Introduction

Contrast media (CM) are widely used in radiology. The purpose of any contrast medium is to enhance the contrast in the image to allow better tissue characterization and improved diagnosis of pathological conditions. Inflammatory and malignant processes will in most cases have increased uptake of CM after intravenous administration. Various types of CM exist and their chemical and physical properties determine their application. Iodine-based CM increases the attenuation of X-ray beams. Hence iodine-based CM are nearly only applied for computed tomography (CT) and interventional procedures using fluoroscopy. Gadolinium-based contrast agents used for magnetic resonance imaging (MRI) shortens tissue T1 and T2 relaxation times leading to altered signal characteristics. Finally, CM used for ultrasound are based on microbubbles leading to increased echoes from the investigated organs.

The ideal CM should improve image contrast and furthermore it should be administered safely without causing any adverse events. Unfortunately, as with other drugs, adverse events may follow the administration of CM. Of particular interest has been the decrease in renal function that can occur after administration of iodine-based CM during CT or interventional procedures (e.g., catheter angiography). This condition is known as contrast medium-induced nephropathy (CIN).

The purpose of this review article is to provide an updated overview of the current knowledge of CIN.

Chemistry of Iodine-Based Contrast Media

This section gives a brief overview of the chemical properties of iodine-based CM. All of these CM diffuse throughout the extracellular compartment after intravascular administration. Iodine-based contrast agents are based on a benzene ring to which 3 iodine atoms are attached. A monomeric contrast agent consists of one benzene ring, whereas a dimeric agent consists of two linked benzene rings. Another important chemical property of the iodine-based contrast agents is the electric charge, i.e., ionic or non-ionic. The water-soluble mechanism is largely dependent on this property.

Both chemical structure (monomer vs. dimer) and electric charge determine the osmolarity of the contrast agents. Osmolarity is an important property as high osmolarity is related to adverse reactions, especially the acute ones. The limit for osmotoxicity is approximately 800 mosm kg−1, and this is also cut-off level between high- and low-osmolar agents.

Iodine-based CM are classified into the following groups:

  • High-osmolar CM (>800 mosm kg−1): monomeric and ionic contrast agents

  • Low-osmolar CM (<800 mosm kg−1): monomeric and non-ionic contrast agents and dimeric ionic contrast agents

  • Iso-osmolar CM (approximately 300 mosm kg−1): dimeric and non-ionic contrast agents

A more comprehensive list of the iodine-based CM is presented in Table 1.

Table 1 Iodine-based contrast media
Full size table

The focus of this review article is the low- and iso-osmolar CM as these are the most used in current radiology practice.

CIN

According to the guidelines on CM published by the European Society for Urogenital Radiology (ESUR) contrast medium-induced nephropathy (CIN) is defined as a condition in which a decrease in renal function occurs within 3 days of the intravascular administration of a CM in the absence of an alternative etiology. An increase in serum creatinine by more than 25% (relative definition) or 44 µmol/l (0.5 mg/dl) (absolute definition) indicates CIN [1]. The term post-contrast acute kidney injury (PC-AKI) is sometimes instead of CIN. Throughout this review article the term CIN is used.

Contrast Media as the Causative Factor

The true incidence of CIN after intravenous injection may previously have been overestimated. Comparative studies including control groups (not injected with CM) have suggested that CIN is over-reported [2•, 3, 4]. A possible explanation is temporal fluctuations in plasma creatinine levels due to physiologic variation and competing nephrotoxic risk factors in the investigated patients. A meta-analysis of 13 controlled clinical trials did not find an increased risk of CIN in patients after intravenous injection of CM [5]. Only four studies were prospective, and there seems a need for future studies with control groups [6].

Direct comparison of CIN incidence in patients after both intravenous and intra-arterial CM administration has been performed in a limited number of clinical trials [7, 8, 9•, 10•]. The overall results of these trials have been that the incidence of CIN, as well as the clinical course of the acute kidney disease, was similar for both methods of CM administration. Furthermore one of these studies showed that CIN rate was similar for unenhanced and CM-enhanced CT [10]. This underlines the above mentioned probable over-reporting of CIN.

The incidence of CIN after intra-arterial CM administration is often reported to be higher than after intravenous administration. However, this statement is controversial. A variety of confounding factors are in play when investigating CIN after interventional procedures such as coronary angiography or PCI. These factors include direct effects from catheter manipulation and comorbidities in (cardiologic) patients. As a precautionary strategy higher eGFR cut-off values are recommended for procedures with intra-arterial CM administration versus intravenous injections (see later section on eGFR cut-off values).

Risk Factors for Development of CIN

Some factors increase the risk of development of CIN. At the time of referral for imaging with use of intravascular CM, a risk assessment must be performed on an individual basis. The risk factors are categorized into the following groups: Patient related factors, procedure related, and CM related.

Patient Related Factors

Pre-existing kidney disease before CM administration is the most important risk factor for development of CIN after CM administration. Accordingly, renal function must be assessed before the imaging procedure (preferably by estimating GFR). Multiple other factors have been descried in the literature, including old age, female gender, heart failure, low BMI, dehydration, diabetes mellitus with albuminuria, inflammation, anemia, proteinuria, and presence of malignancy [11]. However, most of these factors are reported in studies without control groups not receiving CM, and the significance of these factors could not be confirmed when subjected to a meta-analysis [5].

CM Related Factors

Type of CM

The risk of CIN is higher for the older group of high-osmolar CM compared with the newer low-osmolar group of iodinated CM [12]. There has been a substantial interest in whether or not iso-osmolar CM are associated with a significantly lower rate of CIN than the low-osmolar CM. However, meta-analyses have shown no evidence to support this [13, 14].

CM Dose

CM dose relates to the amount of iodine administered—i.e., CM dose = CM concentration x CM volume. For intravenous injections there is little evidence that CM dose is an important factor; however, with newest CT-scanners the dose is lower than it was in the past. On the contrary, for intra-arterial injections (coronary angiography and interventions) it is advisable to keep the CM dose (gram iodine) to the absolute eGFR (ml/min, corrected for body surface area) ratio below 1.1 [15]. Another recommendation is to keep the CM volume (in ml) to eGFR (in ml/min/1.73 m2) ratio below 3.0 [16].

Assessment of Risk Factors Prior to Contrast-Enhanced Imaging

As already mentioned, the most important risk factor for CIN is kidney disease. The assessment of renal function can be performed by measuring plasma creatinine and calculation of the estimated glomerular filtration rate (eGFR). Another strategy is to use standardized questionnaires which include information on history of renal disease, renal surgery, diabetes, heart failure, hypertension, proteinuria, and gout. An example is the Choyke-questionnaire [17]. It has been shown that such questionnaires may detect patients with eGFR < 45 ml/min with adequate sensitivity, especially in the patient group below 70 years of age [18]. It remains to be determined if questionnaires can be used to detect patients with more severe kidney disease and renal function <30 ml/min. Although questionnaires may provide risk stratification of patients prior to contrast-enhanced imaging, it seems preferable to use eGFR testing in all patients. Indeed, eGFR testing has been proven to pick-up more patients with impaired renal function than questionnaires [19]. Furthermore, patient logistics are easier and cost-effectiveness similar [20].

eGFR Cut-Off Values for Prevention of CIN

The assessment of renal function (eGFR calculation) prior to contrast-enhanced imaging is used to determine if it is safe to administer CM or if preventive strategies must be applied. Different cut-off values are recommended for intravenous and intra-arterial administration [21]:

  • Preventive strategies are needed for patients with eGFR < 30 ml/min/1.73 m2 after intravenous or indirect arterial (extermities, carotids, mesenteric arteries or aorta below the renal arteries) administration of iodinated CM.

  • For patients with direct arterial (supra-renal aorta) administration of iodinated CM preventive strategies are needed for patients with eGFR < 45 ml/min/1.73 m2.

Strategies to Prevent CIN

Alternative Imaging Procedures

In patients with increased risk of CIN it shall be considered if other imaging procedures without the use of iodine CM are available to answer the clinical question. Examples are CT without use of intravenous CM, MRI, or ultrasound. However, it should always be remembered that inferior imaging studies without the use of CM can be harmful for the patient if they fail to diagnose severe pathology.

Withholding Nephrotoxic Medications

Temporary withholding of nephrotoxic medications (if clinically possible) is an important part of optimal nephrologic care of patients at risk of CIN. Such medications include NSAID, selective cox-2 inhibitors, several antibiotics, and chemotherapeutic agents. Few studies have investigated the effect of nephrotoxic agents on CIN. One retrospective study showed that concurrent use of 4 nephrotoxic medications was significantly predictive for CIN [22•]. Furthermore, a meta-analysis has shown that use of NSAID is an independent risk factor of CIN [11].

Hydration with Saline or Bicarbonate

A potential reno-protective strategy is to keep the patients well hydrated. Multiple studies have investigated the impact of various hydration strategies to prevent CIN. These strategies include oral hydration and intravenous hydration with saline or bicarbonate. Most studies have been performed in cardiologic patients undergoing percutaneous interventions. The following section summarizes the current evidence for the different hydration regimes.

Theoretically, oral hydration with intake of clear fluids by patients is an easy and cheap strategy. However, in practice compliance and monitoring is challenging. A number of randomized clinical trials have investigated the effect of oral hydration compared to intravenous hydration [23,24,25,26]. In recent years meta-analyses have been published with the conclusion that oral hydration is not associated with an increased risk of CIN compared to IV hydration [27, 28]. It should be noticed that the studies are heterogeneous and lack hard-clinical outcomes. Oral hydration with clear fluids should not be applied as an isolated measure to prevent CIN, but unrestricted oral fluids in addition to IV hydration may be beneficial.

A series of randomized clinical trials have compared the effectiveness of hydration to no hydration. The majority of these studies have been performed in patients undergoing coronary angiography and/or intervention [29,30,31]. One study has been performed in patients undergoing pulmonary CT-angiography [32]. All these studies show that hydration prior to contrast administration is a statistically significant effective method to reduce the incidence of CIN. Although the results are positive regarding incidence of CIN, no differences in mortality and other adverse events were seen between the study groups. Furthermore, few patients with severe renal impairment (defined as eGFR < 30 ml/min/1.73 m2) were included in the studies.

Most studies on hydration have been performed with normal saline or bicarbonate (NaHCO3). The rationale of bicarbonate administration is that alkalinization can reduce the formation of hydroxyl radicals and can directly trap reactive-oxygen species. Comparative studies between saline and bicarbonate hydration regimes have reported both favorable [33, 34] and non-favorable [35, 36] results of bicarbonate in reducing the incidence of CIN. In conclusion, hydration with bicarbonate seems comparable to normal saline and the evidence does not support use of one over the other.

Hydration regimes including infusion both before and after contrast administration have been applied. With bicarbonate hydration the infusion regimes are usually shorter (<6 h), compared to saline regimes (12–24 h). There is conflicting reports on whether the shorter bicarbonate regimes or the longer saline regimes are most effective. Although some studies are in favor of bicarbonate, and others of saline, most studies do not favor one regime over the other [37, 38]. It has also been shown that pre-hydration with bicarbonate before contrast administration is effective in preventing CIN [32].

As already mentioned most studies on hydration have been performed in cardiologic patients undergoing invasive procedures. Few studies have evaluated hydration in patients undergoing contrast-enhanced CT. These studies did not favor hydration with bicarbonate over normal saline [32, 39•].

Based on current evidence, practical, and economic considerations it seems reasonable to recommend a short bicarbonate pre-hydration scheme in patients at risk of CIN undergoing contrast-enhanced imaging with intravenous or indirect (extremities, carotids, mesenteric arteries, or aorta below the renal arteries) arterial injection. Conventional injection protocols with bicarbonate or saline infusion both before and after contrast administration should be reserved for patients at risk of CIN undergoing imaging with direct intra-arterial injections (in the renal arteries or supra-renal aorta).

Recommendation from the ESUR on hydration regimes:

  • For intravenous or indirect intra-arterial contrast media administration: pre-hydrate the patient with 3 ml/kg/h bicarbonate 1.4% for 1 h.

  • For direct intra-arterial contrast media administration: pre-hydrate the patient with 3 ml/kg/h bicarbonate 1.4% for 1 h followed by post-hydration with 1 ml/kg/h bicarbonate 1.4% for 4–6 h.

In patients with severe congestive heart failure (NYHA stage 3 and 4) or severe renal impairment (CKD stage V) intravenous hydration with large volumes can exacerbate acute heart failure and cause pulmonary edema. Hydration in these patients should therefore be individualized, and the strategy should be determined along with cardiologists and nephrologists.

N-acetylcysteine (NAc)

In a study nearly two decades ago, administration of the antioxidant NAc along with hydration was reported to reduce the incidence of CIN [40]. However, the majority of recent randomized clinical trials or meta-analyses have failed to demonstrate a protective effect of NAc against CIN. This applies both to patients undergoing angiography (coronary or peripheral) and CT [38, 41,42,43]. A recent meta-analysis showed the benefit of NAc with or without administration of statins when added to hydration in CIN prevention [44].

Statins

The use of statins to reduce hypercholesterolemia is well established. These drugs are reversible inhibitors of the microsomal enzyme HMG-CoA reductase which converts HMG-CoA to mevalonate. This is a crucial step in cholesterol synthesis. Other effects of statins, including modulation of endothelial function, inflammation, coagulation, and plaque vulnerability have been described [45]. Furthermore, short-term administration of statins at high dose has been proposed as a specific prophylactic measure against CIN. Recent meta-analyses have showed lower-CIN incidence in patients receiving statins compared to controls [46,47,48,49].

Although the results have been positive, generalizability of these results seems problematic. First of all, the study populations have only included cardiologic patients scheduled for percutaneous coronary intervention (PCI). The administration regime for statins is different and so are the combination with other preventive strategies such as NAc administration and/or hydration with saline or bicarbonate. Yet another limitation is that most patients scheduled for PCI are already on chronic statin therapy, and the protective effect of statins have only been demonstrated in statin-naïve patients. With the current evidence it does not seem advisable to apply short-term high dose statins as a single preventive strategy.

Renin-Angiotensin System Blockage: ACE Inhibitors and Angiotensin-II Receptor Blockers

Blockage of the renin-angiotensin system with either ACE inhibitors or angiotensin-II receptor blockers leads to decreased blood pressure levels. From a theoretical point of view high blood pressure may be important to maintain glomerular filtration. Accordingly withholding these medications have been suggested as a preventive strategy for CIN. Multiple studies have investigated this effect. However, data from meta-analyses are conflicting [50,51,52]. Thus it does not seem reasonable to recommend withholding ACE inhibitors/angiotensin-II blockers to prevent CIN.

Renal Replacement Therapy

Hemodialysis safely removes iodinated contrast media. A number of factors influence the effectiveness of hemodialysis (flow of blood and dialysate, duration of dialysis, membrane permeability, contrast media characteristics including protein binding, electric charge, molecular size and hydrophilicity) [53]. Studies have shown that multiple hemodialysis sessions are needed to remove 95% of administered iodine-based contrast media. The reason for this is a modest fractional removal of iodine-based contrast medium [54].

Patients with chronic kidney disease undergoing hemodialysis may have some preserved urine production. In this subgroup of patients there has been interest in whether or not, hemodialysis after contrast media administration is an effective means of reducing the risk of CIN. However, most studies do not demonstrate a protective effect of hemodialysis against CIN in patients with CKD [55, 56]. The majority of studies have been performed in cardiologic patients undergoing coronary angiography. A single study from 2014 demonstrates a protective effect against CIN in coronary angiography patients with simultaneous hemodialysis [57].

As the benefit of hemodialysis for prevention of CIN has not been demonstrated conclusively in patients undergoing imaging procedures with intravenous or intra-arterial administration of CM, its use is not recommended. The complexity of the hemodialysis procedure also discourages the routine application of this method as a preventive method.

In patients where volume overload should be avoided it may be relevant to perform hemodialysis after contrast administration. This applies to patients with severe congestive heart failure and severe kidney disease.

Metformin and Iodinated Contrast Media

The biguanide agent metformin is used in first-therapy for type 2 diabetes mellitus. Metformin lowers plasma glucose levels. Pharmacodynamically metformin works by inhibiting the production of hepatic glucose, reducing glucose absorption in the bowel, and by improving cellular-glucose uptake and utilization [58]. Although the main indication is type 2 diabetes mellitus, metformin is also used in treatment of polycystic ovary syndrome. A potential adverse effect of metformin treatment is induction of metabolic lactate acidosis [59]. Metformin is eliminated by renal excretion. Consequently, any condition that impairs renal function increases the risk of metformin induced lactate acidosis. As intravascular administration of iodinated CM may cause nephropathy, the following guidelines have been developed by the American College of Radiology and the Federal Drug Administration:

Patients scheduled for intravenous (or indirect intra-arterial) CM administration

  • eGFR > 45 ml/min/1.73 m2 (CKD stage 1-3A)—Continue metformin normally prior to and after imaging.

  • eGFR 30–44 ml/min/1.73 m2 (CKD stage 3B)—Continue metformin normally prior to and after imaging. Alert prescribing physician to frequently check eGFR and evaluate metformin dosage.

  • eGFR < 30 ml/min/1.73 m2 (CKD stage 4–5)—Stop metformin temporarily prior to imaging. Alert prescribing physician to frequently check eGFR 48 h after CM administration. Evaluate anti-diabetic medication with the help of a nephrology specialist.

Patients with acute kidney injury of any type scheduled for any type of CM administration OR patients with CKD scheduled for direct intra-arterial (supra-renal aorta or renal artery injection) CM administration

  • Stop metformin temporarily prior to imaging. Alert prescribing physician to frequently check eGFR 48 h after CM administration. Evaluate anti-diabetic medication with the help of a nephrology specialist.

Conclusions

The reduction in renal function that may occur after intravascular injection of iodine-based CM is known as CIN. This condition has been extensively investigated in the literature, and recent important findings from these studies have been described in the present article.

Recent studies have showed that the true incidence of CIN is lower than previously assumed. This is especially true after intravenous injection. It is more difficult to draw conclusions from studies with intra-arterial injections of CM, as the vast majority of these studies are performed in cardiology patients with multiple confounding comorbidities that influence renal function. An eGFR cut-off of 30 ml/min/1.73 m2 is now recommended for intravenous and indirect intra-arterial injections. For direct intra-arterial injection eGFR 45 ml/min/1.73 m2 is the cut-off level.

Hydration with either saline or bicarbonate is the best strategy to reduce the risk of CIN. A variety of hydration schemes have been investigated. There is no evidence that hydration with bicarbonate is more beneficial than hydration with normal saline. However, the bicarbonate injection schemes are most often of shorter duration, and could therefore be preferred from a practical point of view.

Treatment with the antioxidant N-acetylcysteine (NAc) has been suggested as a protective drug against CIN. However, the majority of recent randomized clinical trials or meta-analyses have failed to demonstrate a protective effect of NAc against CIN. Use of statins as an isolated preventive strategy is also not recommended.

Hemodialysis effectively removes contrast media, but it has not a protective effect against CIN in patients with CKD.

Patients treated with metformine should be evaluated before CM administration, as a metabolic acidosis can be the result if renal function is decreased by the CIN mechanism.

A key point for further studies on CIN should be to include control groups (not injected with CM) whenever possible.