动脉瘤性蛛网膜下腔出血后发生卒中相关性肺炎的预测模型
A Prediction Model for Predicting Stroke-Associated Pneumonia in Aneurysmal Subarachnoid Hemorrhage Patients

背景：卒中相关性肺炎(SAP)是动脉瘤性蛛网膜下腔出血(aSAH)患者常见但可以预防的并发症，已被证实与aSAH患者的不良预后和较长的住院时间有关。本研究旨在探讨aSAH患者SAP的危险因素并建立预测模型。方法：回顾性的收集318例在青岛市市立医院神经外科接受治疗的aSAH患者的临床资料。采用单因素分析和多因素logistic回归分析探讨SAP的危险因素。并使用logistic回归方法建立SAP的预测模型。通过计算ROC曲线下面积(AUC)来评价模型的准确性，并通过Bootstrap自助采样法进行模型内部验证。结果：63例aSAH患者发生SAP，发生率19.8%。多因素Logistic回归分析显示，年龄 > 65岁(OR = 3.17, P = 0.002)、吸烟史(OR = 2.824, P = 0.026)、世界神经外科学会联合会(WFNS)分级 ≥ 4 (OR = 2.465, P = 0.029)、改良Fisher分级 ≥ 3(OR = 3.114, P = 0.01)、白细胞(OR = 1.176, P < 0.001)是aSAH患者SAP的独立危险因素。由这5个因素组成的模型AUC为0.856，内部验证中该模型的平均AUC为0.856，95%置信区间为：0.854~0.858。结论：高龄、有吸烟史、高WFNS分级、高改良Fisher分级、入院时白细胞升高是aSAH患者SAP的独立危险因素。结合这5个因素的新预测模型有利于临床医生在入院时评估患者发生SAP的风险，并及时调整治疗策略以预防aSAH患者SAP的发生。
Background: Stroke-associated pneumonia (SAP) is a common but preventable complication in pa-tients with aneurysmal subarachnoid hemorrhage (aSAH). It has been proved to be related to the poor prognosis and long hospital stay of aSAH patients. The purpose of this study was to explore the risk factors of SAP in patients with aSAH and establish a predictive model. Methods: The clinical data of 318 patients with aSAH treated in the Department of Neurosurgery of Qingdao Municipal Hospital were collected retrospectively. Univariate analysis and multivariate logistic regression analysis were used to explore the risk factors of SAP. The logistic regression method was used to establish the prediction model of SAP. The accuracy of the model is evaluated by calculating the ar-ea under the ROC curve (AUC), and the internal verification of the model is carried out by Bootstrap Method. Results: 63 aSAH patients developed SAP with incidence of 19.8%. Multivariate logistic re-gression analysis showed age > 65 (0R = 3.17, P = 0.002), smoking history (0R =2.824, P = 0.026), World Federation of Neurosurgical Societies (WFNS) score ≥ 4 (0R = 2.465, P = 0.029), the modified fisher scale ≥ 3 (0R = 3.114, P = 0.01) and white blood cell count (0R = 1.176, P < 0.001) were inde-pendent risk factors of pneumonia in aSAH patients. Consisted of these five factors, the constructed model was valuable in predicting SAP with AUC of 0.856 and 95% CI is 0.854~0.858. Conclusion: Advanced age, smoking history, high WFNS score, high modified Fisher scale and white blood cell count are independent risk factors for SAP in patients with aSAH. The new predictive model com-bined with these five factors is helpful for clinicians to evaluate the risk of SAP on admission and adjust treatment strategies in time to prevent the occurrence of SAP in patients with aSAH.