Construction and verification of the prediction model of pulmonary infection in patients with aneurysmal subarachnoid hemorrhage after craniotomy
10.3760/cma.j.cn211501-20250113-00124
- VernacularTitle:基于LASSO回归的动脉瘤性蛛网膜下腔出血开颅夹闭术后患者肺部感染风险预测模型构建与验证
- Author:
Shufang SHI
1
;
Yanjun ZHANG
;
Mingxia GUO
;
Jingwen CHEN
;
Kexing JI
;
Xiaolong CHEN
;
Jing ZHAO
;
Xinmin DING
Author Information
1. 山西白求恩医院（山西医学科学院）山西医科大学第三医院 同济山西医院神经外科，太原 030032
- Publication Type:Journal Article
- Keywords:
Subarachnoid hemorrhage;
Infection;
Nomograms;
Craniotomy;
Prediction model
- From:
Chinese Journal of Practical Nursing
2025;41(34):2685-2693
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct and verify a risk prediction model for pulmonary infection in patients with aneurysmal subarachnoid hemorrhage (aSAH) after craniotomy and clipping, providing theoretical basis and practical guidance for improving the quality of postoperative care.Methods:Using the convenience sampling method, a retrospective selection was made of 397 patients with aSAH after craniotomy and clipping who were hospitalized in the Department of Neurosurgery of Shanxi Bethune Hospital (Shanxi Academy of Medical Sciences) from January 2019 to December 2023 as the modeling group. They were randomly divided into the training set and the test set at a ratio of 7:3, with 278 cases in the training set and 119 cases in the test set. Patients were divided into the infection group and the non-infection group based on whether they developed pulmonary infection. Univariate analysis was used to model the risk factors of pulmonary infection after aSAH craniotomy and clamping in the group, and Lasso regression was used to construct a predictive model. A total of 119 patients with aSAH admitted to the neurosurgery department of the same hospital from January to April 2024 were selected for the external validation of the model. The predictive effect of the model was evaluated through the receiver operating characteristic (ROC) curve.Results:In the modeling group, there were 216 male patients and 181 female patients. The incidence of pulmonary infection was 38.54% (153/397). Finally, five influencing factors, namely stroke, Hunt-Hess classification, mechanical ventilation, indwelling nasogastric tube and the timing of initiating enteral nutrition, were included to construct a predictive model. The areas under the ROC curves of the nomogram prediction models of this model in the training set, test set, and external validation group were 0.859(95% CI 0.791-0.928), 0.843(95% CI 0.796-0.890), and 0.800(95% CI 0.711-0.889), respectively. The calibration curve shows that the model's prediction fits well with the actual situation and has a high degree of calibration. Decision curve analysis indicates that this model has high clinical application value under different risk thresholds. Conclusions:The risk prediction model for pulmonary infection in patients after craniotomy and clipping with aSAH has good discrimination and calibration, which can provide reference for medical staff to identify high-risk patients at an early stage and take preventive intervention measures.
