Contrast medium-induced nephropathy: critical review of the existing clinical evidence

Abstract

Contrast agents have evolved remarkably since their use was first considered in 1896, shortly after the first description of X-rays by Roentgen. First attempts include sodium iodide, among many other both simple and complex compounds. By the 1950s, using both science and trial-and-error, radiographic contrast agents had evolved to ionic monomers (e.g. Conray, Renografin, Urografin) that consist of a fully substituted benzene ring with three attached iodine atoms to impart radio-opacity, and a dissociating side chain. The osmolality of this class of contrast agents is in the range of 1200–2000mOsm/kg. In the late 1970s, to address concerns relating to hypertonicity and related toxicity, Almen developed the first non-ionic monomer, a contrast agent with identical radio-opacity but, because there was no longer a dissociating side chain, with markedly reduced osmolality. These were released in the 1980s and have been in wide use since then. Simultaneously, an ionic dimer was developed, with similarly decreased osmolality. This osmolality was closer to that of blood ( 300mOsm/kg), and led to less discomfort and a marked decrease in the incidence of at least minor adverse events. In the 1990s, two non-ionic dimeric contrast agents were developed. Currently, only one is commercially available for intravenous application in Europe and the USA. This is iodixanol. Since this is a non-ionic dimeric contrast agent with six iodine atoms per molecule, rather than three as in the non-ionic monomer, its osmolality is theoretically decreased by 50%, and is close to that of blood. In theory, this has led to a further decrease in adverse events. With lower osmolality and a decrease in the incidence of at least minor adverse events, there has been increasing focus on events that may have clinical significance. Contrast nephrotoxicity is one such area of concern. It has generated much interest over the last few years, for several reasons: first, it is potentially a significant clinical concern. Secondly, with the ageing of the population, the incidence of renal dysfunction is increasing. Simultaneously, the utilization of contrast agents is increasing markedly. A final, very important consideration is that, as a number of recent papers have suggested, the incidence and severity of contrast medium-induced nephropathy (CIN) can be decreased. With this in mind, it is important to review the available data regarding all aspects of CIN. CIN goes by many different names: contrast nephropathy, contrast nephrotoxicity, contrast media nephropathy, contrast agent nephropathy, radiocontrast-induced nephropathy, and others. The multiplicity of names is, perhaps, emblematic of the level of understanding of this entity. Accompanying this lack of a clear name is lack of a clear definition of the entity. The definitions that have been used include a 50% increase in serum creatinine, a 25% increase in serum creatinine, a 0.5 or 1.0mg/dl increase in serum creatinine or a percentage decrease in actual or calculated creatinine clearance. As will be discussed subsequently, it is important to unify the definition of CIN, as it is important to define the best name for it. In this review, this entity will be referred to as contrast medium-induced nephropathy (CIN), but a universally acceptable definition is more difficult to arrive at. The most apt definition currently is probably a defined decrease in calculated creatinine clearance. The history of CIN is interesting. It was perhaps first described in 1955 by Alwall et al., in an article describing the course of renal failure after intravenous urography [1]. In 1968, the entity had achieved sufficient importance to be discussed in an article in theNew England Journal of Medicine [2]. An influential article in 1978 [3] described renal failure following major angiography, and was perhaps the first to indicate that the key risk factor was the presence of underlying renal dysfunction. Subsequent articles [4,5] suggested that CIN was one of the major causes of in-hospital renal failure. The incidence of CIN is difficult to define, since it is a function of its specific definition and the presence or absence of risk factors. There are several important Correspondence and offprint requests to: Michael A. Bettmann, MD, Department of Radiology, Dartmouth-Hitchcock Medical Center, Lebanon, NH 03756, USA. Email: mab9099@nyp.org