Work

A Two-Variable Model Combining the Admission Blood Glucose Level and APACHE II Score for Predicting Weaning Success in Patients with Moderate-to-Severe Traumatic Brain Injury: A Single-Center Retrospective Cohort Study in China

Year: 2026
Type: article
Study design: Observational Study
Abstract: Background: Liberation from invasive mechanical ventilation (MV) remains challenging in patients with moderate-to-severe traumatic brain injury (TBI). Early identification of patients at risk of weaning failure may facilitate individualized management. Purpose: To evaluate whether admission blood glucose levels and the Acute Physiology and Chronic Health Evaluation (APACHE II) score, alone and in combination, were associated with successful liberation from invasive MV and to assess the predictive performance of the combined model. Patients and methods: This single-center retrospective cohort study included 126 adults with moderate-to-severe TBI admitted between January 2021 and May 2025. Successful weaning was defined as ≥48 consecutive hours of spontaneous breathing without invasive MV support. Data collected within 24 hours of admission were analyzed via univariate and multivariate logistic regression. Predictive performance was compared using receiver operating characteristic (ROC) curves and the DeLong test. The optimism-corrected area under the ROC curve (AUC) was estimated using bootstrap internal validation. Results: Fifty-seven patients (45.2%) achieved successful weaning. In multivariate analysis, higher admission glucose levels and APACHE II scores remained independently associated with lower odds of successful weaning. The combined model had an apparent AUC of 0.848 (95% CI, 0.779-0.918), which was significantly higher than that of glucose alone but not significantly different from that of the APACHE II score alone. Bootstrap internal validation yielded an optimism-corrected AUC of 0.847. Conclusion: Higher admission blood glucose levels and APACHE II scores were independently associated with lower odds of successful liberation from invasive MV in patients with moderate-to-severe TBI. The combined model showed good apparent discriminative performance (AUC, 0.848) but did not significantly outperform the APACHE II score alone, and the estimate may be optimistic. Prospective multicenter validation is warranted before clinical use.
Language: English
Field: Medicine
Open Access status: gold
Estimated APC paid: $2,836
OpenAlex ID: W7225252503
FWCI:
Cites: 37
Cited by:
Related to: 10