Contrast-induced nephropathy (CIN) represents an increasing healthcare burden and challenge as the frequency of diagnostic imaging and interventional procedures increases, particularly among patients at risk for developing CIN. As the population ages, decreased renal function and increased atherosclerotic cardiovascular disease become more prevalent. An increasing incidence of obesity with resultant metabolic syndrome and/or adult diabetes also increases the population at risk for CIN.
Clinical Implications of Contrast-induced Nephropathy
CIN is the third most common cause of hospital-acquired acute renal failure.1 The incidence of CIN is <5% in patients with normal renal function and 15–50% in patients with renal dysfunction. The incidence of dialysis-dependent acute tubular necrosis is 1.3–19%. CIN is an indicator of a marked increase in short-term and late mortality.2 Acute renal failure after coronary intervention is associated with a 36% in-hospital mortality rate and a 19% two-year survival rate.2
Strategies to Reduce the Risk for Contrast-induced Nephropathy
The pathogenesis of CIN is complex, with a cascade of contributing factors (see Figure 1). Besides CIN, there are other causes of acute renal dysfunction after angiography, including atheroemboli and hypotension.3 Examples of sudden hypotension during angiography include acute ST-elevation myocardial infarction and balloon dilatation during carotid stenting. At present, CIN cannot be prevented, but it can be attenuated.4 Universally accepted strategies to reduce the risk for CIN include careful patient screening and selection, adequate patient hydration, limiting the volume of contrast medium (CM) administered, and choosing a safe, non-ionic, low-osmolar contrast agent.5 In view of the clinical relationship between CIN and patient morbidity and mortality, identifying patients at risk and considering whether exposure to CM is necessary for diagnosis and/or intervention is implicit. Recent guidelines have advocated routine use of the estimated glomerular filtration rate (eGFR) to identify patients with chronic kidney disease because serum creatinine is an insensitive measure of kidney dysfunction.6 Patients who are at particular risk for developing CIN include those with an eGFR <60ml/min/1.73m2.3
All patients receiving CM should maintain adequate hydration. Intravenous volume expansion with isotonic saline (1.0–1.5ml/kg/h) for three to 12 hours before the procedure and continued for six to 24 hours has been shown to attenuate CIN.5 An additional intravenous volume expansion strategy that is under study is administering isotonic sodium bicarbonate solution at 3ml/kg/h for one hour before angiography and at 1ml/kg/h for six hours afterward if 151–299ml of CM is administered and for 12 hours if ≥300ml of CM is used.7 It is thought that an alkaline environment reduces oxygen-free radical formation in the renal tubule.
Because CM is primarily eliminated through the kidneys, it is reasonable to assume that CM molecules interact with the tubular epithelium. Over the years, there has been ongoing debate as to whether different types of CM have different mechanisms of nephrotoxicity or produce different degrees of renal effects. Table 1 lists by osmolality the physiochemical characteristics of water-soluble, iodinated CM available for use in the US for cardiac procedures. When choosing contrast agents to reduce renal risk, each molecule may stress the kidney through osmotic load and/or direct toxicity (see Figure 1). Each molecule is unique and requires comparison with the other CM.
Comparing Contrast Media in Cell Culture
The direct cellular toxicities of commonly used CM have been compared in renal proximal tubular cells in culture.8 All CM, whether high-osmolar or iso-osmolar, ionic or non-ionic, stress renal proximal tubular cells. When renal proximal tubular cells in culture are exposed to CM, they respond by increasing extracellular adenosine concentrations because of the depletion of adenosine triphosphate due to osmotic load and the large size of CM molecules. The renal toxicity from the direct effects of CM is reversible, as has been shown in in vitro studies in which renal tubular cells responded to CM exposure by decreasing the activity of mitochondrial enzymes without altering viability. While all CM may reduce mitochondrial enzyme activity, differences among agents are seen, with iopamidol and iodixanol being the least ‘toxic’ and the two ionic CM, ioxaglate and diatrizoate, being the most ‘toxic.’ The lesser the depression of mitochondrial function, the more rapidly renal proximal tubular cells can recover after CM has been eliminated from the kidney, suggesting that iopamidol and iodixanol are the least nephrotoxic molecules when studied in cell culture.
Clinical Trial Comparison of Contrast Media Administered Intra-arterially
Earlier clinical trials comparing CM risk for CIN had limitations that prevented valid conclusions: the studies were not prospective or not blinded, the number of patients was small, the timing of outcome assessment was unclear, or the patients received contrast intra-arterially and/or intravenously. More recent trials have been head-to-head, robust, and prospective in high-risk patients with chronic renal failure (eGFR 59–20ml/min) with intra-arterial CM administration. For clinical and research purposes, CIN is defined as an acute decline in renal function (rise in serum creatinine level by 25% or >0.5mg/dl from baseline or fall in eGFR of >25%) after systemic contrast administration in the absence of other causes.9 Based on animal studies, concern about osmotoxicity in the pathogenesis of CIN emerged, which led to the development of low-osmolar and, later, iso-osmolar CM. Of interest, in human prospective trials and meta-analyses, no statistically significant differences in nephrotoxicity between high-osmolar (1,400–2,000mOsm/kg) and non-ionic low-osmolar (600–800mOsm/kg) contrast has been found in patients with normal renal function.10 Clinical trials published in the 1990s, however, clearly showed that use of the high-osmolar ionic monomer diatrizoate in patients with chronic renal disease who were undergoing coronary angiography had a higher incidence of CIN than that associated with non-ionic low-osmolar CM.11,12 In general, lower-osmolar CM have now replaced diatrizoate for routine clinical use.
It is less clear whether there are appreciable differences among the various non-ionic low-osmolar CM regarding the incidence of CIN. Because individual CM have specific effects on renal tubular cells, head-to-head studies are required to compare the safer CM in at-risk patients undergoing angiography. With the development of the non-ionic iso-osmolar dimer iodixanol, subsequent studies have evaluated whether a further reduction in osmolarity would result in still more protection against CIN. In the first of these comparative studies (Nephrotoxicity in High-Risk Patients Study of Iso-Osmolar and Low-Osmolar Non-Ionic Contrast Media [NEPHRIC]), 129 patients with diabetes with chronic renal failure (baseline eGFR 48ml/min; serum creatinine level 1.5–3.5mg/dl) were randomized to the low-osmolar CM iohexol or the iso-osmolar CM iodixanol.13 A higher rate of CIN (defined as a serum creatinine rise of >0.5mg/dl after angiography) was noted after iohexol (26.2%) than after iodixanol (3.1%).
Since the publication of NEPHRIC, subsequent prospective randomized trials involving high-risk patients with renal insufficiency and the intra-arterial administration of CM have not consistently found a lower incidence of CIN associated with iso-osmolar contrast. In general, a benefit favoring the non-ionic iso-osmolar dimer iodixanol is seen compared with the non-ionic low-osmolar moner iohexol or the ionic low-osmolar dimer ioxaglate.13–16 In contrast, when comparing non-ionic low-osmolar monomers (iopamidol or ioversol) in high-risk patients with iso-osmolar iodixanol, no additional protective effect is seen.17,18 There have been no head-to-head comparisons of iohexol with any other non-ionic low-osmolar contrast agents in high-risk patients undergoing intra-arterial administration of contrast. Based on a recent systematic review of angiographic CM in high-risk patients, however, the likely explanation for the NEPHRIC findings is that low-osmolar iohexol appears to be more nephrotoxic than other low-osmolar CM, such as iopamidol and ioversol.10 Furthermore, a systematic review of angiographic CM in high-risk patients found no statistically significant difference in the risk for CIN between low-osmolar iopamidol and iso-osmolar iodixanol.10 These comparisons of low-osmolar and iso-osmolar CM administered intra-arterially in high-risk patients suggest that each molecule’s safety must be based on robust clinical trials. These data also suggest that physical and/or chemical properties of each molecule other than osmolarity are implicated in the pathogenesis of CIN.
With proper hydration and identification of high-risk patients, the incidence of CIN after intra-arterial CM administration is about 10%.19 Given this incidence of CIN, prospective randomized trials need larger patient populations to be properly powered to be certain of the conclusions. The recently published Cardiac Angiography in Renally Impaired Patients (CARE) study fits these criteria by being a multicenter, double-blind, randomized study designed to prospectively compare the incidence of CIN after intra-arterial administration of low-osmolar iopamidol or iso-osmolar iodixanol in 414 patients with moderate to severe chronic kidney disease (eGFR 20–59ml/min/1.73m2) who underwent cardiac angiography or percutaneous coronary interventions.18 All patients received intravenous bicarbonate prophylaxis. The primary CIN end-point was a post-dose serum creatinine increase of ≥0.5mg/dl over baseline. The renal effects of these two agents were comparable, with no significant difference in the occurrence of CIN overall and no significant difference in the occurrence of CIN in the subgroup of chronic kidney disease patients with diabetes.
Clinical Trial Comparison of Contrast Media Administered Intravenously
More recent trials in high-risk patients given CM intravenously suggest that the incidence of CIN is about 5%. Given this lower incidence of CIN, fewer prospective randomized trials have been performed, and those that have been are likely underpowered. For the commonly used CM in the US, there has been one randomized double-blind comparison of the renal tolerability of low-osmolar iopamidol and iso-osmolar iodixanol intravenously administered for computed tomography in 153 high-risk patients (baseline eGFR 44ml/min/1.73m2).20 CIN was defined as a post-contrast rise in serum creatinine of at least 0.5mg/dl. The renal effects of these two contrast agents were comparable (CIN rates: iopamidol 3.9%; iodixanol 4.0%).
Summary
For both intra-arterial and intravenous use, all ionic and non-ionic iodinated contrast agents may further impair renal function in high-risk patients. Non-ionic low-osmolar CM are less nephrotoxic than ionic high-osmolar CM in patients with chronic renal failure, however. Data suggest that there may be significant differences among low-osmolar CM in terms of their nephrotoxic potential, and the degree of nephrotoxicity of low-osmolar CM may be associated more with the molecular properties than with osmolality. The risk for the development of CIN in patients with renal impairment appears lower after iso-osmolar iodixanol or low-osmolar iopamidol than after iohexol use. During both intra-arterial and intravenous use, the risk for CIN is similar with iodixanol and iopamidol in patients with pre-existing renal disease.
Conclusions
Because only a small number of different low-osmolar CM have been studied in robust head-to-head prospective trials with iso-osmolar CM, it is not possible to conclude that all low-osmolar CM are equal. In high-risk patients, use of an iso-osmolar CM or one of the low-osmolar CM with a documented low incidence of CIN is recommended. At this time, iodixanol and iopamidol appear to be the contrast agents of choice to reduce renal risk.
Practical Clinical Applications. The renal effects of iodixanol and iopamidol are comparable in high-risk patients requiring CM for diagnostic studies or intervention. The non-renal differences between these contrast agents should be considered when selecting an agent and when obtaining informed patient consent, however (see Table 2). Using a CM with a higher iodine concentration per milliliter may allow for a smaller volume to be used intravenously without sacrificing image quality. The higher iodine concentration may make complex plaques easier to appreciate during coronary angiography and intervention. In addition, iodixanol is three-fold more likely to cause contrast-induced delayed skin reactions compared with iopamidol.21 After percutaneous coronary intervention with bare-metal or drug-eluting stent placement, cardiac patients require antiplatelet therapy with clopidogrel, which may cause a rash within five to seven days of treatment initiation, a timeline similar to that for delayed contrast-induced skin reactions. In clinical trials with clopidogrel, skin reactions are noted in 4.2% of patients.22 The choice of a contrast molecule that is less likely to cause a delayed contrast-induced skin rash makes decisions regarding the discontinuation of clopidogrel less frequent and more definitive. Finally, iopamidol is less expensive.
Gadolinium-based contrast agents were introduced partly because of the discovery that iodine-based CM could cause CIN; however, recent reports suggest that gadolinium-based agents may also be nephrotoxic.23 Furthermore, after exposure to gadolinium-based contrast, some patients with renal insufficiency have developed nephrogenic systemic fibrosis with scleroderma-like changes in the skin, connective tissues, and other organs, which has sometimes been fatal.24–26 In patients with renal insufficiency, use of any contrast agent should be avoided if possible, but, if required, small volumes of contrast agents, such as iodixanol or iopamidol, with proper hydration appears to be the safest clinical management approach to reduce renal and long-term patient risk.