EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Establishment and Validation of a Prognostic Model for 28‐Day Mortality Incorporating Inflammatory Mediators and Organ Function Indicators in Patients With Sepsis: A Retrospective Multicenter Study

Entwicklung und Validierung eines Prognosemodells für die 28-Tage-Mortalität mit Entzündungsmediatoren und Organfunktionsindikatoren bei Patient:innen mit Sepsis: retrospektive multizentrische Studie

Meng-Xiao Feng, Cong-Ying Song, Wei Huang, Jin-Min Chen, Zheng Zhang, Zhi-Yuan Fan, Yuan-Qiang Lu

Mediators of Inflammation · 2026-01 · 2026 (1)

DOI: 10.1155/mi/5567993 ↗

Auf einen Blick

Bei 200 Patient:innen mit Sepsis wurde ein Prognosemodell für die 28-Tage-Mortalität entwickelt und intern validiert. Es zeigte eine gute Trennschärfe in Trainings- und Validierungsdatensatz.

Fragestellung

Lässt sich mit Entzündungsmediatoren und Organfunktionsindikatoren ein Prognosemodell zur Vorhersage der 28-Tage-Mortalität bei Sepsis erstellen und validieren?

Methode

Retrospektive multizentrische Studie mit 200 Patient:innen; Randomisierung in Trainingsdatensatz (n = 140) und Validierungsdatensatz (n = 60). LASSO- und logistische Regression, ROC, Kalibrierung, Decision Curve und Clinical Impact Curve.

Ergebnisse

Das Nomogramm enthielt AST, log(IL-6) und IFN-γ (Basis 10). AUC 0,889 (95 %-KI 0,808–0,970) im Training und 0,808 (95 %-KI 0,690–0,925) in der Validierung.

Schlussfolgerung der Autor:innen

Das intern validierte Prognosemodell mit AST, IL-6 und IFN-γ zeigte eine gute Diskriminierung für die 28-Tage-Mortalität bei Sepsis und könnte frühe Risikostratifizierung unterstützen.

Kernaussagen

  • AUC 0,889 im Trainingsdatensatz
  • AUC 0,808 im Validierungsdatensatz
  • AST, IL-6 und IFN-γ waren im Modell enthalten

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Mediators of Inflammation, online erschienen am 09.11.2026. Verfasst von Meng-Xiao Feng und 6 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China.

Auf einen BlickAbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Conventional prognostic markers have limited discriminative power for mortality prediction in septic patients. Building on the fact that dysregulated inflammatory responses are central to sepsis pathophysiology, this study aimed to establish and validate a novel prognostic model incorporating inflammatory mediator‐related and organ function‐related indicators to predict 28‐day mortality in septic patients.

Methods

This retrospective multicenter study enrolled 200 eligible septic patients from January 2019 to April 2022. The primary outcome was defined as 28‐day all‐cause mortality. The patients were randomized into training dataset ( n = 140) and validation dataset ( n = 60) at a ratio of 7:3. A comprehensive set of clinical and laboratory parameters, including organ‐related indicators and inflammatory mediators, was systematically collected. The risk factors for predicting death were selected using the least absolute shrinkage and selection operator (LASSO) logistic regression and the binary logistics regression analysis. The receiver operating characteristic (ROC) curve was drawn to determine the model discrimination. Model performance was further evaluated by calibration curves, decision curve analysis, and clinical impact curve.

Results

The training dataset included 110 survivors and 30 nonsurvivors. After a multistep screening process, the prognostic factors in the nomogram included aspartate aminotransferase (AST), the logarithmic form of interleukin (IL)‐6, and interferon (IFN)‐γ with base 10. The model exhibited a good predictive power for death with an area under the curve of 0.889 (95% confidence interval [CI]: 0.808–0.970) in the training dataset and 0.808 (95% CI: 0.690–0.925) in the validation dataset. The calibration curves showed a good agreement between the predicted and actual probabilities. Both decision curve analysis and clinical impact curve analysis suggested potential clinical utility.

Conclusion

The prognostic model developed and internally validated in this study incorporates AST and two inflammatory mediators, IL‐6 and IFN‐γ, which demonstrated good discriminative performance in predicting 28‐day mortality among septic patients. The findings suggest that inflammatory mediator‐related variables, when combined with an organ function‐related indicator, may facilitate early risk stratification and provide complementary prognostic information for clinical decision‐making.

Förderung

  • National Key Research and Development Program of China

Zitieren

APA

Feng, M., Song, C., Huang, W., Chen, J., Zhang, Z., Fan, Z., & Lu, Y. (2026). Establishment and Validation of a Prognostic Model for 28‐Day Mortality Incorporating Inflammatory Mediators and Organ Function Indicators in Patients With Sepsis: A Retrospective Multicenter Study. Mediators of Inflammation, 2026(1). https://doi.org/10.1155/mi/5567993

Vancouver

Feng M, Song C, Huang W, Chen J, Zhang Z, Fan Z, et al. Establishment and Validation of a Prognostic Model for 28‐Day Mortality Incorporating Inflammatory Mediators and Organ Function Indicators in Patients With Sepsis: A Retrospective Multicenter Study. Mediators of Inflammation. 2026;2026(1). doi:10.1155/mi/5567993

BibTeX
@article{Feng2026,
  author = {Feng, Meng-Xiao and Song, Cong-Ying and Huang, Wei and Chen, Jin-Min and Zhang, Zheng and Fan, Zhi-Yuan and Lu, Yuan-Qiang},
  title = {Establishment and Validation of a Prognostic Model for 28‐Day Mortality Incorporating Inflammatory Mediators and Organ Function Indicators in Patients With Sepsis: A Retrospective Multicenter Study},
  journal = {Mediators of Inflammation},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/mi/5567993}
}
Kostenlos starten →