Endothelial Activation and Stress Index Scores for 2,593 Urosepsis Patients

A retrospective study of 2,593 patients with urosepsis admitted to ICUs, using data from the MIMIC-IV database and West China Hospital. The dataset was created by Jiaqi Tang and last updated in May 2026. It investigates the association between the Endothelial Activation and Stress Index (EASIX) biomarker and 28-day mortality.

Use Cases

• Predicting 28-day ICU mortality based on EASIX scores and other clinical indices.
• Developing machine learning models for patient risk stratification using features like Charlson Comorbidity Index and SAPS II.
• Analyzing the non-linear relationship between endothelial dysfunction biomarkers and short-term mortality.
• Validating prognostic models for sepsis in external patient cohorts.

Strengths

• Includes data from 2,593 patients, providing a substantial cohort for analysis.
• External validation was performed using a separate cohort, suggesting methodological rigor.
• The predictive model incorporating EASIX achieved AUC values of 0.70–0.73 across training and validation sets.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

MIMIC-IV database and West China Hospital.

Collection Method

Retrospective cohort study.

Freshness

Last updated 2026-05-12 05:42:33; freshness should be verified.