Impact of hemoglobin glycation index and stress hyperglycemia ratio on short-term mortality in critically ill patients with heart failure: a retrospective cohort study using the MIMIC database.

内容摘要

To evaluate the prognostic value of the combined assessment of hemoglobin glycation index (HGI) and stress hyperglycemia ratio (SHR) for short-term mortality in critically ill patients with heart failure (HF). ICU data on HF patients were retrospectively retrieved from the MIMIC-III and IV databases between 2001 and 2022. SHR and HGI were calculated at admission and stratified into tertiles. The endpoint was in-hospital and 90-day mortality. Logistic and Cox models evaluated associations between SHR, HGI, and their combination with mortality in both diabetic and non-diabetic patients. Exploratory interpretable machine learning prognostic models were developed to evaluate predictive performance. A total of 1573 HF patients were included. In-hospital and 90-day mortality rates were 14.1% and 24.3%, respectively. High SHR was independently associated with increased in-hospital mortality (OR: 2.01, 95% CI: 1.31-3.09) and 90-day mortality (HR: 1.54, 95% CI: 1.18-2.00). Higher HGI was associated with a reduced risk of in-hospital mortality (OR: 0.88, 95% CI: 0.78-0.99) and 90-day mortality (HR: 0.91, 95% CI: 0.84-0.98). High SHR and low HGI had the highest mortality risk for both in-hospital mortality (OR: 1.90, 95% CI: 1.288-2.85) and 90-day mortality (HR: 1.45, 95% CI: 1.13-1.87). The XGBoost model demonstrated optimal predictive accuracy, with an AUROC of 0.810 for in-hospital mortality and 0.808 for 90-day mortality. High SHR and low HGI are significantly associated with short-term mortality in HF patients, suggesting that their combined assessment may improve short-term prognostic assessment in critically ill patients with HF.