The modified HALP score is associated with short-term mortality in critically ill patients with sepsis - A cohort study

Abstract

Introduction: To systematically appraise the prognostic predictive value of the modified HALP (m-HALP) score in critically ill septic patients.

Methodology: The m-HALP scores were computed for septic patients within the initial 24 hours of admission to the intensive care unit (ICU) utilizing data from the MIMIC-IV database. The association between the m-HALP score and 30-day mortality was evaluated using restricted cubic splines and Cox regression. Kaplan-Meier (K-M) analysis was employed to estimate survival differences. Logistic regression was conducted using data from the eICU database to validate the findings. Receiver operator characteristic (ROC) curves were generated to assess predictive value.

Results: The m-HALP score exhibited an L-shaped association with 30-day mortality upon adjustment for multiple variables (HR: 0.84, 95% CI: 0.74-0.96). K-M curves revealed a favorable survival outcome in patients with high m-HALP scores (p < 0.001). In the validation cohort, the m-HALP score proved to be an independent factor influencing in-hospital mortality. The ROC curves suggested that the m-HALP score had a better predictive value for short-term sepsis mortality than the HALP and qSOFA score.

Conclusions: The m-HALP score demonstrated a noteworthy correlation with short-term mortality of septic patients, making it a potentially promising biomarker of prognostic relevance.

Keywords: Sepsis; critically ill patients; m-HALP score; mortality; prognostic.