摘要： 目的 筛查急性缺血性卒中患者整体脑白质高信号（WMH）以及不同部位脑白质高信号［脑室旁白质高信号（PWMH）和脑深部白质高信号（DMWH）］高负荷的影响因素。方法 选择 2023 年7-12 月郑州大学第一附属医院收治的 1161 例急性缺血性卒中患者，根据 Fazekas 评分评估脑室旁白质高信号和脑深部白质高信号负荷程度，再以二者评分之和评估整体脑白质高信号负荷。采用单因素和多因素逐步法 Logistic 回归分析筛查脑白质高信号、脑室旁白质高信号和脑深部白质高信号高负荷的影响因素。结果 共 1161 例患者根据 Fazekas 评分分为 WMH 低负荷组（0 ~ 2 分，493 例）和 WMH 高负荷组（3 ~ 6 分，668 例）、PWMH 低负荷组（0 ~ 1 分，560 例）和 PWMH 高负荷组（2 ~ 3 分，601 例）、DWMH 低负荷组（0 ~ 1 分，670 例）和 DWMH 高负荷组（2 ~ 3 分，491 例）。基于插补后数据集的 Logistic 回归分析显示，年龄增长（OR = 1.060，95%CI：1.046 ~ 1.075，P = 0.000；OR = 1.060，95%CI：1.046 ~ 1.075，P = 0.000；OR =1.047，95%CI：1.033 ~ 1.060，P = 0.000）、缺血性卒中病史（OR = 1.881，95%CI：1.389 ~ 2.546，P = 0.000；OR = 1.508，95%CI：1.132 ~ 2.010，P = 0.005；OR = 1.833，95%CI：1.380 ~ 2.434，P = 0.000）是脑白质高信号、脑室旁白质高信号和脑深部白质高信号高负荷共同的危险因素，肾小球滤过率估计值（eGFR）增加是其低负荷共同的保护因素（b = - 0.012，OR = 0.988，95%CI：0.979 ~ 0.996，P = 0.004；b = - 0.012，OR =0.988，95%CI：0.980 ~ 0.996，P = 0.004；b = - 0.008，OR = 0.992，95%CI：0.983 ~ 1.000，P = 0.048）；此外，肥胖是脑白质高信号（OR = 1.620，95%CI：1.100 ~ 2.385；P = 0.014）和脑室旁白质高信号（OR = 1.712，95%CI：1.176 ~ 2.494；P = 0.005）高负荷的危险因素，而女性（OR = 1.521，95%CI：1.118 ~ 2.068；P = 0.008）、高血压（OR = 1.892，95%CI：1.391 ~ 2.573；P = 0.000）是脑深部白质高信号高负荷的危险因素。结论 年龄增长、肥胖、缺血性卒中病史、高血压、eGFR 下降是急性缺血性卒中脑白质高信号高负荷的危险因素，而脑室旁白质高信号和脑深部白质高信号高负荷的危险因素不同，应积极探寻并控制脑白质高信号危险因素，改善患者预后。

关键词: 缺血性卒中, 脑白质疏松症, 磁共振成像, 危险因素, Logistic模型

Abstract: Objective To explore the influencing factors of overal white matter hyperintensity (WMH) load and WMH load in different brain regions [periventricular white matter hyperintensity (PWMH) and deep white matter hyperintensity (DWMH)] load in patients with acute ischemic stroke (AIS). Methods A total of 1161 patients hospitalized in The First Affiliated Hospital of Zhengzhou University between July and December 2023 for acute ischemic stroke were included. The load of PWMH and DWMH were graded according to the Fazekas visual grading scale separately. The overal load of WMH (referred to as total WMH) was assessed according to the sum of PWMH and DWMH scores. The influencing factors of total high WMH load, PWMH load and DWMH load were investigated by univariate and multivariate Logistic regression analyses. Results According to Fazekas score, the 1161 patients were divided into low WMH load group (0-2 points, n = 493) and high WMH load group (3-6 points, n = 668), low PWMH load group (0-1 points, n = 560) and high PWMH load group (2-3 points, n = 601), low DWMH load group (0-1 points, n = 670) and high DWMH load group (2-3 points, n = 491). In multivariate Logitic analysis, based on the multiple imputation dataset, age (OR = 1.060, 95%CI: 1.046-1.075, P = 0.000; OR = 1.060, 95%CI: 1.046-1.075, P = 0.000; OR = 1.047, 95%CI: 1.033-1.060, P = 0.000) and a history of ischemic stroke (OR = 1.881, 95%CI: 1.389-2.546, P = 0.000; OR = 1.508, 95%CI: 1.132-2.010, P = 0.005; OR = 1.833, 95%CI: 1.380-2.434, P = 0.000) were the common risk factors of high load of WMH, PWMH and DWMH, while an increase in estimated glomerular filtration rate (eGFR) served as a common protective factor (b = -0.012, OR = 0.988, 95%CI: 0.979-0.996, P = 0.004; b = -0.012, OR = 0.988, 95%CI: 0.980-0.996, P = 0.004; b = -0.008, OR = 0.992, 95%CI: 0.983-1.000, P = 0.048). In addition, obesity were the risk factors of high WMH (OR = 1.620, 95%CI: 1.100-2.385; P = 0.014) and PWMH (OR = 1.712, 95%CI: 1.176-2.494; P = 0.005) load, while female (OR = 1.521, 95%CI: 1.118-2.068; P = 0.008) and hypertention (OR = 1.892, 95%CI: 1.391-2.573; P = 0.000) were the risk factors of high DWMH load. Conclusions Aging, obesity, a history of ischemic stroke, hypertension and a decline in eGFR are independent risk factors of high WMH load in patients with acute ischemic stroke. The risk factors of high PWMH and DWMH load are different. Attention should be paid to explore more risk factors of WMH and controlling them to improve the prognosis of acute ischemic stroke.

Key words: Ischemic stroke, Leukoaraiosis, Magnetic resonance imaging, Risk factors, Logistic models