A Novel Tool For Prediction Of Contrast-induced Acute Kidney Injury In Patients Undergoing Percutaneous Coronary Intervention

Objective:To develop a new and reliable tool for predicting the risk of contrast-induced acute kidney injury(CI-AKI)in patients undergoing percutaneous coronary intervention(PCI)Methods:This study retrospectively analyzed the clinical data of patients who treated PCI in our hospital from January 2020 to August 2020.The variables related to CI-AKI after PCI were found by univariate analysis.The receiver-operating characteristic curve(ROC)was used to test the accuracy of all the significant variables in the univariate analysis,and the area under the curve(AUC)was used as the measurement standard of accuracy.Five variables with the largest AUC were selected,then,multivariate regression analysis was used to test their independent effect on CI-AKI.The nomogram diagram of the risk model was drawn with the variables that meet the above conditions,and the regression equation of the risk model is calculated.Finally,the prediction performance of the model is tested by ROC curve,calibration curve and decision curve.Results:A total of 853 patients were included in this study,including CI-AKI group(n =86)and non-CI-AKI group(n = 767).The independent factors related to CI-AKI are evaluated by univariate analysis,a total of 19 variables are significantly correlated with CI-AKI,among which the five largest independent factors of AUC are BNP(AUC 0.853),LVEF(AUC 0.826),e GFR(AUC 0.774),HB(AUC 0.758)and age(AUC 0.757).Through the logistic regression model,the principle of AIC minimization was adopted.The regression equation of this model is 2.26979+0.06313*AGE-0.09118*LVEF-0.02389*HB +0.00045*BNP-0.01815*EGFR.The accuracy of this model is compared with that of other risk models by ROC,and it is found that this model has the highest accuracy(AUC 0.91)and is the best predictor with good calibration and clinical benefit.Conclusion:In this study,We have developed a simple tool limited to BNP,LVEF,e GFR,HB and age that can accurately predict the risk of CI-AKI in patients undergoing PCI.