Construction And Evaluation Of Short~term Prognosis Prediction Scoring System For Acute Stroke Based On MEWS

Objective On the basis of MEWS score,combined with risk factors affecting short~term prognosis of acute stroke,a prognostic prediction scoring system was established to optimize the predictive efficacy for short~term prognosis of acute stroke,optimize clinical management and rational allocation of clinical resources,improve the ability of nurses to observe the disease,and provide a basis for clinical decision~making.Methods From August 1,2018 to September 30,2018,200 adult patients with acute stroke were enrolled in the first affiliated hospital of guangxi medical university,to construct a prognostic scoring system;104 adult patients with acute stroke who were first diagnosed from November 1,2018 to November 30,2018,used for validation followed up for 30 days,and the clinical outcomes after 30 days was counted.Outcome evaluation indexes were assessed by modified Rankin scale(m RS)score,with 0~2points being diagnosed as good prognosis and 3~6 points as poor prognosis.At admission,risk factors that may affect the short~term prognosis of patients were collected,including age,WBC,RBC distribution width,fibrinogen,INR,blood glucose,glycosylated hemoglobin,homocysteine,albumin,CRP,uric acid,urea nitrogen and 12 other factors.Univariate and multivariate Logistic regression analysis was used to finally determine independent risk factors affecting poor prognosis of acute stroke patients at 30 days,Scores were assigned based on and combined with OR values,and a prognostic scoring system was established in conjunction with MEWS.The ROC curve and the Hosmer Lemeshow method were used to comprehensively evaluate and verify the predictive efficacy of the prognostic scoring system,so as to test its clinical application value.Results(1)after univariate analysis,there were statistically significant differences in age,WBC,fibrinogen level,INR,admission blood glucose,and prognosis between the group with good prognosis and the group with poor prognosis in CRP(P <0.05).There were no statistically significant differences in RBC distribution width,hba1 c,albumin,homocysteine,uric acid,and urea nitrogen levels between the two groups(P >0.05).(2)Independent predictors of short-term prognosis in patients with acute stroke were determined by multivariate Logistic regression analysis,which were age,white blood cells and c-reactive protein,and their risk(OR)values were 1.047 times,1.469 times and 1.071 times,respectively.(3)The statistical equation model of risk factors was established as:Logit P=-7.127+0.046X1+0.385X2+0.067X3(note: X1: age,X2:leukocyte,X3: CRP),and the equation module with a total score of 40 was obtained,which was combined with MEWS to form the short~term prognosis prediction scoring system for stroke.(4)The total score of the final constructed prognostic prediction scoring system was 53 points,and the ROC curve was drawn.The area under the ROC curve was 0.875(P<0.001).Through paired chi~square test and consistency test(kappa value),the sensitivity was 82.2%,the specificity was 80.9%,the overall consistency was 0.815,and the kappa value was 0.628.By analyzing the coordinate value of ROC curve,it can be seen that when the score is 7,the prediction efficiency of the scoring system is the strongest,that is,7 is the optimal diagnostic boundary value of the scoring system.(5)the prognostic prediction score system was used to predict ischemic stroke and hemorrhagic stroke respectively.The area under the ROC curve predicted for patients with ischemic stroke was 0.889(P <0.001),and the prediction efficiency was the strongest when the score was 7 points.The area under the ROC curve predicted for patients with hemorrhagic stroke was 0.828(P <0.001),and the prediction efficiency was the strongest when the score was 11 points.(6)The results of the empirical group showed that the score of the prognostic prediction scoring system was closely related to the rate of poor prognosis,which proved that the prognostic prediction scoring system had moderate predictive ability.The area under the ROC curve was 0.823,the sensitivity was 86.7%,the specificity was 87.0%,the overall consistency was 0.817,and the kappa value was 0.635.According to the prognostic scoring system,when the total score ?7,the rate of poor prognosis was 75.5%.The total score <7was 11.5%.Conclusion(1)Patients' age,white blood cell count and c-reactive protein level at admission have significant influence on short-term prognosis of acute stroke patients,which can be used as independent predictors of short-term prognosis of acute stroke.(2)The short~term prognostic and scoring system for acute stroke is simple and convenient to use.(3)The prognostic scoring system has been verified to have a better predictive effect on the short~term prognosis of patients with acute stroke,which is better than that of MEWS.