The relationship between hemoglobin glycation index and all-cause mortality in ill critically patients with heart failure: a retrospective study in MIMIC-IV database

BMC Cardiovasc Disord. 2025 Apr 24;25(1):317. doi: 10.1186/s12872-025-04711-x.

ABSTRACT

BACKGROUND: Heart failure (HF) is a major cause of mortality in critically ill patients and often requires intensive care. The hemoglobin glycation index (HGI), defined as the difference between predicted glycated hemoglobin (HbA1c) and measured HbA1c, may provide additional prognostic insights beyond traditional glycemic metrics.

METHODS: We conducted a retrospective analysis of 8,098 adult patients with HF from the MIMIC-IV database (2008-2022). All were first-time ICU admissions with available hematologic and metabolic data. Patients were stratified into three groups (T1 ≤ - 1.26, - 1.26 < T2 < 1.74, T3 ≥ 1.74) based on HGI. Baseline characteristics were recorded within 24 h of ICU admission, including demographic data, disease severity scores, comorbidities, and medication use. Logistic regression and Cox proportional hazards models assessed the associations between HGI and in-hospital, 30-day, and 1-year all-cause mortality, adjusting for age, sex, race, comorbidities, laboratory results, and relevant treatments. Restricted cubic spline (RCS) analysis was performed to examine potential non-linear relationships. We used sensitivity analyses to increase the confidence in our primary outcome.

RESULTS: Patients in the lowest HGI group (T1) had significantly higher in-hospital, 30-day, and 1-year mortality than those in the other two groups. Specifically, T1 showed an 18.6% in-hospital mortality rate, compared with 12.3% and 9.7% in T2 and T3, respectively (p < 0.001). Fully adjusted models revealed that each 1-unit increase in HGI was associated with an approximate 12% reduction in in-hospital mortality risk (OR = 0.88; 95%CI: 0.83-0.93), and an 3% decreased risk of 1-year all-cause mortality (HR 0.97; 95%CI0.94~1.00). RCS analysis indicated a J-shaped relationship between HGI and mortality, underscoring the heightened risk associated with very low HGI. We conducted sensitivity analyses by separately excluding missing data, diagnosed sepsis, and diagnosed hepatic impairment, consistent with the primary analysis.

CONCLUSIONS: In critically ill HF patients, extremely low HGI levels correlate with poorer short- and long-term survival. These findings suggest that HGI could serve as an adjunct risk stratification tool, prompting closer monitoring and potential intervention in patients with markedly low HGI.

PMID:40275131 | DOI:10.1186/s12872-025-04711-x