Association of preoperative albumin-corrected anion gap with 28-day mortality in cardiac surgery patients: a retrospective cohort study

Cardiac surgery patients face significant postoperative risks. The albumin-corrected anion gap (ACAG) has emerged as a potential prognostic marker, but its role in predicting outcomes following cardiac surgery remains understudied. This study examined the association between preoperative ACAG levels and 28-day mortality following ICU admission in patients undergoing cardiac surgery. This retrospective cohort study analyzed 5,006 cardiac surgery patients from the MIMIC-IV database. Patients were categorized based on ACAG levels: low ( 20 mmol/L). Hierarchical Cox proportional hazards models, restricted cubic spline (RCS) analysis, and exploratory mediation analysis were used to evaluate the association between preoperative ACAG and 28-day mortality following ICU admission. Incremental predictive analyses were additionally performed to assess whether ACAG improved the prognostic performance of models based on SOFA and SAPS II. Elevated ACAG levels (> 20 mmol/L) were significantly associated with a heightened risk of 28-day mortality following ICU admission across all models. In the most comprehensively adjusted model, the hazard ratio was 1.049 (95% CI 1.024–1.075, P < 0.001). ACAG demonstrated superior predictive value for 28-day mortality following ICU admission compared to the traditional anion gap (AUC 0.776 vs. 0.690). RCS revealed a nonlinear relationship, with mortality risk sharply increasing at ACAG levels above 20 mmol/L.In incremental predictive analyses, adding ACAG improved the prognostic performance of both SOFA- and SAPS II-based models. Exploratory mediation analyses suggested that SAPS II, GCS, and several electrolyte-related variables accounted for part of the observed association between preoperative ACAG and 28-day mortality. Elevated preoperative ACAG levels were associated with increased 28-day mortality risk in cardiac surgery patients. Although part of this association may reflect overall illness severity and postoperative physiological stress, ACAG may serve as a useful early risk-stratification marker and could complement existing prognostic assessment frameworks. However, its causal role cannot be inferred from this observational study, and prospective studies are needed to validate these findings.