Predictive value of geriatric nutritional risk index in stroke-associated pneumonia
10.3969/j.issn.1009-0126.2023.12.019
- VernacularTitle:老年营养风险指数对急性缺血性脑卒中患者肺部感染的预测价值
- Author:
Jianmo LIU
1
;
Jingyi LI
;
Haowen LUO
;
Pengfei YU
;
Yongsen CHEN
;
Bin WU
;
Yingping YI
;
Kai WANG
Author Information
1. 330000 南昌大学第二附属医院科技处
- Keywords:
stroke;
forecasting;
pneumonia;
retrospective studies;
geriatric nutritional risk index
- From:Chinese Journal of Geriatric Heart Brain and Vessel Diseases
2023;25(12):1310-1314
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate the predictive value of geriatric nutritional risk index(GNRI)for stroke-associated pneumonia in elderly patients with acute ischemic stroke(AIS).Methods A total of 1505 elderly patients with AIS admitted to Department of Neurology of the Second Affili-ated Hospital of Nanchang University from January 2017 to October 2022 were included in this retrospective study.According to GNRI nutritional assessment,they were divided into T1(high nutritional risk,GNRI＜82,n=49),T2(moderate nutritional risk,GNRI 82-91,n=305),T3(low nutritional risk,GNRI 92-98,n=555),and T4(no nutritional risk,GNRL＞98,n=596)groups.Additionally,based on the discharge diagnosis,they were further classified into pulmonary infection group(150 cases)and non-infection group(1355 cases).These subjects were also ran-domly assigned into training,validation,and testing sets in a ratio of 16∶4∶5.Multivariate logis-tic regression analysis was performed to identify the risk factors for pulmonary infection in stroke patients.Logistic regression and XGBoost algorithms were used to establish prediction models for pulmonary infection.The models were evaluated with their AUC value,accuracy,sensitivity,and specificity based on ROC curve analysis.Results Multivariate logistic regression analysis revealed that hypertension,invasive procedures,consciousness disorders,CRP,lymphocyte count,hemoglo-bin and GNRI were independent risk factors for pulmonary infection in stroke patients(P＜0.05).The AUC value of the GNRI model for predicting pulmonary infection in the testing set was 0.742(95％CI:0.651-0.833),with an accuracy of 71.8％,sensitivity of 76.7％,and specificity of 71.2％.The combined model of clinical indicators(hypertension,invasive procedures,conscious-ness disorders,CRP,lymphocyte count,hemoglobin)and GNRI achieved an AUC value of 0.776(95％CI:0.700-0.853),accuracy of 74.8％,sensitivity of 83.3％,and specificity of 73.8％in the test set.Conclusion GNRI is an independent risk factor for pulmonary infection in elderly pa-tients with AIS and has a certain value in predicting pulmonary infection after AIS.
