Clinical profiles and mortality risk modeling in neurocritical care: A retrospective cohort study from high-altitude...

Neurocritical care patients residing in high-altitude regions such as the Xizang Autonomous Region encounter distinct physiological and logistical challenges, including hypobaric hypoxia, limited medical infrastructure, and delayed access to specialized care. Despite these factors, few studies have systematically characterized their clinical profiles or developed region-specific mortality prediction models. This study aimed to investigate the clinical features of neurocritical care patients admitted to a tertiary hospital in Lhasa, Xizang Autonomous Region, and to construct a predictive model for in-hospital mortality tailored to this high-altitude population.

This single-center retrospective study included 400 neurocritical care patients admitted between October 2021 and December 2023. Demographic data, types of neurological disorders, clinical scores and comorbidities, laboratory examination, pharmacological treatments and blood product transfusion conditions were collected. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of in-hospital mortality. A nomogram-based prediction model was developed and internally validated using bootstrap resampling. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis (DCA).

The overall in-hospital mortality rate was 15.8%. Multivariate analysis identified younger age (95% CI: 0.95-0.97, p = 0.013), higher APACHE II scores (95% CI: 1.02-1.16, p = 0.007), Glasgow Coma Scale ≤8 (95% CI: 1.17-4.56, p = 0.016), and elevated APTT levels (95% CI: 1.00-1.02, p = 0.007) as independent risk factors for mortality. Diuretic use was associated with a protective effect (95% CI: 0.20-0.77, p = 0.007).The final model demonstrated good discrimination (AUC = 0.749), calibration, and clinical utility. The nomogram provides a practical tool for early risk stratification in resource-limited high-altitude neurocritical care settings.

Neurocritical care patients in Xizang exhibit distinct clinical profiles and mortality risks. The developed prediction model may assist clinicians in early identification of high-risk patients and guide targeted interventions to improve outcomes in high-altitude regions.