GLR Biomarker and Mortality in Sepsis-Induced ARDS: Multicenter Retrospective Cohort Study

A retrospective cohort study of 2,485 patients from the MIMIC-IV database and 298 patients from a Chinese hospital, investigating the association between the Glucose-to-Lymphocyte Ratio (GLR) and 28-day mortality in sepsis-induced Acute Respiratory Distress Syndrome (ARDS). The data was used to derive and validate a prognostic biomarker model, authored by Shucun Liu and published on figshare in March 2026.

Use Cases

• Predicting short-term mortality in sepsis-induced ARDS patients based on admission biomarker levels.
• Evaluating the incremental prognostic value of GLR over established severity scores like APS III and SAPS II.
• Conducting subgroup analyses to identify patient populations where the biomarker is most effective.
• Training Cox proportional hazards models for clinical risk stratification.
• Performing decision curve analysis to assess the clinical utility of a new composite biomarker.

Strengths

• The study uses a primary cohort of 2,485 patients and an external validation cohort of 298 patients.
• The analysis employs multivariable Cox modeling, LASSO-Boruta selection, and restricted cubic splines.
• Results show GLR incorporation improved AUC values for APS III (0.694 to 0.708) and SAPS II (0.678 to 0.696).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small (14.2 KB), suggesting it may contain summary statistics or a limited subset of the original clinical data.

Provenance

Source

MIMIC-IV database and Affiliated Hospital of Xuzhou Medical University.

Collection Method

Multicenter retrospective cohort analysis.

Freshness

Last updated 2026-03-19 06:34:34; freshness should be verified.