External Validation Of The Risk Assessment Tool At The End Of Operation

Background:More than 200 million adults worldwide undergo major non-cardiac surgery each year,and the number of such patients continues to grow.The current researches focus on the evaluation of the patient’s condition before and after the operation.There has not been lots of reports on the objective criteria of the intra-operative evaluation to determine whether patients transfer to the intensive care unit(ICU)after surgery.However,for critically ill patients undergoing elective major surgery,they often experience surgical trauma,stress response,hemorrhage and blood transfusion,organ hypoperfusion,ischemia and hypoxia,body temperature change,and drug residual during the operation.The patient’s organ function and internal environment balance have been greatly changed.All of this requires the anestbhesiologist to re-evaluate the patient’s risk during the operation or at the end of the operation to clarify the optimal plan for the patient’s condition during the operation.To provide a relatively objective standard of transferring to ICU or the post-anesthesia care unit.However,there is currently lack of a relatively objective intra-operative risk assessment tool,which require researchers to further exploring a prediction model for this type of patients.The first step,we conducted a retrospective study to establish a model-"Intra-operative Anesthesia Assessment Score",which showed that:1)In patients who died in ICU,the "Intra-operative Anesthesia Assessment Score" was more relevant than the APACHE Ⅱ score system;2)The "Intra-operative Anesthesia Assessment Score" is more relevant than the APACHE Ⅱ score system in terms of ICU stay time.In order to establish the universality of the "Intra-operative Anesthesia Assessment Score",we try to use patient data different from the established model to conduct external verification experiments on the basis of the previous retrospective study.Objective:For critically ill patients undergoing elective major abdominal surgery,where to go after surgery is often one of the common concerns of clinicians and anesthesiologists.On the basis of the preliminary multi-center research,we need to provide effective external verification.Method:Patients admitted to the Third Affiliated Hospital of Southern Medical University for abdominal surgery from January 2019 to July 2020.We put all variables into the Logistic regression model with whether the patient went to the intensive care unit(ICU)as the outcome variable,using the receiver operating characteristic curve area and The Hosmer-Lemeshow goodness-of-fit evaluates the discriminability of the model,comparing the observed patients entering the ICU with the predicted.Results:The p value of Hosmer-Lemeshow goodness-of-fit test is 0.821,P>0.05,indicating that the whole model is a significant fit.The Logistic model constructed by the 18 indicators of the risk score model has an accurate prediction rate of 88.7%,and its receiver operating characteristic curve area is 0.936,which has high accuracy.Conclusion:Considering that there are many factors affecting the admission of patients to ICU,the regression coefficients of individual indicators may differ greatly from the scores given by Intra-operative Anesthesia Assessment Score.However,the accuracy of the entire model and the area under the receiver operating characteristic curve shows that the Intra-operative Anesthesia Assessment Score can provide a reference for critically ill patients undergoing elective major surgery.It has extremely high clinical application value.