The predictive value of glycemic coefficient of variation(GLUcv)for sepsis-associated encephalopathy in emergency sepsis patients

Chen Wei; Shen Qi; Chen Yuhong

doi:10.3969/j.issn.1002-1949.2024.09.016

Abstract：Objective To analyze the predictive value of glycemic coefficient of variation(GLUcv)for sepsis-associated encephalopathy(SAE)in emergency sepsis patients.Methods A prospective cohort study was used to select sepsis patients who were treated in Emergency EICU Department of Shangyu People's Hospital from March 2022 to February 2024.The peripheral blood glucose of the patients within 24 h of admission were measured and GLUcv was calculated.The SAE events within 28 d of sepsis diagnosis were considered as research endpoint.Patients who experienced SAE within 28 d were included in the SAE group,while those who did not experience SAE were included in the N-SAE group.The differences in main data and indicator levels were compared between two groups after grouping,and the relationship of sepsis complicated by SAE with GLUcv and other major indicators were analyzed by constructing a Logistic regression model.The restricted cubic spline method was used to analyze the dose-response relationship between GLUcv and the risk of sepsis complicated by SAE.The receiver operating characteristic(ROC)curves and decision curves were drawn to analyze the predictive value of GLUcv for sepsis complicated by SAE.Results A total of 135 patients completed the study,with GLUcv measurements ranging from 12.59 to 33.60 within 24 h of admission.According to the low to high quartile of GLUcv,135 patients were divided into Q1 to Q4 groups.The incidence of SAE in Q4 group was higher than that in Q1 to Q3 group,and the acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ),serum amyloid A protein(SAA),interleukin-6(IL-6),and neuron specific enolase(NSE)levels were higher in Q3 and Q4 groups than in Q1 and Q2 groups.The 24-hour lactate clearance rate was lower in Q3 and Q4 groups than in Q,and Q2 groups(P＜0.05).57 cases of SAE occurred within 28 d after diagnosis,with an incidence rate of approximately 42.22％.At admission,the APACHE Ⅱ score,GLUcv,IL-6 and NSE levels in the SAE group were higher than those in the N-SAE group,and the 24-hour lactate clearance rate was lower than that in the N-SAE group(P＜0.05).The results of Logistic regression model showed that sepsis complicated by SAE may be related to APACHE Ⅱ score at admission,24-hour lactate clearance rate,GLUcv,serum IL-6,NSE and SAA(P＜0.05).Drawing ROC curve,the area under curve(AUC)of GLUcv predicting the SAE for sepsis patients was 0.797,with an optimal cutoff value of 23.24.The restricted cubic spline method showed there was a linear dose-response relationship between GLUcv within 24 h of admission and the risk of SAE within 28 d of admission in sepsis patients(P＜0.05).When the GLUcv threshold reached 23.24,the risk of SAE increased with the increase of GLUcv.Drawing a decision curve,the clinical net benefit of GLUcv predicting SAE in sepsis patients was ideal,and GLUcv combined with other indicators can improve the overall decision-making benefit.Conclusions The occurrence of SAE in sepsis patients is related to an increase of GLUcv,and the risk of SAE increases with the increase of GLUcv.Calculating the GLUcv value within 24 h of admission for sepsis patients can predict the high-risk population of SAE,and combining with other indicators can improve the net benefit rate of overall risk prediction.