Is the Stress Hyperglycemia Ratio a Predictor of Sepsis-Related Mortality? A Systematic Review and Meta-Analysis

Document Type

Conference Proceeding

Publication Date

10-2025

Publication Title

Chest

Abstract

PURPOSE: SHR (Stress Hyperglycemia Ratio) is estimated based on admission blood glucose and average glycemic level (HbA1c). Prior studies have indicated that increased SHR levels are associated with higher mortality in sepsis patients, reflecting the impact of glucose dysregulation. The strength and consistency of this relationship remain unclear. This meta-analysis aims to evaluate the association between SHR and in-hospital and one-year mortality in sepsis which could potentially be helpful in the risk stratification of sepsis patients. METHODS: A thorough search was conducted until January 2025 using PubMed and Google Scholar databases in accordance with PRISMA guidelines to examine the effect of SHR on sepsis-related mortality. Articles were included based on the primary outcome of SHR’s association with in-hospital mortality and with one-year mortality as the secondary outcome. Random-effects models were utilized to calculate pooled hazard ratios (HR) and odds ratios (OR) with 95% confidence intervals (CI). Heterogeneity was evaluated with the I 2 statistic, and publication bias was analyzed through funnel plots. RESULTS: Our meta-analysis included 18261 patients from four studies. The pooled mean age was 68.94 ± 13.39 years. The median was converted to an estimated mean using the method described by Luo et al. (2016) to ensure comparability. The weighted mean percentage of males across all studies was 60.65% (pooled SD: 1.74%). After adjusting for confounders, higher SHR was associated with a 2.62-fold increase in the odds of in-hospital all-cause mortality (aOR-2.62, 95% CI: 1.73-3.95, p< 0.01). For one-year all-cause mortality, adjusted odds ratio (aOR) was 1.58 (95% CI: 1.35–1.86, p< 0.01), indicating a continued risk beyond hospitalization. Heterogeneity was high (I2 = 88.18%) for in-hospital allcause mortality and substantial (I2 = 65.18%) for one-year all-cause mortality, indicating variability among the included studies. The leave-one-out sensitivity analysis confirmed the robustness of the overall findings, showing that no single study disproportionately influenced these outcomes. CONCLUSIONS: Higher SHR remained a significant predictor of both in-hospital and one-year all-cause mortality in sepsis patients after adjusting for confounders. However, heterogeneity was high for in-hospital mortality and substantial for one-year mortality, which warrants cautious interpretation. Potential contributors to this variability include differences in study design, patient populations, and SHR threshold definitions. CLINICAL IMPLICATIONS: Our findings suggest that SHR may serve as a valuable prognostic biomarker for early risk stratification in sepsis patients, aiding clinicians in identifying high-risk individuals. Given the significant association between elevated SHR and mortality, integrating SHR assessment into clinical decision-making may enhance risk prediction models. Additionally, interventions targeting glucose control in critically ill septic patients, such as tailored glycemic management strategies, may have potential benefits in improving outcomes. Further research is needed to establish optimal SHR thresholds, evaluate its incorporation into sepsis management guidelines, and determine whether targeted interventions can effectively mitigate its impact on mortality.

Volume

168

Issue

4S

First Page

2071A

Last Page

2072A

Comments

American College of Chest Physicians CHEST Annual Meeting, October 19-22, 2025, Chicago, IL

DOI

10.1016/j.chest.2025.07.1182

Link to Full Text

Share

COinS