Association of hemoglobin glycation index with all-cause mortality in critically

The hemoglobin glycation index (HGI), which quantifies individual variation in the relationship between hemoglobin A1c and glucose levels, has been associated with adverse outcomes across various clinical populations. However, the prognostic significance of HGI in critically ill patients with atrial fibrillation remains unknown. We investigated the association between HGI and all-cause mortality in this population. We conducted a retrospective cohort study using the MIMIC-III and MIMIC-IV databases. Adult patients (≥ 18 years) with atrial fibrillation during ICU admission were included. HGI was calculated as the difference between observed and predicted hemoglobin A1c values (predicted HbA1c = 0.013 × glucose + 4.804). Primary outcomes were in-hospital and 30-day mortality. Patients were stratified into HGI quartiles. Statistical analyses included Kaplan-Meier survival analysis, restricted cubic splines, and multivariable logistic and Cox regression models. Subgroup and sensitivity analyses were performed. Among 9,924 critically ill patients with atrial fibrillation (median age 72.3 years, 63.1% male), 714 (7.2%) died during hospitalization and 912 (9.2%) died within 30 days of ICU admission. Non-survivors had significantly lower HGI values than survivors (-2.1 vs. -2.0, P < 0.001). Kaplan-Meier analysis demonstrated markedly inferior survival in the lowest HGI quartile (P < 0.001). Restricted cubic spline analysis revealed an L-shaped dose-response relationship with threshold effects. In fully adjusted models, each 1-unit HGI increase was associated with reduced mortality (in-hospital: OR 0.89, 95% CI 0.82–0.97; 30-day: HR 0.91, 95% CI 0.85–0.97). These associations remained consistent across subgroups and sensitivity analyses. Higher HGI values are independently associated with reduced short-term mortality risk in critically ill patients with atrial fibrillation. HGI represents a valuable prognostic biomarker for risk stratification and clinical decision-making in this population.