Stress hyperglycemia ratio is associated with delirium in elderly cardiac surgery patients in the postoperative intensive care unit: a multicenter retrospective cohort study.

@article{da2026stress,
  title = {{Stress hyperglycemia ratio is associated with delirium in elderly cardiac surgery patients in the postoperative intensive care unit: a multicenter retrospective cohort study.}},
  author = {Wei Da and Luyao Huang and Shu Zhou and Hong Wei and Yihong Yang and Juan Wang and Hong Zhang},
  journal = {BMC geriatrics},
  year = {2026},
  month = aug,
  doi = {10.1186/s12877-026-08083-5},
  url = {https://europepmc.org/article/MED/42972259},
  note = {CC BY-NC-ND},
}

TY  - JOUR
TI  - Stress hyperglycemia ratio is associated with delirium in elderly cardiac surgery patients in the postoperative intensive care unit: a multicenter retrospective cohort study.
AU  - Wei Da
AU  - Luyao Huang
AU  - Shu Zhou
AU  - Hong Wei
AU  - Yihong Yang
AU  - Juan Wang
AU  - Hong Zhang
PY  - 2026
DA  - 2026/08/05
JO  - BMC geriatrics
DO  - 10.1186/s12877-026-08083-5
UR  - https://europepmc.org/article/MED/42972259
AB  - Background  Postoperative delirium(POD) is severe in elderly cardiac surgery patients.Elevated glucose levels trigger oxidative stress, endothelial dysfunction, and cerebral metabolic disturbances, which are key mechanisms linking dysglycemia to delirium. Current models lack predictors, and dysglycemia interpretation is confounded by diabetes. The stress hyperglycemia ratio (SHR) integrates acute glucose and glycated hemoglobin (HbA1c) to quantify acute metabolic stress, but its predictive value for POD needs validation.Methods  This was a multicenter retrospective cohort study involving elderly patients (≥ 65 years) admitted to the ICU following cardiac surgery. The discovery cohort was derived from the MIMIC-IV database (n = 1,087), while the external validation cohort consisted of patients from six Chinese hospitals (n = 982). The primary outcome was POD. The optimal threshold for the SHR was identified using restricted cubic splines (RCS). Subsequently, the independent association between SHR and POD was evaluated via multivariable logistic regression with sequential covariate adjustment. The performance of the prediction model based on this threshold was assessed using the receiver operating characteristic curve (ROC), calibration plots, and decision curve analysis. To further ensure the robustness of the findings, we performed propensity score matching (PSM) and comprehensive subgroup analyses.Results  RCS analysis revealed a nonlinear relationship between the SHR and the risk of POD, with an inflection point at 1.09. After extensive covariate adjustment, an SHR ≥ 1.09 was strongly and independently associated with a significantly increased risk of POD (adjusted odds ratio [aOR], 4.152; 95% CI, 3.054-5.676, P < 0.001). This robust association was consistently validated in the external cohort (aOR, 3.856; 95% CI, 2.687-5.593, P < 0.001) and persisted in the PSM analysis. Notably, the univariate model based solely on continuous SHR exhibited good discriminative performance (area under the curve [AUC] 0.734), along with favorable calibration and clinical utility, compared with conventional severity scores, indicating SHR's potential value for risk stratification.Conclusion  An elevated SHR (≥ 1.09) was independently associated with a higher odds of POD in elderly cardiac surgery patients. This readily calculable metric may help identify patients at increased risk and guide enhanced monitoring or future intervention trials.
SN  - 1471-2318
ER  -