In order to make appropriate decisions about clinical management, it is important for physicians to be able to stratify patients according to their risk for contrast-induced nephropathy (CIN). The most important risk marker for nephropathy after exposure to iodinated contrast media is preexisting renal impairment. The risk of CIN is elevated and becomes clinically important in patients with chronic kidney disease characterized by an estimated glomerular filtration rate <60 mL/min per 1.73 m(2). In patients with renal impairment, diabetes mellitus amplifies the risk of CIN and complicates postprocedure management. Other markers associated with an increased risk of CIN include cardiovascular disease, periprocedural hemodynamic instability, use of nephrotoxic drugs, and anemia. The effect of risk factors is additive, and the presence of multiple risk factors in the same patient can create a very high risk for CIN and acute renal failure requiring dialysis. Risk models incorporating baseline and periprocedural characteristics have been developed using data from large databases of percutaneous coronary intervention patients. These schemes are potentially valuable, but at present the most practical approach to risk prediction is based on a simple model incorporating renal function and diabetes mellitus.