; the proportion of the patients with cardloembollsm was significantly lower than that of the non-diabetic group （2. 8% vs. 9. 1%;X^2= 4. 428, P =0. 035）. The proportion of patients with posterior circulation middle segment infarction in the diabetic group was significantly higher than that of the non-diabetic group （49. 1% vs. 31.4% ; X^2 = 9. 726, P = 0. 002）. The proportions of the patients with brainstem infarction （60. 2% vs. 48. 2% X^2 =4. 182, P =0. 041） and single brainstem infarction （55.6% vs. 30. 5% ;X^2 = 19. 235, P = 0. 000） in the diabetic group were significantly higher than those in the non-diabetic group. In patients with single brainstem infarction, the proportions of the patients with pontine infarction （43.5% vs. 25.9% ;X^2 = 10. 374, P = 0. 001） and medulla oblongata infarction （7.4% vs. 1.8% ; P = 0. 023） in the diabetic group were significantly higher than those in the non-diabetic group. Conelusiom Diabetes and previous stroke or TIA history are the independent risk factor for posterior circulation stroke. Diabetes is closely associated with brainstem infarction, and it is more likely to result in pontine infarction.