Metabolic acidosis is a common acid-base imbalance in critically ill patients often associated with poor outcomes. However, the prognostic value of albumin-corrected anion gap (ACAG) in diabetic patients with severe heart failure (HF) requiring intensive care unit (ICU) admission remains unclear. This study aimed to evaluate the predictive utility of ACAG for adverse outcomes in diabetic heart failure patients.

Cardiovascular diseases and diabetes mellitus (DM) burdens were first characterized using Global Health Data Exchange (GHDx) data, with incidence, prevalence, and disability-adjusted life years (DALYs) as metrics. Temporal trends were analyzed via Joinpoint regression models to calculate average annual percentage changes (AAPC). A retrospective analysis was then performed on 1,577 diabetic heart failure patients from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Patients were stratified into survival/death groups based on ICU outcomes and divided into quartiles according to ACAG levels. Kaplan-Meier survival curves with log-rank tests were used to assess short- and long-term outcomes. Cox proportional hazards models evaluated independent associations between ACAG and mortality. Restricted cubic spline (RCS) analysis examined dose-response relationships, and a nomogram prediction model incorporating ACAG was developed and validated.

The study cohort included 1,577 patients (54.85% male) with in-hospital and ICU mortality rates of 19.59% and 80.41%, respectively. Patients with ACAG ≥18.75 demonstrated significantly higher mortality compared to those with ACAG <18.75 (p<0.05). After adjusting for confounding variables, ACAG ≥18.75 remained independently associated with increased mortality risk. RCS analysis confirmed a linear relationship between ACAG and mortality risk (p<0.001). The validated nomogram model incorporating ACAG showed excellent discriminative ability.

ACAG is independently associated with in-hospital and ICU mortality in diabetic patients with severe heart failure. This finding suggests ACAG could serve as a valuable prognostic marker for identifying high-risk individuals in this population.