血清脂蛋白相关磷脂酶A2预测大动脉粥样硬化性卒中患者转归的价值

Predictive value of serum iipoprotein-associated phospholipase A2 for the outcome in patients with large atherosclerotic stroke

目的 探讨血清脂蛋白相关磷脂酶A2（lipoprotein-associated phospholipase A2, Lp-PLA2）与大动脉粥样硬化性卒中（large artery atherosclerotic stroke, LAA）患者转归的相关性.方法 回顾性纳入2015年3月至2018年1月连云港市第二人民医院收治的LAA患者.发病后90 d时采用改良Rankin量表进行转归评价,0~2分定义为转归良好.应用多变量logistic回归分析确定转归不良的独立危险因素.应用受试者工作特征（receiver operating characteristic, ROC ）曲线分析Lp-PLA2对临床转归的预测价值.结果 共纳入121例LAA患者,其中男性64例（52.9%）,女性57例（47.1%）,年龄（63.5 ±9.5）岁;72例（59.5%）转归良好,49例（40.5%）转归不良.转归良好组糖尿病患者构成比（26.4%对65.3%;χ2=18.110,P〈0.001）以及糖化血红蛋白[（6.39 ±2.33）%对（7.58 ±3.12）%;t=1.663,P=0.041]、基线美国国立卫生研究院卒中量表（National Institutes of Health Stroke Scale, NIHSS）评分[5（3~6）分对10（7~14）分;Z= -7.498,P〈0.001]和Lp-PLA2[（194.7 ±84.3）μg/L对（291.4 ±82.6）μg/L;t= -5.447,P〈0.001]与转归不良组差异有统计学意义.多变量logistic回归分析显示,糖尿病[优势比（ odds ratio, OR） 1.215,95%可信区间（ confidence interval, CI） 1.102~1.601;P=0.046]、糖化血红蛋白（OR 2.275,95% CI 1.065~4.865;P=0.037）、基线 NIHSS 评分（OR 2.113,95% CI 1.585~ 2.734;P=0.015）和Lp-PLA2（OR 5.183,95% CI 3.203~8.134;P〈0.001）为LAA患者转归不良的独立危险因素. ROC分析显示,Lp-PLA2预测转归不良的曲线下面积为0.792（95% CI 0.713~0.872）;最佳截断值为260.5 μg/L,预测转归不良的敏感性为79.6%,特异性为84.7%.结论 血清Lp-PLA2水平较高是LAA患者转归不良的独立预测因素,其对转归不良的预测价值较高.

Objective To investigate the predictive value of serum lipoprotein-associated phospholipase A2（Lp-PLA2） for the outcomes in patients with large atherosclerotic stroke （LAA）. Methods Patients with LAA admitted to the Second People＇s Hospital of Lianyungang from March 2015 to January 2018 were enrolled retrospectively. The outcomes were evaluated by the modified Rankin Scale at 90 d after onset, 0-2 was defined as good outcome. Multivariate logistic regression analysis was used to identify the independent risk factors for poor outcome. The receiver operating characteristic （ROC） curve was used to analyze the predictive value of Lp-PLA2for outcomes. Results A total of 121 patients with LAA were enrolled, including 64 males （52.9%） and 57 females （47.1%）, aged 63.5 ±9.5 years; 72 （59.5%） had good outcome and 49 （40.5%） had poor outcome. The differences were statistically significant in the proportion of diabetic patients （26.4% vs.65.3%; χ2=18.110, P〈0.001） and glycated hemoglobin （ 6.39% ±2.33% vs. 7.58% ±3.12%; t=1.663, P=0.041）, baseline National Institutes of Health Stroke Scale （NIHSS） score （5 [3-6] vs.10[7 -14]; Z= -7.498, P〈 0.001）, and Lp-PLA2（194.7 ±84.3 μg/L vs.291.4 ± 82.6 μg/L; t= -5.447, P〈0.001） between the good outcome group and the poor outcome group. Multivariate logistic regression analysis showed that diabetes （ odds ratio [ OR] 1.215, 95% confidence interval [CI] 1.102-1.601; P=0.046）, glycosylated hemoglobin （ OR 2.275, 95% CI 1.065-4.865; P=0.037）, baseline NIHSS score （ OR 2.113, 95% CI 1.585-2.734; P=0.015）, and Lp-PLA2（OR 5.183, 95% CI 3.203-8.134; P〈0.001） were the independent risk factors for poor outcomes in patients with LAA. ROC analysis showed that the area under the curve of Lp-PLA2predicting poor outcome was 0 .792 （95% CI 0.713-0.872）; the optimal cut-off value was 260.5 μg/L, the sensitivity for predicting poor outcome was 79.6%, and the specificity was 84.7%. Conclusion The higher serum Lp-PLA2level is an independent predictive factor for poor outcome in patients with LAA. It has a higher predictive value for poor outcome.