Contrast-Induced Nephropathy

Contrast media (CM) act on distinct anatomic sites within the kidney and exert adverse effects via multiple mechanisms. They cause a direct cytotoxic effect on the renal proximal tubular cells, enhance cellular damage by reactive oxygen species, and increase resistance to renal blood flow. They also exacerbate renal vasoconstriction, particularly in the deeper portions of the outer medulla. This is especially important in patients with chronic kidney disease (CKD), because their preexisting abnormal vascular pathobiology is made worse by the effects of CM.[13, 14]

Renal (particularly medullary) microcirculation depends on a complex interplay of neural, hormonal, paracrine and autocrine influences. Of note are the vasodilator nitric oxide (NO) and the vasoconstrictors vasopressin, adenosine (when it acts via the high-affinity A1 receptors), angiotensin II, and endothelins. Prostaglandins cause a redistribution of blood flow to the juxtamedullary cortex and therefore are protective.

NO, in particular, seems to be very important, with antiplatelet, vasodilatory, insulin sensitizing, anti-inflammatory, and antioxidant properties. It has been suggested that plasma levels of asymmetrical dimethylarginine (ADMA), which is an endogenous inhibitor of all NO synthase isoforms, can be used as a marker of CIN, especially in patients with unfavorable outcomes.

CM-mediated vasoconstriction is the result of a direct action of CM on vascular smooth muscle and from metabolites such as adenosine and endothelin. Additionally, the osmotic property of CM, especially in the tubular lumen, decreases water reabsorption, leading to a buildup of interstitial pressure. This, along with the increased salt and water load to the distal tubules, reduces GFR and causes local compression of the vasa recta. All of this contributes to worsening medullary hypoxemia and renal vasoconstriction in patients who are already volume depleted.

Finally, CM also increase resistance to blood flow by increasing blood viscosity and by decreasing red cell deformability. The resulting intravascular sludging generates local ischemia and causes activation of reactive oxygen species that cause tubular damage at a cellular level.

Comparison of contrast-agent nephropathy potential

The ability of different classes of CM to cause CIN is influenced by their osmolality, ionicity (the ability of the CM to dissociate in water), and molecular structure. Each of those characteristics, in turn, influences their behavior in body fluid and their potential to cause adverse effects.[15] See Table 1, below.

Table 1. Physiochemical Properties of Contrast Media ^[16] (Open Table in a new window)

Agents are classified as high, low, or iso-osmolar depending on their osmolality in relation to blood. The term low-osmolarity contrast media (LOCM) is actually somewhat of a misnomer, since these agents have osmolalities of 600-900 mOsm/kg and so are 2-3 times more hyperosmolar than blood. High-osmolarity contrast media (HOCM) are 5-7 times more hyperosmolar than blood, with osmolalities greater than 1500 mOsm/kg.

Molecular structure of CM refers to the number of benzene rings. Most CM that were developed in the 1990s are dimers with 2 benzene rings. Dimeric CM, while nonionic and with low osmolarity, have high viscosity, which may influence renal tubular blood flow.

The ratio of iodine to dissolved particles describes an important relationship between opacification and osmotoxicity of the contrast agent. The higher ratios are more desirable. High-osmolar agents have a ratio of 1.5, low-osmolar agents have a ratio of 3, and iso-osmolar agents have the highest ratio, 6.

While the safety of LOCM over HOCM in terms of CIN seems intuitive, clinical evidence of it came from a meta-analysis by Barrett and Carlisle that showed the benefit of using LOCM over HOCM, mostly in high-risk patients.[17] The Iohexol Cooperative Study was a large, prospective, randomized, double-blinded, multicenter trial that compared the risk of developing CIN in patients receiving the low-osmolarity agent iohexol versus the high-osmolarity agent diatrizoate. While the HOCM group was 3.3 times more likely to develop CIN compared with the LOCM group, this was seen only in patients with preexisting CKD (baseline SCr of 1.5 mg/dL or higher). In addition to CKD, other independent risk factors were diabetes mellitus, male sex, and higher contrast volume.

Even within the LOCM category, the risk is not the same for all agents. High-risk patients have a higher likelihood of developing CIN if they receive iohexol than if they receive another agent (eg, iopamidol) in the same class.

A comparison of two iso-osmolar LOCM (iohexol and iodixanol) in the Nephrotoxicity in High-Risk Patients Study of Iso-Osmolar and Low-Osmolar Non-Ionic Contrast Media (NEPHRIC study), arguably the most definitive study in this category to date, found that the odds of developing CIN in high-risk patients were almost 9 times greater for the study's iohexol group than for the iodixanol group. The incidence of CIN was 3% in the iodixanol group versus 26% in the iohexol group.[18] However, these results were not duplicated in some subsequent studies.

When iodixanol was used, the Rapid Protocol for the Prevention of Contrast-Induced Renal Dysfunction (RAPPID) trial found a 21% incidence of CIN,[19] and the Contrast Media and Nephrotoxicity Following Coronary Revascularization by Angioplasty (CONTRAST) trial found a 33% incidence of CIN.[20] Finally, the Renal Toxicity Evaluation and Comparison Between Visipaque (Iodixanol) and Hexabrix (Ioxaglate) in Patients With Renal Insufficiency Undergoing Coronary Angiography (RECOVER) trial compared the iso-osmolar agent iodixanol with the low-osmolarity agent ioxaglate and found a significantly lower incidence of CIN with iodixanol than with ioxaglate (7.9% vs 17%, respectively).[21]

Thus, although the data are by no means uniform, they seem to suggest that the iso-osmolar contrast agent iodixanol may be associated with smaller increases in SCr and lower rates of CIN when compared with other LOCM, especially in patients with CKD and in those with CKD and diabetes mellitus.[22]  

Risk factors

Risk factors for CIN can be divided into patient-related, procedure-related, and contrast-related factors (although the risk factors for CIN are still being identified and remain poorly understood).

Patient-related risk factors are as follows[23] :

• Older age
• CKD
• Diabetes mellitus
• Hypertension
• Metabolic syndrome
• Anemia
• Multiple myeloma
• Hypoalbuminemia
• Kidney transplant
• Hypovolemia and decreased effective circulating volumes - As evidenced by chronic heart failure (CHF), an ejection fraction (EF) of less than 40%, hypotension, and intra-aortic balloon counterpulsation (IABP) use

Procedure-related risk factors are as follows:

• Urgent versus elective
• Arterial versus venous
• Diagnostic versus therapeutic

Contrast-related risk factors are as follows:

• Volume of contrast
• Contrast characteristics, including osmolarity, ionicity, molecular structure, and viscosity

The single most important patient-related risk factor is preexisting CKD, even more so than diabetes mellitus.[24] Patients with CKD in the setting of diabetes mellitus have a 4-fold increase in the risk of CIN compared with patients without diabetes mellitus or preexisting CKD.

In contrast, a review of patients with acute kidney injury (AKI) on arrival at the emergency department found that CM administration was not associated with either persistent AKI at hospital discharge or initiation of dialysis within 180 days. These authors concluded that current consensus recommendations for use of intravenous CM in patients with stable kidney disease may also be followed in patients with pre-existing AKI.[4]

Occurrence in the United States

CIN is the third leading cause of hospital-acquired AKI. Decreased renal perfusion causing either prenal injury or acute tubular necrosis is the commonest cause.[25]

Despite numerous studies, the actual incidence of CIN is not clear. The rate varies, depending on the following:

• The definition used for CIN
• The contrast agent characteristics, including the type, amount, duration, and route of administration
• Preexisting risk factors
• Length of follow-up (including the day of measurement of postcontrast serum creatinine)

In patients without risk factors, the incidence may be as low as 2%. In those with risk factors, such as diabetes, the rate rises to 9%, and to as high as 90% in patients with diabetic nephropathy. Therefore, the number and the type of preexisting risk factors directly influence the incidence of kidney insufficiency. Incidence rates are also procedure dependent, with reports in the literature ranging from 1.6-2.3% for diagnostic interventions to 14.5% overall in patients undergoing coronary intervention.[26]

Although the risk for CIN with CT scans has been a long-standing concern, a retrospective study by Hinson et al of 17,934 emergency department visits from 2009 to 2014 found that IV contrast was not associated with an increased frequency of AKI. Whether patients underwent contrast-enhanced, unenhanced, or no CT, there were no significant differences in the incidence of AKI, dialysis, or mortality, regardless of baseline kidney function.[9]

The main strengths of this study were the large number of patients and the use of two controls. However, this was a single-center study and it excluded patients with serum creatinine levels > 4.0 mg/dL as well as those with kidney transplants.

In another study, of 13,126 patients who underwent percutaneous vascular interventions, the incidence of CIN was noted to be 3%. Of the 400 patients who developed CIN, 26 (6.5%) required dialysis.[27]

 A meta-analysis of 259 studies found a 9% incidence of CIN following angiography with a 0.5% incidence of kidney failure requiring hemodialysis.[28]  

Race- and age-related demographics

Overall, race has not been found to be a risk factor for CIN. A review of 4070 predominantly male patients undergoing peripheral and coronary angiography and percutaneous coronary and endovascular intervention found no association between race and the development of CIN at 72 hours, or the development of kidney dysfunction at 3 months post procedure. In the long term, the rate of initiation of dialysis was significantly lower in White patients than in non-Whites, but mortality was not.[29]

The incidence of CIN in patients older than 60 years has been variously reported as 8-16%. It has also been shown that in patients with acute myocardial infarction (MI) who undergo coronary intervention, age of 75 years or older is an independent risk factor for CIN.

CIN is normally a transient process, with kidney function reverting to normal within 7-14 days of contrast administration. Less than one-third of patients develop some degree of residual renal impairment.

Dialysis is required in less than 1% of patients, with a slightly higher incidence in patients with underlying renal impairment (3.1%) and in those undergoing primary percutaneous coronary intervention (PCI) for MI (3%). However, in patients with diabetes and severe kidney failure, the rate of dialysis can be as high as 12%.

Of the CIN patients who need dialysis, 18% end up requiring it permanently. However, many of these patients have advanced kidney insufficiency and concomitant diabetic nephropathy and would have progressed to needing dialysis regardless of the episode of CIN.

A growing body of knowledge indicates that AKI after contrast medium exposure can be a harbinger of CKD or ESRD. In one observational study that included 3986 patients who underwent coronary angiography, 12.1% of patients experienced contrast-induced AKI, and of those, 18.6% suffered persistent kidney damage.[30] The population studied appeared representative of the general population undergoing angiography and the rate of AKI was consonant with other studies. The finding that persistent kidney damage can occur after contrast-induced AKI highlights the potential for acceleration of the progression of kidney injury in individuals with preexisting CKD.

Mortality

Patients who require dialysis have a considerably worse prognosis, with a reported rate of 35.7% inhospital mortality (compared with 7.1% in the nondialysis group) and a 2-year survival rate of only 19%.

CIN by itself may be an independent mortality risk factor. Following invasive cardiology procedures, patients with normal baseline kidney function who develop CIN have reduced survival compared with patients with baseline chronic CKD who do not develop CIN.

In one study of the effect of CIN on long-term mortality after percutaneous coronary intervention in patients with or without CKD, CIN was found to be significantly correlated with long-term mortality in the entire cohort (hazard ratio [HR] 2.26, 95% confidence interval [CI] 1.62 to 2.29, P< 0.0001) and in patients with CKD (HR 2.62, 95% CI 1.91 to 3.57, P< 0.0001) but not in patients without CKD (HR 1.23, 95% CI 0.47 to 2.62, P = 0.6). The rate of CIN was 11% in patients with CKD and 2% in patients without CKD.[31]

Gadolinium-based agents

Gadolinium-based CM (used for magnetic resonance imaging [MRI]), when compared with iodine-based CM, have a similar, if not worse, adverse effect profile in patients with moderate CKD and eGFR of less than 30 mL/min. Their use has been implicated in the development of nephrogenic systemic fibrosis, a chronic debilitating condition with no cure.

A review of 3 series and 4 case reports suggested that the risk of kidney insufficiency with gadolinium is similar to that of iodinated radiocontrast dye. The reported incidence varies from 4% in stage 3 CKD to 20% in stage 4 CKD, but may even be worse, as suggested by some investigators. In a study of 57 patients, AKI was seen in 28% of patients in the gadolinium group, compared with 6.5% of patients in the iodinated CM group, despite prophylactic saline and N-acetylcysteine (NAC).[32]

The risk factor profile with gadolinium-based CM is similar to that for iodinated CM; the incidence of AKI is higher in patients with any of the following:

• Older age
• Lower baseline creatinine clearance
• Diabetic nephropathy
• Anemia
• Hypoalbuminemia

Risk stratification scoring systems

CIN is the result of a complex interplay of many of the above risk factors. The presence of 2 or more risk factors is additive, and the likelihood of CIN rises sharply as the number of risk factors increases. Researchers have tried to quantify the contribution of each risk factor to the ultimate outcome of CIN.

Risk stratification scoring systems have been devised to calculate an individual patient’s risk of developing CIN. This has mostly been done in patients undergoing percutaneous coronary intervention (PCI), especially those with preexisting risk factors. Mehran et al developed a scoring system based on points awarded to each of the following multivariate predictors[33] :

• Hypotension = 5 points
• Intra-aortic balloon pump (IABP) use = 5 points
• CHF = 5 points
• SCr > 1.5 mg/dL = 4 points
• Age > 75 years = 4 points
• Anemia = 3 points
• Diabetes mellitus = 3 points
• Contrast volume = 1 point for each 100 mL used

Risk categories by total calculated score, CIN rates, and requirements for dialysis were as follows:

• Low risk (score of ≤5): CIN rate 7.5%, dialysis in 0.04%
• Moderate risk (score of 6-10): CIN rate 14%, dialysis in 0.12%
• High risk (score of 11-15): CIN rate 26.1%, dialysis in 1.09%,
• Very high risk (score of ≥16): CIN rate 57.3%, dialysis in 12.6%

Bartholomew et al derived and validated a scoring system for CIN risk after PCI that took into consideration the following 8 variables[34] :

• Creatinine clearance < 60 mL/min
• IABP use
• Urgent coronary procedure
• Diabetes mellitus
• CHF
• Hypertension
• Peripheral vascular disease
• Volume of contrast used

Lin et al validated a simpler scoring system for predicting the risk of CIN in patients undergoing emergent PCI.[35] The system consists of the following risk factors, each of which is assigned 1 point:

• Age > 75 years
• Baseline SCr > 1.5 mg/dL
• Hypotension
• IABP use

Risk categories by scores and CIN incidence were as follows:

• Low risk: 0 points; incidence 1.0%
• Moderate risk: 1-2 points; incidence 13.4%
• High risk: ≥3 points; incidence 90.0%

Prognostic Nutritional Index

A number of studies have evaluated the effectiveness of the Prognostic Nutritional Index (PNI) to predict CIN in patients with acute coronary syndrome (ACS).[36, 37]  The PNI is based on the serum albumin concentration and peripheral blood lymphocyte count, and has been used primarily with cancer patients as a reliable indicator of nutritional and immune status. A meta-analysis of 17,590 patients from 10 studies was conducted to determine the reliability of PNI in predicting CIN in patients with ACS receiving angiography or PCI. The results of the analysis found a higher risk of CIN in patients with a low PNI compared to those with a high PNI.[37]  

Patients with contrast-induced nephropathy usually present 24-48 hours after receiving intravenous contrast during a diagnostic or therapeutic procedure (eg, percutaneous coronary intervention [PCI]). The acute kidney injury is usually nonoliguric.

A physical examination is useful for ruling out other causes of acute nephropathy, such as cholesterol emboli (characteristic findings of which include blue toe and livedo reticularis) or drug-induced interstitial nephritis (which typically involves a rash). Patients may have evidence of volume depletion or may be in decompensated heart failure. See DDx/Diagnostic Considerations.

In contrast-induced nephropathy (CIN), the serum creatinine (SCr) concentration usually begins to increase within 24 hours after contrast agent administration, peaks between days 3 and 5, and returns to baseline in 7-10 days. Serum cystatin C (which has been suggested as a surrogate marker of kidney function in lieu of SCr) is increased in patients with CIN.

Nonspecific formed elements can appear in the urine, including renal tubular epithelial cells, pigmented granular casts, urate crystals, and debris. However, these urinary findings do not correlate with severity.

Urine osmolality tends to be less than 350 mOsm/kg. The fractional excretion of sodium (FENa) may vary widely. In the minority of patients with oliguric CIN, the FENa is low in the early stages, despite the absence of clinical evidence of volume depletion.

Histology

Contrast media cause direct toxic effects on renal tubular epithelial cells, characterized by cell vacuolization, interstitial inflammation, and cellular necrosis. In one study, these characteristic changes, called osmotic nephrosis, were observed in 22.3% of patients undergoing kidney biopsy within 10 days after contrast exposure.[38]

Prevention is the cornerstone of contrast-induced nephropathy (CIN) management, and hydration is the cornerstone of CIN prevention. Renal perfusion is decreased for up to 20 hours following contrast administration. Intravascular volume expansion maintains renal blood flow, preserves nitric oxide production, prevents medullary hypoxemia, and enhances contrast elimination. 

A number of other therapies for CIN have been investigated, including the following:

• Sodium bicarbonate
• N-acetylcysteine (NAC)
• Statins
• Ascorbic acid ^[39]
• The adenosine antagonists theophylline and aminophylline
• Vasodilators
• Forced diuresis
• Renal replacement therapy
• Prostaglandin E1 ^[40]

A systematic review and meta-analysis of prevention strategies for CIN found that, compared with intravenous (IV) saline only, the following had clinically important and statistically significant benefits when used in combination with IV saline[41] :

• Low-dose NAC: Risk ratio (RR), 0.75
• NAC, in patients receiving low-osmolar contrast media (LOCM): RR, 0.69
• Statins plus NAC (versus NAC plus IV saline): RR, 0.52

A clinically important difference that was not statistically significant was found with the following, when compared with IV saline only[41] :

• Sodium bicarbonate, in patients receiving LOCM
• Statins plus IV saline
• Ascorbic acid

A systematic review and meta-analysis by Zaki and colleagues found that oral hydration produced similar a similar reduction in CIN compared with IV fluids. The researchers concluded that patients should be encouraged to drink fluids and salts prior to undergoing procedures requiring contrast medium.[42]

The first study revealing the benefit of hydration in CIN prevention came from Solomon et al.[43] They also found forced diuresis to be inferior to hydration with 0.45% saline. Fluids with different compositions and tonicity have since been studied, as well as the addition of bicarbonate and mannitol.

Intravenous normal saline has been found to be superior to half-normal saline in terms of its enhanced ability to produce intravascular volume expansion. It also increases delivery of sodium to the distal nephron, prevents renin-angiotensin activation, and thus maintains increased renal blood flow. Oral fluids, while beneficial, have not been considered as effective as intravenous hydration.[44, 45]  

The first study comparing no prophylaxis versus hydration, the phase 3 AMACING trial, found no prophylaxis to be non-inferior and cost-saving in preventing CIN compared with intravenous hydration. The AMACING trial included 600 high-risk patients with an estimated glomerular filtration rate (eGFR) of 30-59 mL/min/1.73 m2) aged 18 years and older who were undergoing an elective procedure requiring iodinated contrast medium administration.[46]

Diuretics on their own are not recommended but a meta-analysis suggested that furosemide with matched hydration by the RenalGuard System may reduce the incidence of contrast-induced acute kidney injury in high-risk patients undergoing percutaneous coronary intervention or transcatheter aortic valve replacement.[47]

The CIN Consensus Working Panel found that adequate intravenous volume expansion with isotonic crystalloids (1-1.5 mL/kg/h), 3-12 hours before the procedure and continued for 6-24 hours afterward, decreases the incidence of CIN in patients at risk. The panel studied 6 clinical trials with different protocols for volume expansion. The studies differed in the type of fluid used for hydration (isotonic vs half-normal saline), route, duration, timing, and amount of fluid used.[48]

For hospitalized patients, volume expansion should begin 6 hours prior to the procedure and be continued for 6-24 hours postprocedure. For outpatients, administration of fluids can be initiated 3 hours before and continued for 12 hours after the procedure. Postprocedure volume expansion is more important than preprocedure hydration. It has been suggested that a urine output of 150 mL/h should guide the rate of intravenous fluid replacement, although the CIN Consensus Working Panel did not find it useful to recommend a target urine output.

Chronic heart failure (CHF) poses a particular challenge. Patients with compensated CHF should still be given volume, albeit at lower rates. Patients with uncompensated CHF should undergo hemodynamic monitoring, if possible, and diuretics should be continued. In emergency situations, clinical judgment should be used and, in the absence of any baseline kidney function measurements, adequate postprocedure hydration should be carried out.

In patients with CHF and chronic kidney disease who were undergoing coronary procedures, Qian et al found that CIN occurred less often in patients who received hydration plus nitrates than in those who received routine hydration (12.8% vs. 21.2%; P=0.018). The treatment group received continuous intravenous infusion of isosorbide dinitrate combined with intravenous infusion of isotonic saline at a rate of 1.5 mL/kg/h.[49]

A study by Yan et al documented the benefit of setting the hydration infusion rate according to the diameter of the inferior vena cava, as measured by ultrasonography (IVCU), to prevent CIN in patients with CHF. Compared with a control group that received isotonic saline at 0.5 mL/kg/h, patients receiving IVCU-guided hydration received significantly higher hydration volume and experienced a significantly lower incidence of CIN (12.5% vs 29.1%, P = 0.004) and a significantly lower rate of major adverse cardiovascular or cerebrovascular events during the 18-month follow-up period (14.4% vs 27.2%, P = 0.027). The study included 207 CHF patients undergoing coronary angiography or coronary angiography with percutaneous coronary intervention.[50]

Statins are widely used in coronary artery disease (CAD) for their pleiotropic effects (favorable effects on endothelin and thrombus formation, plaque stabilization, and anti-inflammatory properties), and it was believed that, given the vascular nature of CIN, they might have similar renoprotective effects. Iniital data for statin use were retrospective and anecdotal, and were taken mostly from patients already on statins who underwent percutaneous coronary intervention (PCI).[51] Subsequently, a meta-analysis of prospective controlled studies found a statistically significant reduction of CIN incidence in patients pretreated with high-dose statins before the procedure (odds ratio [OR], 0.45; 95% confidence interval, 0.34-0.58; P < 0.0001).[52]

Another meta-analysis showed that in patients undergoing coronary angiography or PCI, short-term statin use reduced the incidence of CIN; these authors concluded that statins should be used even in patients with low levels of low-density lipoprotein (LDL) cholesterol. Of a total of 4734 patients, CIN occurred in 79 of 2,358 patients (3.3%) who were treated with statins, versus 153 of 2,376 patients (6.4%) who were given placebo (OR 0.50, P< 0.00001). Benefits were observed with both high-dose short-term statins (OR 0.44, P< 0.0001) and low-dose short-term statins (OR 0.58, P = 0.010).[53]

Yet another meta-analysis demonstrated that preprocedural rosuvastatin treatment could significantly reduce the incidence of contrast-induced acute kidney injury (OR 0.49, P < 0.001). However, rosuvstatin treatment did not seem to be effective for preventing contrast-induced acute kidney injury in patients with chronic kidney disease who underwent elective cardiac catheterization.[54]

Bicarbonate therapy alkalinizes the renal tubular fluid and thus prevents free radical injury. In the Harber-Weiss reaction, which is activated in an acidic environment, hydrogen peroxide and an oxygen ion (from superoxide) react to form a hydroxide ion, all agents of free-radical injury. Bicarbonate, by alkalinizing the environment, slows down that reaction. It also scavenges reactive oxygen species (ROS) from nitric oxide, such as peroxynitrite.

Bicarbonate protocols most often include infusion of sodium bicarbonate at the rate of 3 mL/kg/hour 1 hour before the procedure, continued at 1 mL/kg/hour for 6 hours afterward. Some investigators have used 1 mL/kg/hour for 24 hours, starting 12 hours before the procedure. The exact duration, however, remains a matter of debate. Hydration with sodium bicarbonate has been found by some researchers to be more protective than normal saline alone.

Treatment controversy

A 2008 retrospective cohort study at the Mayo Clinic assessed the risk of CIN associated with the use of sodium bicarbonate, NAC, and the combination of sodium bicarbonate with NAC and found that, compared with no treatment, sodium bicarbonate used alone was associated with an increased risk of CIN. NAC alone or in combination with sodium bicarbonate did not significantly affect the incidence of CIN. The results were obtained after adjusting for confounding factors, including total volume of hydration, medications, baseline creatinine, and contrast iodine load.[55] Given the above information, further evaluation of the use of sodium bicarbonate to prevent CIN was recommended.

Subsequently, a prospective, double-blind, multicenter randomized clinical trial in 391 patients with an estimated glomerular filtration rate (eGFR) < 45 mL/min/1.73 m2 undergoing elective coronary or peripheral angiography found no statistically significant difference between sodium bicarbonate and sodium chloride in terms of the incidence of the composite of death, dialysis, or sustained 6-month reduction in eGFR or contrast-induced acute kidney injury. This trial used a high dose of isotonic sodium bicarbonate (target 2.0 mEq/kg) or a similar molar amount of isotonic sodium chloride.[56]

Furthermore, a meta-analysis and systematic review of 29 studies concluded that overall, hydration with sodium bicarbonate could significantly reduce CIN and the length of hospital stay compared with sodium chloride and that the addition of NAC as a supplement to sodium bicarbonate could increase prophylactic effects against nephropathy. In addition, hydration with sodium bicarbonate was found to be more effective in emergency coronary imaging and high-risk patients than in elective coronary imaging.[57]

In the Kompas randomized clinical trial, which included 523 adults with stage 3 chronic kidney disease, hydration with sodium bicarbonate before contrast-enhanced computed tomography (CT) imaging showed no benefit in terms of renal safety compared with withholding hydration. The relative increase in creatinine level 2 to 5 days after contrast administration compared with baseline was 3.5% in study patients who received prehydration with 250 mL of 1.4% sodium bicarbonate administered in a 1-hour infusion before CT imaging, versus 3.0% in those who received no prehydration (mean difference, 0.5; 95% confidence index, −1.3 to 2.3; P <  0.001 for noninferiority).[58]

NAC is acetylated L-cysteine, an amino acid. Its sulfhydryl groups make it an excellent antioxidant and scavenger of free oxygen radicals. It also enhances the vasodilatory properties of nitric oxide.

In a systematic review and meta-analysis of 61 randomized controlled trials that included 11,480 patients, Xu et al determined that NAC supplementation was associated with a significant decrease in CIN risk and blood creatinine level, but not with a reduction in mortality or nephropathy requiring dialysis. In addition, NAC supplementation did not reduce the risk of CIN in patients with diabetes.[59]

Xu et al noted that NAC prophylaxis provides much more important benefit in patients with kidney dysfunction and high contrast medium dose than in those with normal kidney function and low dose of contrast agent. Their findings also suggested that administering NAC orally provides increased protection against CIN. The authors concluded that, “it is reasonable to administer NAC by the oral route for patients who are undergoing coronary angiography and who have renal dysfunction or who are receiving high doses of contrast agent.”[59]

The standard oral NAC regimen consists of 600 mg twice daily for 24 hours before and on the day of the procedure. Higher doses of 1 g, 1200 mg, and 1500 mg twice daily have also been studied, with no significant dose-related or route-related (oral vs intravenous) difference. NAC has very low oral bioavailability; substantial interpatient variability, and inconsistency between the available oral products, which obscures the picture further.[10, 45, 60]

Treatment controversy

A controversy relating to NAC therapy involves the parameter on which its effectiveness is assessed. It was suggested that the beneficial effect of NAC in CIN is related to its ability to lower serum creatinine (SCr) rather than to improve GFR. It was believed that NAC directly reduces SCr by increasing creatinine excretion (tubular secretion), decreasing its production (augments activity of creatine kinase), or interfering with its laboratory measurement, enzymatic or nonenzymatic (Jaffe method).

This was supported by a study that demonstrated a significant decrease in SCr after four doses of 600 mg of oral NAC in healthy volunteers with normal kidney function and no exposure to radiocontrast media.[61] This would cast doubt on the results of at least 13 randomized, controlled trials that showed NAC to be protective in CIN, with SCr used as the endpoint. However, Haase et al compared the effect of NAC on SCr by simultaneously studying its effect on cystatin C and found that NAC did not artifactually lower SCr when measured by the Jaffe method.[62]

The CIN Working Panel concluded that the existing data on NAC therapy in CIN is sufficiently varied to preclude a definite recommendation.[48]  The KDIGO guidelines do recommend use of NAC in conjunction with hydration. In clinical practice, NAC therapy remains part of the standard of care; NAC is routinely administered because of its low cost, lack of adverse effects, and potential beneficial effect, as demonstrated by the relative risk reduction of CIN, ranging from 0.37-0.73, as reported in several meta-analyses.

Fewer than 1% of patients with CIN ultimately go on to require dialysis, the number being slightly higher in patients with underlying renal impairment (3.1%) and in those undergoing primary percutaneous coronary intervention (PCI) for myocardial infarction (3%). However, in patients with diabetes and severe kidney failure, the rate of dialysis can be as high as 12%. Patients who get dialyzed do considerably worse, with in-hospital mortality rates of 35.7% (compared with 7.1% in the nondialysis group) and a 2-year survival rate of only 19%.

Contrast media (CM) have low lipophilicity, low plasma protein binding, and minimal biotransformation. They quickly equilibrate across capillary membranes and have volumes of distribution equivalent to that of the extracellular fluid volume. In patients with normal kidney function, CM are excreted with the first glomerular passage and the decrease in their plasma concentration follows a two-part exponential function: a distribution phase and an elimination phase. In patients with renal impairment, the renal clearance values are reduced. For example, 50% of the low-osmolarity contrast agent iomeprol is eliminated within 2 hours in healthy subjects, compared with 16-84 hours in patients with severe renal impairment.

In patients already on dialysis, the commonly cited issues with contrast administration include volume load and direct toxicity of contrast to the remaining nonfunctional nephrons and nonrenal tissues. These issues underlie the perceived need for emergent dialysis and contrast removal.

Rodby attempted to address these concerns, calculating that the administration of 100 mL of hyperosmolar contrast would move 265 mL of water from the intracellular to the extracellular compartment, resulting in an increase in extracellular volume by 365 mL. The increase in intravascular space would therefore be only a third, or 120 mL. Fluid shifts with low-osmolarity CM are even less. Rodby also found that extrarenal toxicity of CM was cited in mostly single case reports, and no objective evidence could be identified in three prospective studies.[63]

The risk of acute damage from contrast is greatest in patients with chronic kidney disease (CKD). This can be explained by the increase in single-nephron GFR and, thus, the filtered load of contrast per nephron. This is akin to a double hit to the remaining nephrons: increased contrast load and prolonged tubular exposure. While this may not seem to be a concern in patients with end-stage renal disease (ESRD) who are already on dialysis, residual kidney function, in fact, plays a major role in their outcome—more so in patients on peritoneal dialysis. Its preservation is therefore important.[63]

CM can be effectively and efficiently removed by hemodialysis (HD). High-flux dialysis membranes with blood flows of between 120-200 mL/min can remove almost 50% of iodinated CM within an hour and 80% in 4 hours. Even in patients with CKD, in whom contrast excretion is delayed, 70-80% of contrast can be removed by a 4-hour HD treatment.

A meta-analysis by Cruz et al—8 trials (6 randomized and 2 nonrandomized, controlled studies) were included in the analysis, with a pooled sample size of 412 patients—indicated that periprocedural extracorporeal blood purification (ECBP) does not significantly reduce the incidence of CIN in comparison with standard medical therapy. ECBP in the study consisted of HD (6 trials), continuous venovenous hemofiltration (1 trial), and continuous venovenous hemodiafiltration (1 trial).[64] Cruz et al found that the incidence of CIN in the standard medical therapy group was 35.2%, compared with 27.8% in the ECBP group. Renal death (combined endpoint of death or dialysis dependence) was 12.5% in the standard medical therapy group, compared with 7.9% in the ECBP group.

An important consideration is the role of ECBP therapy in patients with severe kidney impairment (ie, stage 5 CKD) not yet on maintenance dialysis. A study by Lee et al indicated that in patients with CKD who are undergoing coronary angiography, prophylactic HD can improve renal outcome. The study included 82 patients with stage 5 CKD who were not on dialysis and who were referred for coronary angiography.[65] The patients were randomly assigned to either undergo prophylactic HD (initiated within 81 ± 32 min) or to receive intravenous normal saline (control group).

Of patients in the control group, 35% required temporary renal replacement therapy, compared with 2% of the dialysis group. In addition, long-term, postdischarge dialysis was required in 13% of the control patients but in none of the dialysis patients. Among those patients who did not require chronic dialysis, an increase in SCr at discharge of over 1 mg/dL from baseline was found in 13 patients in the control group and in 2 patients in the dialysis group.[65]

The study, though hopeful, does raise some concerns. While the change in creatinine clearance on day 4 from baseline was statistically significant, the day 4 creatinine clearance itself was not significantly different between the 2 groups. Also, the results were not expressed as CIN incidence.  How much time off dialysis a single HD session was able to buy these patients was not discussed. The duration of follow-up was also not clear.

Marenzi et al found better outcomes in patients who received venovenous hemofiltration both pre- and post-CM administration than in patients who received post-CM hemofiltration or no hemofiltration at all. These outcomes included a lower likelihood of CIN, no need for HD, and no 1-year mortality, in the pre-/post-CM group.[66]

The biggest confounder in studies of continuous renal replacement therapy (CRRT) is that the outcome measure (SCr) is affected by the treatment itself. While the advantage of CRRT is the lack of delay in its institution, contrast clearance rates would be 1 L/h (16.6 mL/min provided a maximal sieving coefficient for contrast across the hemofiltration membrane of 1), substantially less than standard HD. Furthermore, continuous venovenous hemofiltration is expensive, highly invasive, and requires trained personnel; the procedure itself needs to be performed in the intensive care unit (ICU). In the face of equivocal benefit of a highly invasive and expensive procedure, continuous venovenous hemofiltration has yet to be accepted as a method for preventing CIN.

Dialysis immediately after contrast administration has been suggested for patients already on long-term HD and for those at very high risk of CIN. Three studies looked at its necessity and found that LOCM can be given safely to patients with ESRD who are being maintained on HD without the added expense or inconvenience of emergent postprocedural HD.

The only condition in which HD might be argued to have a beneficial role is in patients on peritoneal dialysis who rely on their residual renal function. In this setting, HD performed soon after CM administration may provide enhanced removal and therefore protect residual renal function. It should be noted, however, that these patients on peritoneal dialysis would therefore need an additional HD procedure with concomitant vascular access, as the clearance with peritoneal dialysis would be far too slow to offer any protection.

Frank et al found that although the overall clearance of contrast was significantly increased by dialysis, the peak plasma concentration of iomeprol 15 minutes after contrast administration was not significantly changed by simultaneous dialysis. These investigators prospectively studied 17 patients with chronic renal insufficiency (SCr >3 mg/dL), dialysis independent, who were then randomized to receive high-flux HD over 6 hours simultaneously with contrast administration.[67]

Studies of HD for CIN vary with respect to the definition of CIN used, the patient population, the type and volume of CM, how long after CM administration HD is started, and, finally, the dialysis treatment modality itself. While existing studies do not show HD to be superior to hydration alone for CIN prevention, if HD is used in conjunction with hydration and CIN protective therapy, such as NAC and bicarbonate, it might prove to be efficacious in some high-risk patients. While the initiation of long-term dialysis was 5-15%, the progression to uremia over a long-term follow-up period is still unanswered.[16] . At this time, routine hemodialysis or hemofiltration either prophylactically or after contrast exposure is not recommended, no matter what level of initial kidney function.[2]

Ascorbic acid, which has antioxidant properties, was studied for its ability to counter the effect of free radicals and reactive oxygen species. One study found that oral ascorbic acid administered in a 3-g dose preprocedure and two 2-g doses postprocedure was associated with a 62% risk reduction in CIN incidence.[68]

Theophylline and aminophylline are adenosine antagonists that counteract the intrarenal vasoconstrictor and tubuloglomerular feedback effects of adenosine. They have been found to have a statistically significant effect in preventing CIN in high-risk patients. However, their use is limited by their narrow therapeutic window and adverse effects profile.

Vasodilators, such as calcium channel blockers, dopamine/fenoldopam, atrial natriuretic peptide, and L-arginine, all with different mechanisms of action, have a favorable effect on renal hemodynamics. However, their use for CIN prevention has not been borne out by most controlled trials, and they are not routinely recommended at this point.

Forced diuresis with furosemide and mannitol was studied in the hope that this procedure would dilute CM within the tubular lumen and enhance their excretion. Furosemide and mannitol in fact worsen CIN by causing dehydration in patients who may already have intravascular volume depletion. Their use at this time is discouraged.

The best therapy for CIN is prevention. Physicians need to be increasingly aware that CIN is a common and potentially serious complication. Patients at risk should be identified early, especially those with CKD (ie, estimated GFR [eGFR] < 60 mL/min/1.73 m2). A detailed history inquiring for risk factors, especially diabetes mellitus, should be ascertained.

A systematic review by Silver et al identified 12 models for predicting risk of CIN. Most of the higher-performing models included the following factors:

• Preexisting chronic kidney disease
• Older age
• Diabetes
• Heart failure or impaired ejection fraction
• Hypotension or shock

However, the review concluded that most of the models have only modest predictive ability, and are relevant only to patients receiving contrast for coronary angiography.[69]

In patients with risk factors for CIN, the possibility of alternative imaging studies that do not need contrast should be explored. MRI with gadolinium is no longer considered a safe alternative to iodinated contrast because of the risk of nephrogenic systemic fibrosis (NSF), an irreversible, debilitating condition seen mostly in patients with an eGFR of less than 30 mL/min/1.73 m2. Of gadolinium-based contrast agents, gadoteridol and gadobenate dimeglumine are considered to pose the lowest risk of NSF; no association of either agent with an unconfounded case of NSF has yet to be established.[70]

In patients with a moderate to severe risk of CIN, creatinine clearance rates or eGFR should be estimated by the Modification of Diet in Renal Disease (MDRD) formula, the Cockroft-Gault formula, or the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation and then measured again 24-48 hours after contrast administration.

In the emergency setting, where the benefit of very early imaging studies outweighs that of waiting, the imaging procedure can be carried out without an initial estimation of SCr or eGFR. A retrospective study of 382 patients undergoing contrast-enhanced computed tomography for suspected acute stroke found that eliminating creatinine screening prior to imaging did not adversely affect rates of CIN, hemodialysis, or mortality.[71]

Intra-arterial administration of iodinated CM poses a greater risk for CIN than does the intravenous approach. For patients at an increased risk for CIN who are receiving intra-arterial contrast, nonionic iso-osmolar agents (iodixanol) are associated with the lowest risk of CIN.

The length of time between two contrast procedures should be at least 48-72 hours. Rapid repetition of contrast administration has been found to be a univariate risk factor for CIN.

If possible, potentially nephrotoxic drugs (eg, nonsteroidal anti-inflammatory drugs [NSAIDs], aminoglycosides, amphotericin B, cyclosporine, tacrolimus) should be withdrawn in patients at risk (eGFR < 60 mL/min).

Metformin, though not nephrotoxic, should be used prudently, because if kidney failure does occur, there is risk of concomitant lactic acidosis. Therefore, metformin should be stopped at the time of the procedure and resumed 48 hours later if kidney function remains normal.

Angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) cause a 10-15% rise in SCr by reducing intraglomerular pressure. While they should not be started at the time of CM use, current guidelines do not recommend stopping them if the patient is already on them.

Minimizing contrast administration

The amount of contrast used during the procedure should be as little as possible and kept under 100 mL. Most investigators have found 100 mL to be the cut-off value below which no patient needed dialysis. The risk of CIN increases by 12% for each 100 mL of contrast used beyond the first 100 mL. Most angiographic diagnostic studies require 100 mL of contrast, compared with 200-250 mL for angioplasty. The maximum amount of contrast that can be used safely should be individualized, taking into account the preexisting kidney function.

Various formulas for calculating the maximal safe CM dose have been suggested. Two most often cited are those suggested by Cigarroa et al and the European Society of Urogenital Radiology (ESUR).[72, 73] Cigarroa et al, in a retrospective study of 115 patients undergoing cardiac catheterization and angiography, using the high-osmolarity CM diatrizoate, suggested that the dose of CM should not exceed 5 mL/kg of body weight (maximum 300 mL divided by SCr [mg/dL]). The ESUR, in turn, has published maximal low-osmolarity CM volumes for various SCr cutoff values.

While the formulas from Cigarroa and the ESUR take into account the SCr, it has been suggested that the eGFR (a more accurate predictor of kidney function) and the iodine dose of CM should be reflected in any estimates or predictions of safe CM dosages. There exists, however, no unimpeachably safe CM dose algorithm for CIN prevention.

Renin-angiotensin-aldosterone system blockade

A prospective, 50-month Mayo study found that renin-angiotensin-aldosterone system (RAAS) blockade, particularly in older patients with coronary heart disease, exacerbates CIN (43% incidence of dialysis and 29% progression to ESRD).[74] The marker used for kidney function was eGFR, as calculated by the MDRD formula. The study recommended withholding RAAS blockade 48 hours prior to contrast exposure.

RAAS blockade, however, can improve renal perfusion and decrease proximal tubular reabsorption, including CM absorption by the tubular cells. This effect can be documented with the increase in the fractional excretion of urea seen with low-dose RAAS therapy in patients with CHF and moderate CKD (the majority of the CIN-susceptible population).[75] In this group, reduction in intraglomerular pressure and filtration fraction from RAAS therapy might decrease tubular CM concentration and therefore lessen its adverse effects.

In a randomized pilot study conducted in 208 patients with moderate kidney insufficiency who were undergoing cardiac catheterization, withholding of an ACEI or ARB 24 hours or longer preprocedure resulted in a non-significant reduction in contrast-induced AKI and a significant reduction in post-procedural rise of creatinine. The authors suggested considering withdrawal of ACEI/ARB therapy as a low-cost intervention when referring a patient for cardiac catheterization.[76]

Despite some of the above data, current KDIGO guidelines do not support stopping RAAS blockade[2]

Hydration therapy, typically with intravenous isotonic saline, is the cornerstone of contrast-induced nephropathy (CIN) prevention. However, other agents have have demonstrated some benefit in prevention of CIN, including N-acetylcysteine (NAC) and statins.

Class Summary

Used for prevention of contrast toxicity.

Class Summary

These agents are used for their favorable effects on endothelin and thrombus formation, plaque stabilization and anti-inflammatory properties by improving lipid profile.

Class Summary

Sodium bicarbonate