A Simple Clinical Rule To Predict The Prognosis Of Acute Cerebral Infarction In One Month

Objectives To develop and validate a simple risk score to evaluate the prognosis of patients with acute cerebral infarction in one month. The purpose is designed to simplify the statistical equation model, so that it is convenient to be used in clinic.Methods The study includes 689 cases of acute cerebral infarction patients admitted within 72 hours. The modified Rankin Scale(m RS) score at one month is used as assessment indicator. Unfavorable outcome is defined as life depend(m RS:3-5 points) or died within one month(m RS:6 points), favorable outcome is defined as m RS: 0-2 points.All patients were divided into favorable outcome group( 541 cases) and unfavorable outcome group( 148 cases). Some factors were collected, including gender, age,previous stroke / TIA history, history of hypertension, history of diabetes, history of coronary artery disease, history of atrial fibrillation, history of smoking, the classification of OCSP, time from onset to admission, NIHSS, body temperature, blood pressure, blood cell, blood glucose, serum creatinine, blood urea nitrogen, uric acid, albumin, fibrinogen,triglycerides, total cholesterol, high density lipoprotein and low density lipoprotein. The univariate analysis and multivariate Logistic regression analysis(Forward LR) were used to select risk factors. The simple risk score for each covariate of the fitted multivariate model was generated by their β coefficients. The overall score was calculated as the sum of the weighted scores. The model was validated by ROC curve, consistency test and Mc Nemar test. At last, a further study is to check the accuracy of the model.Results 1 Univariate analysis: previous stroke /TIA history, history of atrial fibrillation,history of diabetes, history of coronary heart disease, hypertension, OCSP, age, NIHSS,fibrinogen, albumin, blood glucose and systolic blood pressure were significantly different(P<0.05) between the favorable outcome group and the unfavorable outcome group; sex, smoking history, time of onset to admission, admission temperature, diastolic blood pressure, blood cell, urea nitrogen, serum creatinine, uric acid and lipid levels were not significant(P>0.05)between two groups. 2 Multivariate Logistic regression analysis:age, NIHSS on admission, blood glucose, previous stroke/TIA history and history of atrial fibrillation were significant(P<0.05) between two groups. The risk of unfavorable outcome were as follows: 1.208 times, 1.841 times, 3.992 times, 2.697 times, 5.730 times.3 The prognosis model of acute cerebral infarction is established for: Logit P=-19.258+0.189X1+0.610X2+1.746X3+ 1.384X4 +0.985X5.(Note: X1: age, X2: NIHSS, X3:history of atrial fibrillation, X4: blood sugar, X5: previous stroke / TIA history). 4 The risk score is 13 points with the AUC of 0.934, the result shows as follows: sensitivity(75.2%), specificity(95.9%), accuracy(0.914), Kappa value(0.738). 5 The model is futher tested by 269 patients, the result shows as follows: sensitivity(75.0%), specificity(91.1%), accuracy(0.885), Kappa value(0.611).Conclusions 1 Age, NIHSS, blood glucose, history of atrial fibrillation and history of previous stroke/TIA are main factors of acute cerebral infarction in one month. 2 The risk score is a simple rule to predict the outcome of acute cerebral infarction. It performed well in both calibration and discrimination. Therefore, It is a useful tool for clinical practice.