From early warning to endotype-guided care in sepsis: A critical narrative review of AI, multi-omics, and clinical translation
Intensive Care Science · 2026
DOI: 10.1016/j.iics.2026.07.002 ↗Kurz gefasst
Fachartikel in der Zeitschrift Intensive Care Science, 2026. Verfasst von Shengze Wu und 5 weiteren Autor:innen. Die Arbeit nennt 240 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Zhejiang University, Wu Jieping Medical Foundation, Natural Science Foundation of Beijing Municipality.
Themen: Sepsis
Abstract (Originalsprache)
Sepsis is a highly heterogeneous syndrome characterized by dynamic host responses, which has limited the effectiveness of conventional “one-size-fits-all” therapeutic strategies. Precision medicine has therefore emerged as a promising approach for improving risk stratification, diagnosis, and individualized treatment throughout the clinical course of sepsis. This narrative review summarizes recent advances in artificial intelligence (AI), multi-omics technologies, pathogen identification, molecular endotyping, biomarker-guided management, and phenotype-directed therapeutic strategies. Relevant literature published up to January 15, 2026, was retrieved from PubMed, EMBASE, and the Cochrane Library, with priority given to randomized controlled trials, externally validated studies, and international guidelines. Current evidence suggests that AI-based early warning systems can identify patients at risk several hours before clinical deterioration, while metagenomic next-generation sequencing (mNGS) substantially improves pathogen detection despite challenges in distinguishing colonization from infection. Transcriptomic and proteomic endotypes have enhanced biological stratification and prognostic assessment, although prospective evidence supporting endotype-guided therapy remains limited. Overall, precision medicine offers a promising framework for addressing the biological heterogeneity of sepsis; however, its successful clinical implementation will require multicenter prospective validation, standardized molecular classification systems, cost-effective diagnostic platforms, and phenotype-enriched interventional trials.
Förderung
- Zhejiang University
- Wu Jieping Medical Foundation
- Natural Science Foundation of Beijing Municipality
- Key Research and Development Program of Zhejiang Province
- Natural Science Foundation of Zhejiang Province
- National Key Research and Development Program of China
- Fundamental Research Funds for the Central Universities
- National Natural Science Foundation of China
Zitieren
APA
Wu, S., Yang, S., Yang, J., Zhou, P., Shen, H., & Zhang, Z. (2026). From early warning to endotype-guided care in sepsis: A critical narrative review of AI, multi-omics, and clinical translation. Intensive Care Science. https://doi.org/10.1016/j.iics.2026.07.002
Vancouver
Wu S, Yang S, Yang J, Zhou P, Shen H, Zhang Z. From early warning to endotype-guided care in sepsis: A critical narrative review of AI, multi-omics, and clinical translation. Intensive Care Science. 2026. doi:10.1016/j.iics.2026.07.002
BibTeX
@article{Wu2026,
author = {Wu, Shengze and Yang, Suibi and Yang, Jie and Zhou, Pengmin and Shen, Hongjie and Zhang, Zhongheng},
title = {From early warning to endotype-guided care in sepsis: A critical narrative review of AI, multi-omics, and clinical translation},
journal = {Intensive Care Science},
year = {2026},
doi = {10.1016/j.iics.2026.07.002}
}
EUVIMED