Risk score model for prediction of contrast-induced nephropathy after percutaneous coronary intervention

Abstract

Background The use of contrast media has been closely related to contrast-induced nephropathy (CIN) in several studies. The volume of contrast media and patient characteristics may be important risk factors for CIN. Identifying and quantifying these risk factors may be useful in predicting risk of CIN. Objectives To (1) identify important risk factors associated with CIN, and (2) develop a risk score model for prediction of CIN after percutaneous coronary intervention (PCI). Methods A total of 181 patients underwent PCI were enrolled. Patient- and procedure-related factors were prospectively collected. CIN was defined as an increase in serum creatinine > 25% or > 0.5 mg/dl from pre-PCI value within 72 hours after PCI or diagnosed with CIN. Receiver-operator characteristics (ROC) methods were used to determine the optimal cutoff point of contrast media volume (V) and volume/body weight (V/BW) ratio. All risk factors including the cutoff point values obtained from ROC curve analysis were tested for an association with CIN by multivariate logistic regression analysis. A CIN risk score model was developed by using odds ratio from multivariate analysis as risk score values for each of risk factor. The relationship between risk score and occurrence of CIN after PCI was evaluated. Results The incidence of CIN after PCI was 6.1%. The ROC curve analysis indicated that V/BW ratio of 2.6 ml/kg was a good discriminator for CIN with concordance statistic (C-statistic) of 0.73. Multivariate logistic regression analysis showed that V/BW ratio  2.6 ml/kg (OR 8.184; p=0.003), congestive heart failure (CHF) (OR 6.465; p=0.010), and creatinine clearance (CrCl) < 30 ml/min (OR 6.141; p=0.019) were associated with CIN after PCI. The CIN risk score model demonstrated good discriminative ability with C-statistic of 0.849. The incidence of CIN increases with risk score. Conclusion The CIN risk score model, incorporating 3 risk predictors including V/BW ratio  2.6 ml/kg, CrCl < 30 ml/min, and CHF, can be used as a tool to assess individual patient risk of developing CIN after PCI.