Prognostic prediction model construction and validation based on regional cerebral oxygen saturation and bispectral index monitoring in pediatric traumatic brain injury
- 1PICU, Qinghai Provincial Women and Children's Hospital
Abstract
Background: Continuous multimodal cerebral monitoring may improve detection and management of secondary injury after pediatric moderate-to-severe traumatic brain injury (msTBI). We developed and validated a prognostic model for 6-month functional outcome integrating regional cerebral oxygen saturation (rScO₂), bispectral index (BIS), clinical and biomarker data.
Methods: Prospective cohort of 93 children with msTBI monitored with rScO₂ and BIS (monitoring group) and a retrospective control cohort of 66 children without such monitoring. Primary outcome was Glasgow Outcome Scale (GOS) at 6 months. Multivariate logistic regression identified independent predictors; a nomogram was constructed and internally validated using receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA).
Results: Total N = 159. Monitoring-group patients had shorter hospital and pediatric intensive care unit (PICU) lengths of stay and shorter invasive ventilation duration than controls ( p < 0.05). Multivariate analysis identified admission Glasgow Coma Scale (GCS) [odds ratio (OR) 1.939, 95% confidence interval (CI): 1.261 ~ 2.984, p = 0.003], mean rScO₂ (OR 1.121, 95%CI: 1.027 ~ 1.225, p = 0.011), mean BIS (OR 1.115, 95%CI: 1.041 ~ 1.193, p = 0.002) and neuron-specific enolase (NSE) (OR 0.843, 95%CI: 0.769 ~ 0.924, p < 0.001) as independent outcome-associated variables. The nomogram achieved AUC 0.91 (95%CI: 0.84–0.97), with favorable calibration and net clinical benefit on DCA.
Conclusion: A prognostic model combining admission GCS, mean rScO₂, mean BIS and NSE provides accurate individual risk estimation for pediatric msTBI and may assist bedside risk stratification and management decisions.
Related
- Construction and validation of a nomogram prediction model for the risk of delirium after hip arthroplasty in elderly patients
- Risk factors for postoperative recurrence of chronic subdural hematoma and construction and validation of an online dynamic nomogram prediction model.
- Development and validation of a prognostic prediction model for elderly patients with sepsis-associated acute kidney injury based on interpretable machine learning
- Development and validation of novel prognostic hepatic-cardiac index and hemodynamic prediction models in pre-capillary pulmonary hypertension patients
- Development and validation of a multiphase CT radiomics-based machine learning model for prognostic prediction in sarcomatoid renal cell carcinoma
BibTeX
@article{li2026prognostic,
title = {{Prognostic prediction model construction and validation based on regional cerebral oxygen saturation and bispectral index monitoring in pediatric traumatic brain injury}},
author = {Baowen Li and Ruhai Ma and Zhuyun Tian and Youcang Wan and Xiaoli Zhong and Wenhai Wang and Weixiao Yan and Shengkui Zhao},
journal = {Frontiers in Neuroscience},
year = {2026},
month = oct,
doi = {10.3389/fnins.2026.1866247},
url = {https://www.frontiersin.org/articles/10.3389/fnins.2026.1866247/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Prognostic prediction model construction and validation based on regional cerebral oxygen saturation and bispectral index monitoring in pediatric traumatic brain injury
AU - Baowen Li
AU - Ruhai Ma
AU - Zhuyun Tian
AU - Youcang Wan
AU - Xiaoli Zhong
AU - Wenhai Wang
AU - Weixiao Yan
AU - Shengkui Zhao
PY - 2026
DA - 2026/10/16
JO - Frontiers in Neuroscience
DO - 10.3389/fnins.2026.1866247
UR - https://www.frontiersin.org/articles/10.3389/fnins.2026.1866247/full
AB - Background: Continuous multimodal cerebral monitoring may improve detection and management of secondary injury after pediatric moderate-to-severe traumatic brain injury (msTBI). We developed and validated a prognostic model for 6-month functional outcome integrating regional cerebral oxygen saturation (rScO₂), bispectral index (BIS), clinical and biomarker data. Methods: Prospective cohort of 93 children with msTBI monitored with rScO₂ and BIS (monitoring group) and a retrospective control cohort of 66 children without such monitoring. Primary outcome was Glasgow Outcome Scale (GOS) at 6 months. Multivariate logistic regression identified independent predictors; a nomogram was constructed and internally validated using receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA). Results: Total N = 159. Monitoring-group patients had shorter hospital and pediatric intensive care unit (PICU) lengths of stay and shorter invasive ventilation duration than controls ( p < 0.05). Multivariate analysis identified admission Glasgow Coma Scale (GCS) [odds ratio (OR) 1.939, 95% confidence interval (CI): 1.261 ~ 2.984, p = 0.003], mean rScO₂ (OR 1.121, 95%CI: 1.027 ~ 1.225, p = 0.011), mean BIS (OR 1.115, 95%CI: 1.041 ~ 1.193, p = 0.002) and neuron-specific enolase (NSE) (OR 0.843, 95%CI: 0.769 ~ 0.924, p < 0.001) as independent outcome-associated variables. The nomogram achieved AUC 0.91 (95%CI: 0.84–0.97), with favorable calibration and net clinical benefit on DCA. Conclusion: A prognostic model combining admission GCS, mean rScO₂, mean BIS and NSE provides accurate individual risk estimation for pediatric msTBI and may assist bedside risk stratification and management decisions.
SN - 1662-453X
SN - 1662-4548
ER -