重症脑卒中并发肺部感染的影响因素分析及预测模型构建-现代医学

重症脑卒中并发肺部感染的影响因素分析及预测模型构建

单位：1. 中国人民解放军海军第九七一医院神经外科, 山东青岛 266071;
2. 潍坊市中医院脑病科, 山东潍坊 261041

分类号：R743.3

出版年·卷·期（页码）：2026·45·第四期（682-690）

摘要：

目的： 探讨重症脑卒中并发肺部感染(PI)的影响因素并构建预测模型。方法： 选取2020年10月至2023年1月中国人民解放军海军第九七一医院与潍坊市中医院重症脑卒中患者206例作为建模集，依据住院期间是否发生PI分为感染组、非感染组。比较两组患者基线资料、格拉斯哥昏迷评分(GCS)、改良版Beck口腔评估量表(MBOAS)评分、血红蛋白-白蛋白-淋巴细胞-血小板评分(HALP)。采用单因素、最小绝对值收敛和选择算子(LASSO)回归及多因素Logistic回归分析探讨重症脑卒中患者并发PI的独立影响因素，建立PI的预测模型。绘制受试者工作特征(ROC)曲线探讨模型的区分度，校准曲线评价校准度，临床决策分析曲线(DCA)量化不同阈值概率下的净收益以评估模型的临床效用。另选2023年2月至2024年12月的100例患者作为验证集(感染组20例，非感染组80例)进行验证。结果： 在建模集数据中，感染组年龄校正的查尔森合并症指数(aCCI)、美国国立卫生院卒中量表(NIHSS)、MBOAS评分及机械通气、吞咽困难患者占比显著高于非感染组，HALP、GCS评分低于非感染组(P＜0.05)。LASSO回归获得6个非零系数变量，分别为aCCI、NIHSS评分、机械通气、HALP、GCS、MBOAS评分；多因素Logistic分析显示，机械通气、HALP、GCS评分、MBOAS评分是重症脑卒中并发PI的独立相关因素(P＜0.05)；ROC曲线显示，建立的重症脑卒中并发PI预警模型的曲线下面积(AUC)为0.952，内部验证获取一致性指数为0.957。Hosmer-Lemeshow χ²=6.688，P=0.571，模型平均绝对误差为0.008。决策曲线显示，在阈概率0.03～0.92，运用该模型预测PI能获得较好临床净获益。在验证集数据中，模型的AUC为0.927，与建模集比较无明显差异(P＞0.05)，提示联合模型稳定性良好。结论： 基于机械通气、HALP、GCS、MBOAS评分构建的重症脑卒中并发PI预警模型可有效识别潜在高风险患者群体，为临床风险评估及早期预防性干预提供参考。

Objective: To explore the influencing factors of pulmonary infection(PI) in patients with severe stroke and to construct a predictive model. Methods: A total of 206 patients with severe stroke admitted to No. 971 Hospital of PLA Navy and Weifang Hospital of Traditional Chinese Medicine from October 2020 to January 2023 were enrolled as modeling set. According to whether PI occurred during hospitalization, patients were assigned to infection group or non-infection group. Baseline data, Glasgow coma scale score(GCS), modified Beck oral assessment scale(MBOAS) score, and hemoglobin-albumin-lymphocyte-platelet score(HALP) were compared between the two groups. Univariate analysis, least absolute shrinkage and selection operator(LASSO) regression, and multivariate logistic regression analysis were used to explore the independent influencing factors of PI in patients with severe stroke, and a predictive model for PI was established. Receiver operating characteristic(ROC) curve was plotted to explore the discriminative ability of the model, calibration curve was used to evaluate the calibration, and decision curve analysis(DCA) was performed to quantify the net benefit under different threshold probabilities to assess the clinical utility of the model. Additionally, 100 patients from February 2023 to December 2024 were selected as validation set(20 cases in the infection group and 80 cases in the non-infection group) for validation. Results: In the modeling set, the age-adjusted Charlson comorbidity index(aCCI)， national institute of health stroke scale(NIHSS) score, MBOAS scores, and the proportions of patients with mechanical ventilation and dysphagia in the infection group were significantly higher than those in the non-infection group, while HALP and GCS were significantly lower than those in the non-infection group(P<0.05). LASSO regression yielded six variables with non-zero coefficients: aCCI, NIHSS score, mechanical ventilation, HALP, GCS, and MBOAS scores. Multivariate Logistic analysis showed that mechanical ventilation, HALP, GCS, and MBOAS scores were independent risk factors for PI in patients with severe stroke(P<0.05). The ROC curve showed that the area under the curve(AUC) of the established predictive model for PI in patients with severe stroke was 0.952, and the concordance index obtained from internal validation was 0.957. The Hosmer-Lemeshow χ²=6.688, P=0.571, and the mean absolute error of the model was 0.008. The decision curve showed that using this model to predict PI could obtain good clinical net benefit when the threshold probability ranged from 0.03 to 0.92. In the validation set, the AUC of the model was 0.927, with no significant difference compared to that of the modeling set(P>0.05), indicating good stability of the combined model. Conclusion: The predictive model for PI in patients with severe stroke constructed based on mechanical ventilation, HALP, GCS, and MBOAS scores can effectively identify potential high-risk patient populations, providing a reference for clinical risk assessment and early preventive intervention.

参考文献：

[1] 中华医学会神经病学分会,中华医学会神经病学分会脑血管病学组.中国重症卒中管理指南2024[J].中华神经科杂志,2024,57(7):698-714.
[2] 兰雅智,刘颖,赵荣荣.重症脑卒中机械通气患者非计划性拔管预测因素[J].现代医学,2025,53(8):1272-1277.
[3] FEIGIN V L,BRAININ M,NORRVING B,et al.World stroke organization:global stroke fact sheet 2025[J].Int J Stroke,2025,20(2):132-144.
[4] KUMAR S,CHOU S H,SMITH C J,et al.Addressing systemic complications of acute stroke:a scientific statement from the American Heart Association[J].Stroke,2025,56(1):e15-e29.
[5] WANG J,JIANG P,HUANG Y,et al.Prognostic value of the cutoffs for HALP in endometrial cancer[J].Am J Clin Oncol,2023,46(3):107-113.
[6] BARON J C.Fifty years of deciphering stroke pathophysiology[J].Stroke,2025,56(7):1947-1957.
[7] SEITZ A,RAVAL A P.Menstruation:an important indicator for assessing stroke risk and its outcomes[J].Stroke,2025,56(2):533-542.
[8] 谢瑞怡.基于循证的集束化护理对预防脑卒中鼻饲患者吸入性肺炎的研究[D].南京:南京中医药大学,2020.
[9] 中华医学会神经病学分会,中华医学会神经病学分会脑血管病学组.中国急性缺血性卒中诊治指南2023[J].中华神经科杂志,2024,57(6):523-559.
[10] DE RUBEIS G,CHATURVEDI S,KAMEL H,et al.Heterogeneity in measurement of NIHSS:a systematic review and meta-analysis[J].Neurol Sci,2025,46(1):227-237.
[11] JAHANSHIR M,NOBAHAR M,GHORBANI R,et al.Effect of clove mouthwash on the incidence of ventilator-associated pneumonia in intensive care unit patients:a comparative randomized triple-blind clinical trial[J].Clin Oral Investig,2023,27(7):3589-3600.
[12] MENON D K,SILVERBERG N D,FERGUSON A R,et al.Clinical assessment on days 1-14 for the characterization of traumatic brain injury:recommendations from the 2024 NINDS traumatic brain injury classification and nomenclature initiative clinical/symptoms working group[J].J Neurotrauma,2025,42(13-14):1038-1055.
[13] HWANG D Y,OCZKOWSKI S J W,LEWIS K,et al.Society of critical care medicine guidelines on family-centered care for adult ICUs:2024[J].Crit Care Med,2025,53(2):e465-e482.
[14] BAI J,ZHAO Y,WANG Z,et al.Stroke-associated pneumonia and the brain-gut-lung axis:a systematic literature review[J].Neurologist,2025,30(4):237-250.
[15] CAI J X,HUANG M Q,ZHOU W Q,et al.Effect of mild moxibustion at Jiuwei(CV15) on the diaphragm and postural control ability of patients after recovery from stroke complicated with pulmonary infection:a randomized controlled trial[J].Acupunct Res,2025,50(4):426-432.
[16] COELHO P M,ALMEIDA P L,CARMEZIM I,et al.Mortality and pulmonary complications of post-stroke dysphagia:a casuistic review of an acute stroke unit[J].Cureus,2024,16(12):e74993.
[17] LIU X C,CHANG X J,ZHAO S R,et al.Identification of risk factors and construction of a nomogram predictive model for post-stroke infection in patients with acute ischemic stroke[J].World J Clin Cases,2024,12(20):4048-4056.
[18] 胡慧勤.改良版Beck口腔评估表指导下的口腔护理对卒中相关肺炎的预防效果[J].全科护理,2024,22(7):1301-1304.
[19] ABUJABER A A,YASEEN S,NASHWAN A J,et al.Prediction of stroke-associated hospital-acquired pneumonia:Machine learning approach[J].J Stroke Cerebrovasc Dis,2025,34(2):108200.
[20] FAN X,XU J,YE R,et al.Identification of risk factors and development of prediction models for Glasgow Coma Scale score deterioration in stroke patients:a study based on the MIMIC-IV database[J].J Intensive Care Med,2025,40(9):937-947.
[21] PADWALE V,CHIVATE C,KIRNAKE V,et al.Comparative prognostic value of the national institutes of health stroke scale(NIHSS) and the Glasgow Coma Scale(GCS) in supratentorial and infratentorial stroke patients in western India[J].Cureus,2024,16(7):e65778.
[22] VREEBURG R J G,VAN LEEUWEN F D,MANLEY G T,et al.Validation of the GCS-pupil scale in traumatic brain injury:incremental prognostic value of pupillary reactivity with GCS in the prospective observational cohorts CENTER-TBI and TRACK-TBI[J].J Neurotrauma,2025,42(9-10):798-813.
[23] VAN VALBURG M K,TERMORSHUIZEN F,GEERTS B F,et al.Predicting 30-day mortality in intensive care unit patients with ischaemic stroke or intracerebral haemorrhage[J].Eur J Anaesthesiol,2024,41(2):136-145.
[24] TUN T Z,HAN S M,MOJI K,et al.A descriptive study of stroke types,risk factors,clinical features,and outcomes in a tertiary hospital in Myanmar[J].Trop Med Health,2024,52(1):26.
[25] CALIANDRO P,LENKOWICZ J,REALE G,et al.Artificial intelligence to predict individualized outcome of acute ischemic stroke patients:the SIBILLA project[J].Eur Stroke J,2024,9(4):1053-1062.
[26] ÇOLAK M,ÇOBAN H,SARIOG∨LU N,et al.Could the HALP score serve as a biomarker of bronchiectasis exacerbation[J].Tuberk Toraks,2024,72(1):202401811.
[27] NERGIZ S,OZTURK U.The effect of prognostic nutritional index on infection in acute ischemic stroke patients[J].Medicina,2023,59(4):679.
[28] HUANG L F,ZHU M L,YE Y R.Association of nutritional indices and prognosis of stroke patients:a systematic review and meta-analysis[J].Eur Rev Med Pharmacol Sci,2023,27(12):5803-5811.
[29] YANG X,SUN H,HOU S,et al.The impact of nutritional and inflammatory status on mortality in stroke patients:Results from NHANES 2005-2018[J].J Stroke Cerebrovasc Dis,2025,34(7):108334.
[30] XU M,CHEN L,HU Y,et al.The HALP(hemoglobin,albumin,lymphocyte,and platelet) score is associated with early-onset post-stroke cognitive impairment[J].Neurol Sci,2023,44(1):237-245.
[31] OZTURK U,NERGIZ S,OZTURK O.The association between HALP score and infection in acute ischemic stroke patients[J].J Stroke Cerebrovasc Dis,2024,33(11):107929.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】

提示：您还未登录，请登录！点此登录