The Real-world Study Of Sepsis-associated Brain Injury Based On A Critical Care Database And Validation Of Basic Experiments

Chapter one:Construction and external validation of a clinical prediction model for prognosis of patients with SAD(Sepsis-associated delirium)Objective:To construct a predictive model for death in the ICU in patients with sepsis-associated delirium based on the critical care database and to explore the risk factors for death in patients with sepsis-associated delirium.Methods:Clinical data of SAD patients were retrospectively collected using the MIMIC-Ⅳ database,and possible potential predictor variables were included.The presence or absence of death in the ICU was used as the final predictor.Nine variables were screened by one-way logistic regression model,two-way stepwise regression method and LASSO regression,and finally a prognostic prediction model was constructed using multi-factor logistic regression and column line plots were presented.The constructed SAD prognostic prediction model was validated using the eICU database.The discrimination and calibration of the model were assessed.Results:The all-cause mortality rate in the ICU for SAD patients was approximately 14.76%.A total of 3422 SAD patients were eventually included in the modeling cohort,and a prediction model consisting of variables(age,type of admission,SOFA score,tumor history,respiratory rate,blood urea nitrogen BUN,coagulation activation time PTT,oxygen partial pressure PO2,and lactate)with a C-index of 0.786 was constructed.The eICU database was used for validation,and a total of 4686 SAD patients,the AUC for predicting mortality in the ICU was 0.754.However,the SOFA score was a poor predictor of prognosis for SAD patients.The true and predicted values on the plotted calibration curves were in good agreement in both the modeling cohort and the validation cohort.Conclusion:In this study,a prognostic prediction model within the ICU for patients with SAD was constructed and externally validated using a multicenter database.The model had good discrimination and calibration and outperformed the SOFA score,a commonly used severity scoring system for critical illness.Chapter two:Prognostic improvement of intravenous magnesium sulfate in patients with SAE(sepsis-associated encephalopathy)Objective:To investigate the relationship between intravenous magnesium sulfate use and the prognosis of SAE(sepsis-associated encephalopathy)patients.Methods:Patients with SAE were extracted from the MIMIC-Ⅳ database and matched 1:1 using the propensity score matching method for patients with intravenous magnesium sulfate and those without magnesium sulfate.The association of intravenous magnesium sulfate with 28-day mortality,total length of hospital stay,and length of ICU stay in SAE patients was assessed using propensity score matching PSM,multivariate Cox model,double robust estimation method,and Inverse Probability Of Treatment Weighting(IPTW,Inverse Probability Of Treatment Weighting).Results:A total of 5134 SAE patients were extracted from this study,and there were 1774 matched pairs of patients in the magnesium sulfate and non-magnesium sulfate groups.Intravenous magnesium sulfate was significantly associated with a lower 28-day mortality(8.57%vs 11.84%;OR 0.70;P=0.001)and with a lower first SOFA 24h upsurge value.However,it was associated with longer total length of hospital stay(10.53 days vs 13.98 days;P<0.001)and ICU stay(4.13 days vs 16.41 days;P<0.001)in patients.Conclusion:Intravenous magnesium sulfate reduces the 28-day morbidity and mortality rate in patients with SAE,but prolongs both the total-length of hospitalization and the length of ICU stay.Chapter three:Discussion on the mechanism of magnesium sulfate improving SAE sepsis-associated encephalopathyObjective:To explore the effect of magnesium sulfate on sepsis-associated encephalopathy(SAE)in mice.To explore the specific mechanism of magnesium sulfate in improving SAE.To provide new clinical treatment options for SAE.Methods:① SAE mouse model was established by cecal ligation and puncture,CLP surgery.② 7-day survival experiment was used to evaluate the effect of magnesium sulfate on the prognosis of SAE mice.③Morris water maze test was used to evaluate the effect of magnesium sulfate on cognitive function in SAE mice.④LPS was used to induce inflammatory response in the Bend3.0 monolayer endothelial cell model,and then magnesium sulfate was used to intervene.⑤Molecular biological experiments were used to detect the expression of ZO-1,HO-1/NRF2 and other proteins,pyroptosis,and cell viability.Results:In vivo experiments showed that:① Magnesium sulfate could improve the cognitive function and survival rate of S AE mice,but it could not improve anxiety and depression disorders.②Magnesium sulfate may improve sepsis-associated encephalopathy by inhibiting pyroptosis of hippocampal neurons.In vitro experiments showed that:①magnesium sulfate may improve sepsis-associated encephalopathy by reducing blood-brain barrier permeability in mice.②LPS could reduce the expression of ZO-1 in brain endothelial cell line Bend3.0,which was positively correlated with the concentration of LPS.③Magnesium sulfate may improve the LPS-induced reduction of ZO-1 expression in endothelial cells by enhancing HO-1/NRF2 signal.④Magnesium sulfate can inhibit endothelial cell death induced by sepsis histone.Conclusions:magnesium sulfate can improve the memory impairment and survival rate of SAE mice with cecal ligation and perforation.The possible mechanism is to reduce the permeability of blood-brain barrier by inhibiting the pyroptosis of hippocampal microglia and increasing the expression of ZO-1 through HO-1/NRF2.Our study further verified the conclusion of retrospective clinical studies that magnesium sulfate improves the prognosis of SAE.It also demonstrated that magnesium sulfate improves the symptoms of memory impairment in SAE,and proposed a possible mechanism for improving the blood-brain barrier function in SAE,providing a potential,safe and cost-effective drug for sepsis-associated encephalopathy.