[Correlation between glycemic variability coefficient and sepsis-induced coagulopathy and analysis of risk factors]

To investigate the association between glycemic variability coefficient (CV) and sepsis-induced coagulopathy (SIC), analyze the related risk factors for SIC in septic patients, and further explore the clinical value of glycemic CV as a potential predictor for SIC.

A retrospective case-control study was conducted. General data, laboratory findings, therapeutic interventions, and disease severity scores of septic patients admitted to the Department of Critical Care Medicine of the First Hospital of Qinhuangdao from July 2021 to June 2024 were collected. Blood glucose CV within 24 hours of admission was used as the exposure variable, and SIC served as the primary outcome. Patients were divided into Q₁, Q₂, Q₃, and Q₄ groups according to quartiles of glycemic CV within 24 hours of intensive care unit (ICU) admission, with Q₁ as the reference. Baseline clinical characteristics were compared among the four groups. Multivariate logistic regression was performed to analyze the effect of glycemic CV on SIC occurrence after adjusting for various confounders. Subgroup analyses were used to explore potential interaction effects and further verify the robustness of the results. A restricted cubic spline model was applied to examine the dose-response relationship between glycemic CV and SIC. Patients were divided into SIC and non-SIC groups based on the presence of SIC. Univariate analysis was used to identify risk factors for SIC, and significant variables were entered into binary logistic regression to determine independent risk factors. Receiver operator characteristic curve (ROC curve) was plotted to assess the predictive value of glycemic CV for SIC.

A total of 440 patients with sepsis were enrolled, including 110 in Q₁ (glycemic CV<19.5%), 109 in Q₂ (19.5%≤glycemic CV<26.0%), 111 in Q₃ (26.0%≤glycemic CV<31.5%), and 110 in Q₄ (glycemic CV≥31.5%). Multivariate logistic regression showed that after adjustment for confounders, each 10% increase in glycemic CV was associated with an increased risk of SIC [odds ratio (OR)=1.049, 95% confidence interval (95%CI) was 1.024-1.074, P<0.001]. Compared with Q₁, the risk of SIC remained significantly higher in Q₄ (OR=2.739, 95%CI was 1.427-5.255, P=0.002). Restricted cubic spline (RCS) analysis revealed a linear dose-response relationship between glycemic CV and SIC risk (overall P<0.001, nonlinear P=0.828), with the risk increasing linearly as glycemic CV rose. Subgroup analyses showed no significant interaction between glycemic CV and sex, age, diabetes, Acute Physiology and Chronic Health Evaluation II(APACHE II), Sequential Organ Failure Assessment (SOFA), use of vasoactive agents, use of corticosteroids, or anticoagulant therapy, consistently supporting elevated glycemic CV as a risk factor for SIC. Among the 440 patients, 280 were in the non-SIC group and 160 in the SIC group. Multivariate logistic regression including significant variables from univariate analysis identified high glycemic CV (OR=1.049, 95%CI was 1.024-1.074, P<0.001), high body mass index (BMI; OR=1.175, 95%CI was 1.106-1.248, P<0.001), and continuous renal replacement therapy (CRRT; OR=3.286, 95%CI was 1.561-6.919, P=0.002) as independent risk factors for SIC. ROC curve analysis showed that the area under the curve (AUC) of glycemic CV for predicting SIC was 0.624, with a cutoff value of 24.5%, yielding a specificity of 52.1% and sensitivity of 70.6%.

There is a positive linear relationship between glycemic CV and the risk of SIC in septic patients. High glycemic CV, high BMI, and CRRT are independent risk factors for SIC. Glycemic CV has certain predictive value for SIC, and monitoring glycemic CV may facilitate early identification and prevention of SIC, thereby improving patient outcomes.