摘要目的:探讨入院时中性粒细胞/淋巴细胞比值（neutrophil-lymphocyte ratio, NLR）对急性缺血性卒中（acute ischemic stroke, AIS）患者卒中相关性肺炎（stroke-associated pneumonia, SAP）的预测价值。方法:回顾性纳入昆明医科大学第一附属医院2019年1月至2020年6月期间收治的AIS患者。收集患者人口统计学信息、血管危险因素、入院时卒中严重程度以及NLR等资料。应用多变量 logistic回归分析NLR与SAP的独立相关性。将NLR进行四分位数分组，进一步分析NLR与SAP之间的趋势关系。采用受试者工作特征（receiver operating characteristic, ROC）曲线评价NLR对SAP的诊断价值。 结果:共纳入316例AIS患者，男性200例（63.29%），年龄（63.86±13.78）岁，中位基线美国国立卫生研究院卒中量表评分为4分（四分位数间距：2~9分），中位NLR为4.42（四分位数间距：3.17~6.70）。93例（29.43%）发生SAP。多变量 logistic回归分析显示，NLR是AIS患者发生SAP的独立危险因素（优势比1.189，95%置信区间1.077~1.313； P<0.001）。而且，SAP风险随NLR升高呈上升趋势（ P趋势<0.001）；与第1四分位数相比，第4四分位数发生SAP的风险增高9.991倍（95%置信区间2.912~34.279； P<0.001）。ROC曲线分析显示，NLR预测SAP的曲线下面积为0.793（95%置信区间0.737~0.850），最佳截断值为5.475，此时预测SAP的敏感性和特异性分别为66.67%和79.82%。 结论:入院时NLR为AIS患者SAP的独立危险因素，对SAP有一定的预测价值。

abstractsObjective:To investigate the predictive value of neutrophil/lymphocyte ratio (NLR) at admission for stroke associated pneumonia (SAP) in patients with acute ischemic stroke (AIS).Methods:Patients with AIS admitted to the First Affiliated Hospital of Kunming Medical University from January 2019 to June 2020 were retrospectively included. The demographic information, vascular risk factors, severity of stroke at admission, and NLR data of the patients were collected. Multivariate logistic regression was used to analyze the independent correlation between NLR and SAP. The NLR was divided into quartile groups to further analyze the trend relationship between NLR and SAP. Receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of NLR for SAP. Results:A total of 316 patients with AIS were enrolled, including 200 males (63.29%) with an age of 63.86±13.78 years. The median baseline Nationanl Institutes of Health Stroke Scale score was 4 (interquartile range, 2-9), and the median NLR was 4.42 (interquartile range, 3.17-6.70). Ninety-three patients (29.43%) experienced SAP. Multivariate logistic regression analysis showed that NLR was an independent risk factor for SAP in patients with AIS (odds ratio 1.189, 95% confidence interval 1.077-1.313; P<0.001). Moreover, SAP risk increases with the increase of NLR ( Ptrend<0.001). Compared to the first quartile, the risk of SAP increased 9.991 times in the fourth quartile (95% confidence interval 2.912-34.279; P<0.001). ROC curve analysis showed that the area under the curve of NLR for SAP prediction was 0.793 (95% confidence interval 0.737-0.850), with an optimal cutoff value of 5.475. The sensitivity and specificity for predicting SAP were 66.67% and 79.82%, respectively. Conclusion:NLR at admission is an independent risk factor for SAP in patients with AIS and has certain predictive value for SAP.