J-shaped relationship between stress hyperglycemia ratio and delirium risk in critically ill patients: A population-based study

Abstract

Background The incidence of delirium in critically ill patients is strongly correlated with poor prognosis. The stress hyperglycemic ratio has emerged as a novel marker for assessing the response to acute hyperglycemia. Glycemic fluctuations during periods of stress play a crucial role in precipitating or directly causing delirium. However, the association between SHR and delirium in hospitalized ICU patients remains uncertain. Objective This study aimed to investigate the potential relationship between SHR and delirium in ICU patie nts and to examine possible subgroup differences in this association. Methods A total of 2,093 Intensive care unit (ICU) patients were included in this retrospective cohort study. The relationship between SHR and delirium was explored using multifactorial logistic regression, subgroup analyses, smoothed curve fitting, and threshold effect analysis models. Results Among the 2,093 participants, 59.05% were male and 40.95% were female, with a mean age of 64.19 ± 16.31 years. We identified a non-linear positive correlation between SHR and delirium, with an inflection point at 0.68, and the odds ratio (95% CI) after the inflection point was 1.88 (1.35, 2.62), P  < 0.001. This interaction was statistically significant concerning the APACHE II scores and C-reactive protein levels at admission. Conclusion We found a nonlinear positive association between SHR and delirium. Our study highlights that managing SHR levels in critically ill patients may help to prevent or mitigate the development of delirium, emphasizing the potential value of SHR as an early intervention and precision treatment for delirium.