Trajectory-Based Modeling Improves Long-Term Outcome Prediction Beyond 90-Day Assessment in ICU-Treated Intracerebral Hemorrhage
Revista de Neurología · 2026
DOI: 10.31083/rn52441 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista de Neurología, 2026. Verfasst von Qing Su und 4 weiteren Autor:innen. Beteiligte Einrichtung: Emergency Medicine Department, The First Affiliated Hospital of Jinan University, 510632 Guangzhou, Guangdong, China. Hauptthemen laut MEDLINE: Cerebral Hemorrhage. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective:To identify distinct long-term functional recovery trajectories among intensive care unit (ICU)–treated patients with spontaneous intracerebral hemorrhage (ICH) and to evaluate whether trajectory-based classification improves prediction of long-term quality of life (QoL) compared with conventional single time-point functional assessment.Methods:We conducted a prospective longitudinal cohort study of consecutive adult patients with spontaneous ICH admitted to a tertiary ICU between January 2024 and January 2025. Survivors to hospital discharge were followed for 12 months. Functional status was assessed using the modified Rankin Scale (mRS) at hospital discharge, 90 days, 6 months, and 12 months. Latent class mixed modeling (LCMM) was used to identify distinct recovery trajectories. Health-related QoL at 12 months was measured using the EuroQol 5-Dimension 5-Level questionnaire (EQ-5D-5L) instrument. Multinomial logistic regression was performed to evaluate associations between early clinical characteristics and trajectory membership. Receiver operating characteristic curve analysis was conducted to compare the predictive performance of trajectory membership and 90-day mRS for 12-month QoL.Results:Among 196 patients (mean age 61.7 ± 12.4 years; 60.2% male), three distinct functional recovery trajectories were identified: good, moderate, and poor recovery. Twelve-month QoL demonstrated a graded decline across trajectory groups. Older age, lower admission Glasgow Coma Scale scores, and larger hematoma volume were independently associated with both moderate and poor recovery trajectories, while intraventricular hemorrhage, mechanical ventilation, and prolonged sedation (>48 hours) were specifically associated with the poor recovery trajectory. Trajectory-based classification demonstrated superior predictive performance for 12-month QoL compared with 90-day mRS (area under the curve [AUC], 0.81 vs 0.72; p = 0.003), with significant improvement in risk discrimination. Trajectory-based classification also improved risk reclassification (net reclassification improvement (NRI) = 0.27; integrated discrimination improvement (IDI) = 0.09).Conclusion:ICU-treated patients with intracerebral hemorrhage exhibit distinct long-term recovery trajectories that are strongly associated with quality of life. Importantly, trajectory-based classification provides incremental prognostic value beyond conventional single time-point functional assessment, supporting a shift toward dynamic outcome modeling for long-term prognostication in ICH.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Emergency Medicine Department, The First Affiliated Hospital of Jinan University, 510632 Guangzhou, Guangdong, China.
Zitieren
APA
Su, Q., Zhang, X., Wang, B., Zhu, L., & Li, Y. (2026). Trajectory-Based Modeling Improves Long-Term Outcome Prediction Beyond 90-Day Assessment in ICU-Treated Intracerebral Hemorrhage. Revista de Neurología. https://doi.org/10.31083/rn52441
Vancouver
Su Q, Zhang X, Wang B, Zhu L, Li Y. Trajectory-Based Modeling Improves Long-Term Outcome Prediction Beyond 90-Day Assessment in ICU-Treated Intracerebral Hemorrhage. Revista de Neurología. 2026. doi:10.31083/rn52441
BibTeX
@article{Su2026,
author = {Su, Qing and Zhang, Xini and Wang, Baogui and Zhu, Liyun and Li, Yufeng},
title = {Trajectory-Based Modeling Improves Long-Term Outcome Prediction Beyond 90-Day Assessment in ICU-Treated Intracerebral Hemorrhage},
journal = {Revista de Neurología},
year = {2026},
doi = {10.31083/rn52441}
}
EUVIMED