Objectives: To develop a simplified scoring system based on preprocedure clinical characteristics to predict contrast-induced nephropathy (CIN) before elective coronary angiography and percutaneous coronary intervention (PCI).
Background: CIN is associated with increased mortality and morbidity following coronary angiography and PCI and accounts for increased hospital costs.
Methods: Several baseline clinical characteristics of 1218 patients were considered as candidate univariate predictors of CIN (increase > or =0.5 mg/dl in serum creatinine within 5 days after contrast exposure). On the basis of the odds ratio at multivariate logistic regression, seven markers (with weighted scores) were identified as independent correlates of CIN: age at least 73 years (1), diabetes mellitus (2), left ventricular ejection fraction 45% or less (2), baseline serum creatinine value at least 1.5 mg/dl (2), baseline creatinine clearance 44 ml/min or less (2), posthydration creatinine > or = prehydration creatinine value (2) and one procedure effected within the past 72 h (3).
Results: CIN occurred in 114 (9.4%) patients [range 1.1-52.1% for a low (< or =3) and very high (> or =9) risk score, respectively]; the odds of CIN increased significantly with each class (Cochran-Armitage chi-square, P < 0.0001) and the risk score allowed us to determine patients with low and high risk for postprocedure CIN (c-statistic = 0.86). These results were reproduced in a validation set.
Conclusion: Preprocedural clinical risk factors have different influences on the likelihood of CIN. Risk classification based on the most significant parameters can be used to predict CIN before contrast exposure. The simple scoring system proposed here provides a good estimate of the risk of CIN, allowing the interventional team to make adequate adjustment to the procedures.