Construction and Evaluation of a Nomogram Prediction Model for the Risk of Stress Hyperglycemia in Patients with Acute Myocardial Infarction
Objective:To analyzed the influencing factors of the risk of stress hyperglycemia(SHG)risk in patients with acute myocardial infarction(AMI),construction and evaluation of nomogram prediction model.Methods:298 AMI patients were divided into occurrence group(n=92)and non-occurrence group(n=206)according to whether SHG occurred.The influencing factors of SHG in AMI patients wewe analyzed by multivariate Logistic regression model,and a nomogram prediction model of influencing factors was con-structed,the predictive value of the nomogram prediction model for SHG in AMI patients was analyzed by receiver operating characteris-tic(ROC)curve.Results:Female,high body mass index(BMI),cardiac function Killip classification ≥ class Ⅱ,advanced age,high hs-CRP and high Cys-C were independent risk factors for SHG in AMI patients(P＜0.05).The results of ROC curve analysis showed that,the area under the curve(AUC)of gender,age,BMI,cardiac function Killip classification,hs-CRP,Cys-C levels and nomogram predic-tion model were 0.609,0.711,0.826,0.618,0.768,0.774 and 0.953 respectively.The internal validation results showed that,the Bias-corrected prediction curve basically coincided with the Ideal line,and the C-index was 0.905,which indicated that the model had good predictive ability.The decision curve shows that,the threshold probability of the model ranges was 0.01～0.97,and its net return was＞0,which was higher than the two null lines.Conclusion:Cardiac function Killip grade ≥ Ⅱ,advanced age,female,high BMI,high hs-CRP and high Cys-C were independent risk factors for SHG in AMI patients.The prediction ability of the nomogram model based on the above factors is better than that of a single index.
Acute myocardial infarctionStress hyperglycemiaNomogramPredictive value
NETL
NSTL
万方数据
