Long-term air pollution exposure as a new predictor of postoperative delirium and the establishment of machine learning model

Introduction Postoperative delirium (POD) is a commonly occurring condition in the postoperative period. Studies have shown that air pollution may be related to cognition. Therefore, the study intends to investigate the relationship between air pollution and POD. Methods The annual exposure of the population to fine particulate matter (PM 2.5 ), ground-level ozone (O 3 ) and nitrogen dioxide (NO 2 ) was estimated at areas of residence. Logistic regression equations were applied to examine the risk factors linked to POD. Mediation effect analyses explored the role of cerebrospinal fluid (CSF) biomarkers between air pollution and POD. Machine learning method was used to construct the prediction model. and the interpretability of the models was evaluated by Shapley additive interpretation (SHAP). Results The prevalence of POD was 7.24%. The results of the logistic regression analysis indicated that PM 2.5 ( OR =1.052, 95% CI =1.004–1.102, P = 0.033) have been identified as a independent risk factor for POD. Mediation analysis demonstrated that the relationship between PM 2.5 and POD was influenced by the mediating effect of T-tau, which mediated approximately 11% across all study patients, while it reached 13.51% in the elderly patients. Five risk factors were included in nine machine learning models, among which the support vector machine (SVM) model was the best. SHAP plots of SVM showed that education, age and PM 2.5 were significant predictors of POD. Conclusions In conclusion, PM2.5 may be a risk factor for POD. This study further developed a reliable SVM model for predicting the occurrence of POD; education, age, and PM2.5 were significant predictors of POD. Clinical Trial Registration https://www.chictr.org.cn/showproj.html?proj=54473 , ClinicalTrials.gov [ChiCTR2000033439].

Authors

Institutions

Publication Details

Journal
Frontiers in Medicine
Published
2026-11-19
DOI
https://doi.org/10.3389/fmed.2026.1941645
Primary Topic
Intensive Care Unit Cognitive Disorders
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Long-term air pollution exposure as a new predictor of postoperative delirium and the establishment of machine learning model

Hongyan Gong, Yanlin Bi, Chen ChunXia, Yuanlong Wang et al.
Frontiers in Medicine
Intensive Care Unit Cognitive Disorders
article

Long-term air pollution exposure as a new predictor of postoperative delirium and the establishment of machine learning model

Hongyan Gong, Yanlin Bi, Chen ChunXia, Yuanlong Wang, X Wang, Wenjie Kong, Xu Lin, Yongxin Liang, Jiayi Chen, Bin Wang
article en

Abstract

Introduction Postoperative delirium (POD) is a commonly occurring condition in the postoperative period. Studies have shown that air pollution may be related to cognition. Therefore, the study intends to investigate the relationship between air pollution and POD. Methods The annual exposure of the population to fine particulate matter (PM 2.5 ), ground-level ozone (O 3 ) and nitrogen dioxide (NO 2 ) was estimated at areas of residence. Logistic regression equations were applied to examine the risk factors linked to POD. Mediation effect analyses explored the role of cerebrospinal fluid (CSF) biomarkers between air pollution and POD. Machine learning method was used to construct the prediction model. and the interpretability of the models was evaluated by Shapley additive interpretation (SHAP). Results The prevalence of POD was 7.24%. The results of the logistic regression analysis indicated that PM 2.5 ( OR =1.052, 95% CI =1.004–1.102, P = 0.033) have been identified as a independent risk factor for POD. Mediation analysis demonstrated that the relationship between PM 2.5 and POD was influenced by the mediating effect of T-tau, which mediated approximately 11% across all study patients, while it reached 13.51% in the elderly patients. Five risk factors were included in nine machine learning models, among which the support vector machine (SVM) model was the best. SHAP plots of SVM showed that education, age and PM 2.5 were significant predictors of POD. Conclusions In conclusion, PM2.5 may be a risk factor for POD. This study further developed a reliable SVM model for predicting the occurrence of POD; education, age, and PM2.5 were significant predictors of POD. Clinical Trial Registration https://www.chictr.org.cn/showproj.html?proj=54473 , ClinicalTrials.gov [ChiCTR2000033439].

Frontiers in MedicineVol. 13
Qingdao University (CN), Qingdao Municipal Hospital (CN), Peking University People's Hospital (CN)
Intensive Care Unit Cognitive Disorders
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon