Predictive value of glucose metabolism indices for short-term outcomes in critically ill patients with congestive heart failure

@article{钟2026predictive,
  title = {{Predictive value of glucose metabolism indices for short-term outcomes in critically ill patients with congestive heart failure}},
  author = {华 钟 and 劢 冯 and 婷 吴 and 晚露 周 and 千慧 罗 and 瑞正 石},
  journal = {Journal of Central South University Medical Sciences},
  year = {2026},
  month = aug,
  doi = {10.11817/j.issn.1672-7347.2026.260338},
  url = {https://doi.org/10.11817/j.issn.1672-7347.2026.260338},
  note = {CC BY-NC-ND 4.0},
}

TY  - JOUR
TI  - Predictive value of glucose metabolism indices for short-term outcomes in critically ill patients with congestive heart failure
AU  - 华 钟
AU  - 劢 冯
AU  - 婷 吴
AU  - 晚露 周
AU  - 千慧 罗
AU  - 瑞正 石
PY  - 2026
DA  - 2026/08/28
JO  - Journal of Central South University Medical Sciences
DO  - 10.11817/j.issn.1672-7347.2026.260338
UR  - https://doi.org/10.11817/j.issn.1672-7347.2026.260338
AB  - Objective: Congestive heart failure (CHF) is a major global public health problem, affecting approximately 64 million people worldwide, with a 5-year survival rate of less than 50%. The short-term risk of death is particularly high among patients admitted to the intensive care unit (ICU). Disturbances in glucose metabolism are an important pathophysiological feature of CHF, and effective glycemic monitoring is of considerable importance for prognostic assessment in critically ill patients with CHF. This study aims to evaluate the predictive value of 3 glucose metabolism-related indices, blood glucose level, stress hyperglycemia ratio (SHR), and hemoglobin glycation index (HGI), for the risk of 7-day mortality in critically ill patients with CHF.  Methods: This study was based on the MIMIC-IV database and included 851 critically ill patients with CHF who met the eligibility criteria between 2008 and 2019. The primary outcome was 7-day all-cause mortality, and the secondary outcome was 30-day all-cause mortality. Kaplan-Meier (K-M) survival curves with log-rank tests were used to assess the associations between each index and survival outcomes. Univariable and multivariable Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs). Three models were constructed to progressively evaluate the influence of potential confounders: an unadjusted model, Model 1 adjusted for demographic and clinical characteristics, and Model 2 further adjusted for laboratory parameters. Restricted cubic spline (RCS) analysis was used to explore the dose-response relationships between the 3 indices and mortality risk. Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were used to evaluate the discriminative ability of each index, and incremental ROC analysis was performed to determine whether adding each index improved the predictive performance of the baseline model. Least absolute shrinkage and selection operator (LASSO) regression was used to select features from 15 candidate variables and to objectively assess the independent predictive contribution of each index in a multivariable setting. In addition, stratified analyses were conducted across multiple clinically relevant subgroups defined by sex, age, type 2 diabetes mellitus (T2DM) status, vasopressin use, acute myocardial infarction (AMI), acute kidney injury (AKI), and sepsis to evaluate the consistency and robustness of the prognostic value of SHR.  Results: A total of 851 critically ill patients with CHF were included, of whom 70 (8.226%) died from any cause within 7 days. Cox regression analysis showed that, when analyzed as a continuous variable, SHR was independently associated with 7-day all-cause mortality in the fully adjusted model (HR=1.760, 95% CI 1.193 to 2.595; P<0.05), whereas the association between blood glucose and 7-day all-cause mortality was no longer significant after full adjustment (HR=1.002, 95% CI 0.999 to 1.004; P>0.05). HGI, when analyzed as a continuous variable, was associated with 7-day all-cause mortality (HR=0.787, 95% CI 0.650 to 0.954; P<0.05); however, no significant association was observed after HGI was categorized into tertiles in the fully adjusted model (P>0.05). Tertile analysis of SHR showed that the highest tertile remained significantly associated with 7-day all-cause mortality in the fully adjusted model (P<0.05). K-M survival analysis further showed that patients with higher blood glucose levels or higher SHR had a greater risk of 7-day all-cause mortality, whereas patients with lower HGI also had a significantly increased risk of 7-day all-cause mortality (all P<0.05). ROC curve analysis showed that SHR tended to have better predictive performance for mortality risk than blood glucose and HGI, although the differences did not reach statistical significance (P>0.05). Notably, adding SHR to the baseline model significantly improved its predictive performance (P=0.048). LASSO regression analysis showed that among the 9 predictors ultimately selected, SHR was the only glucose-related variable retained, whereas blood glucose and HGI were excluded.  Conclusion: SHR is an independent predictor of short-term mortality risk in critically ill patients with CHF and has considerable potential for clinical application in risk stratification and prognostic assessment.
SN  - 1672-7347
ER  -