Association between albumin-corrected anion gap level and all-cause mortality in patients with gastrointestinal bleeding: A retrospective cohort study of the MIMIC-IV database

This study explores the association of albumin-corrected anion gap (ACAG) level and all-cause mortality in patients with gastrointestinal bleeding (GIB). Data in this retrospective cohort study were extracted from the medical information mart for intensive care IV database. Univariate and multivariate Cox proportional hazards models were utilized for evaluation of hazard ratio (HR) of 30- and 90-day all-cause mortality in different ACAG levels. Subgroup analysis of age, gender, complications and mechanical ventilation was conducted to investigate this relationship in patients with different characteristics. Among 1933 eligible participants, 390 died in 30 days and 540 died in 90 days. After adjusting for covariates, the concentration of ACAG elevated 1 mmol/L, the risk of 30-day all-cause mortality and 90-day all-cause mortality increased 0.064 (95% confidence interval [CI]: 1.046-1.084) and 0.061 (95% CI: 1.044-1.078), respectively. Compared to a normal ACAG level, a high ACAG level was associated with increased risk of both 30-day (HR = 1.620, 95% CI: 1.281-2.049) and 90-day (HR = 1.572, 95% CI: 1.287-1.921) all-cause mortality. The restricted cubic spline curves indicated a linear correlation between ACAG and all-cause mortality in GIB patients. Additionally, this positive association between ACAG and increased mortality risk was significant in age, gender, non-hypertension, non-diabetes mellitus, sepsis, atrial fibrillation, non-heart failure, nonmechanical ventilation, and mechanical ventilation subgroups (all P < .05). An elevated ACAG level was associated with increased risk of mortality in GIB patients, with a linear correlation. However, the specific mechanism that ACAG level linked to mortality risk in GIB patients still needs clarification.