Clinical Study On The Value Of NLR,HS-mGPS And GCS Scores In Predicting Agi In Patiens With Severe Ischemic Stroke

Objective: To explore the correlation between blood test results and scoring scales collected within 24 hours of admission of patients with severe ischemic stroke and the occurrence of acute gastrointestinal injury(AGI),and to explore their predictive value for acute gastrointestinal injury.Methods:A collection of 207 patients with severe ischemic stroke who were hospitalized in the Intensive Care Unit of the Department of Neurology,affiliated Hospital of Yanbian University from September 2017 to September 2020.The NIHSS score and GCS score at admission were recorded,and the results of neutrophil count,lymphocyte count,albumin and hs-CRP within 24 hours after admission were collected.According to the acute gastrointestinal injury classification standard,the patients were classified on the 4th day of hospitalization,and they were divided into no AGI group and AGI group.The Logistic regression method was used to analyze the factors affecting the occurrence of AGI.ROC curve was used to analyze the predictive value of NLR,GCS score,HS-m GPS score and combined indexes in the occurrence of AGI in patients with severe ischemic stroke.Results:1.Comparing the age,gender,ethnicity,hypertension,and diabetes between the AGI-free group and the AGI group at the 4 days hospitalization,the difference was not statistically significant(P>0.05).2.In comparison between the two groups,the white blood cell,neutrophil count,lymphocyte count,hs-CRP,NIHSS score,GCS score,NLR,HS-m GPS score were significantly different(P < 0.05 or P < 0.001).There was no statistically significant difference among platelets,blood glucose,blood K,blood Na,blood Cl,prealbumin,and albumin(P>0.05)3.Multivariate Logistic regression analysis showed that NLR,GCS score and HS-m GPS score were independent predictors of AGI in patients with severe ischemic stroke.4.The ROC curve results of NLR,GCS score and HS-m GPS score for predicting the occurrence of AGI in patients with severe ischemic stroke are as follows:The area under the curve NLR was 0.779,GCS score was 0.689,HS-m GPS score was 0.617,HS-m GPS+GCS score was 0.743,NLR+HS-m GPS score was 0.790,NLR+GCS score was 0.824,the combination of the three is 0.833,the area under the curve is the largest,the specificity is 77.9%,and the sensitivity is 79.9%.5.The comparison of ROC curve showed that NLR was superior to HS-m GPS in predicting the occurrence of AGI in patients with severe ischemic stroke(P<0.001).The combined prediction of NLR and GCS score and the combination of the three were better than those of single index(P<0.05 or P<0.001).Conclusions:1.NLR,HS-m GPS,GCS score has a certain predictive value of AGI in patients with severe ischemic stroke.2.The combination of the three indexes has higher predictive value than a single index,which can provide a reference basis for early identification of the occurrence of AGI in clinical work.