Impact of time-varying blood glucose exposure intensity and variability on the prognosis of sepsis patients
10.3760/cma.j.cn114656-20241028-00756
- VernacularTitle:时变血糖暴露强度及变异性对脓毒症患者预后的影响
- Author:
Qixiang HONG
1
;
Wentao WU
;
Xiaomin LI
Author Information
1. 徐州医科大学附属连云港医院（连云港市第一人民医院），连云港 222000
- Keywords:
Critical illness;
Sepsis;
Glucose;
Metabolism disorders;
ICU
- From:
Chinese Journal of Emergency Medicine
2025;34(9):1229-1236
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the correlation between the intensity of glycemic exposure over time [time-weighted average glucose (TWA-G)] and time-weighted average glucose variation (TWA-GV) with the prognosis of septic patients.Methods:Data for this study were obtained from the MIMIC-IV database and inpatients from The First People's Hospital of Lianyungang between 2020 and 2024. The study included all intensive care unit (ICU) patients who had blood glucose measured, were aged ≥18 years, and met the Sepsis-3.0 diagnostic criteria. The primary outcome was 28-day mortality. The relationship between the time-varying intensity of glycemic exposure and mortality was estimated using a piecewise exponential additive mixed model. Restricted cubic splines were further employed to assess the optimal proportion of time spent within specific blood glucose risk thresholds for septic patients.Result:This study incorporated 15 116 sepsis hospitalization records from the MIMIC-Ⅳ database and 565 septic patients from The First People's Hospital of Lianyungang. After adjusting for baseline characteristics and time-dependent confounding factors, the study found that both TWA-G and TWA-GV were associated with an increased risk of 28-day mortality. TWA-G values outside the range of 86-172 mg/dL or TWA-GV greater than 8% were significantly associated with an increased hazard ratio for mortality. Mild glycemic variability in the early stage was associated with slightly lower mortality. The short-term impact of TWA-G and TWA-GV on septic patients was significant, particularly affecting mortality in elective surgical, female, and obese diabetic patients. Subgroup analysis revealed that the diabetic population tolerated a wider glycemic range (86-215 mg/dL) compared to the non-diabetic population (86-172 mg/dL). Similarly, low-risk septic patients tolerated a wider range (87-193 mg/dL) compared to high-risk septic patients (84-170 mg/dL). Furthermore, the study found that an accumulated time spent outside the target glycemic range exceeding 72% ( HR=1.01, 95% CI: 1.00-1.02, P=0.04) was associated with increased patient mortality risk. Conclusions:Blood glucose levels and glycemic variability should be closely monitored during the early and middle phases following ICU admission in septic patients. High-intensity glycemic exposure is associated with an increased risk of mortality. Mild glycemic variability may be more physiologically adaptive in the early stages of sepsis.
