Table of Contents

 

Introduction

Definition

Clinical Features

Pathophysiology

Risk Factors

Treatment History

Conclusions

 

N-Acetylcysteine and Radiocontrast-Induced Nephropathy

 

 

Jeff S. Rose, MD
Resident Grand Rounds

October 6, 2004

 

Introduction

 

Radiocontrast media is the third leading cause of hospital-acquired acute renal failure being preceded by hypotension and surgery (1).  Occurrence varies by study and ranges between 11% and 45% depending on comorbid conditions and defining parameters in patients undergoing coronary angiography/intervention (2).  Radiocontrast-induced nephropathy (RIN), also known as contrast media-associated nephrotoxicity or contrast nephropathy, occurs at rate of approximately 150,000 cases/year, out of approximately 1,000,000 cardiac catheterizations/year (3).  It is estimated an additional nine million contrasted computed tomography scans are performed yearly, and there is a growing preponderance of invasive contrasted interventional procedures putting more patients at risk of RIN.

 

The economic impact and natural history of RIN cannot be ignored.  Second generation nonionic low-osmolar contrast agents (Optiray at Baptist Hospital) cost the patient at least three times the cost of first generation contrast with the wholesale cost estimated at approximately $106 per 120 mL.  These agents are quoted at being up to ten times the cost of older agents and estimated to cost the health care system an additional one billion dollars per annum (4).  These measures have been taken to reduce the incidence of RIN.  McCullough (5) demonstrated a significant risk (relative risk of 28.6%) of in-hospital mortality when dialysis was required (Figure 1) in the setting of acute renal failure.  For those not requiring dialysis, hospital stay is estimated to be prolonged by 2 days at expense of $148 million annually (6).  Most cases of RIN are reversible and nonoliguric in nature, however, up to 30% may have some degree of permanent renal insufficiency.  A study by Levy et al (7) compared patients who developed acute renal failure in the setting of RIN to patients matched for age, baseline serum creatinine level and procedure type to those who did not.  They found a high rate of mortality (34%) in those who developed overt acute renal failure compared to only 7% (p<0.001) of  patients without acute renal failure (7).  When severity of comorbid conditions were matched by using APACHE II scores, mortality from acute renal failure increased (odds ratio of mortality 5.5).  Nevertheless, the majority of patients with RIN will recover their preprocedural renal function.

 

In the following pages, I hope to define RIN, describe the clinical features, outline the pathophysiology and associated risk factors and, most importantly, analyze the role of NAC in RIN.  I will briefly gloss over other preventative measures which have been studied in the past and their role in the future.

 

 

 

 

 

 

 

 

 

 

 

 

                           Figure 1

 

Definition

 

Radiocontrast-induced nephropathy (RIN) is typically defined as a sudden decline in renal function after radiocontrast administration.  Defining parameters vary throughout the literature, and I will focus on studies with the following criteria:

  • absolute rise in serum creatinine (sCr) of 0.5 mg / dL within 48 to 72 hours after contrast administration

  • 25% or greater increase of sCr within 48 to 72 hours after contrast administration regardless of baseline sCr

  • 25% or greater decrease of glomerular filtration rate within 72 hours after contrast administration

 

It must be noted, however, there is no “standard” for the definition of RIN, and many older studies have used an absolute increase of sCr of 0.25 to 1.0 mg / dL.

 

Clinical Features

 

Clinically, the diagnosis of RIN is made after an abrupt deterioration of renal function after radiocontrast exposure in the absence of other causes of renal failure.  The above criteria can be applied clinically to make the diagnosis.  Typically, RIN presents as an asymptomatic, non-oliguric rise in serum creatinine 24 to 48 hours after contrast exposure, peaks within three to five days and typically resolves with a return to baseline sCr by days seven to ten (8).  The time course for severe cases is prolonged with an increase of sCr up ten days and a gradual return over the following two to four weeks.  Patients with advanced renal insufficiency experience oliguric renal failure in up to 30% of cases.

 

Examination of the serum and urine are relatively nonspecific.  Urinalysis usually yields renal tubular epithelial cells, granular casts and low-grade proteinuria; however the urinalysis may also be bland.  Occasionally, calcium oxalate or uric acid may be seen.  The fractional excretion of sodium is typically less than 1%, mimicking extracellular volume depletion (8).  This phenomenon must also be differentiated from atheroembolic disease and cholesterol emboli syndrome after intravascular manipulation.  The distinction can generally be made by the delayed onset and protracted course of embolic disease as opposed to the quick onset and resolution of RIN.  One may also see the stigmata of embolic disease including worsening or new onset hypertension, livedo reticularis, distal digital infarcts and ischemic digits.  Laboratory examination of embolic disease may reveal eosinophilia, decreased complement levels and elevated inflammatory markers.  This distinction must be made as atheroembolic syndromes carry a high overall mortality ranging 60-80%, and aggressive supportive measures need to be pursued.  This is contrasted to RIN which typically has a benign course.

 

Pathophysiology

 

The pathophysiology of RIN is poorly understood, under significant debate and is presumed multifactorial.  It is believed four dominant injury pathways exist, and it is likely that these factors work together in concert to induce RIN in a given patient.  The predominant factor appears to be intrarenal vasoconstriction causing renal medullary hypoxia culminating in cell detachment, apoptosis and necrosis.  Since the renal medulla is normally deficient in oxygen, with a PaO2 of 10 to 20 mm Hg, it is readily susceptible to further hypoxia.  Second, radiocontrast medium may precipitate in the distal tubule lumen, along with glycoproteins, forming casts.  Third, radiocontrast medium may directly damage tubular cells via the difference in osmolality or direct cytotoxicity.  Finally, reperfusion injury may occur after initial tissue ischemia via reactive oxygen species production (3,9). 

 

Alteration of renal hemodynamics, namely vasoconstriction, is the primary proposed mechanism of RIN and is affected by a variety of factors.  These include adenosine, calcium and endothelin (21).  There also appears to be an association between comorbid conditions such as age and diabetes that decrease vasodilating factors including nitric oxide and prostaglandins. We will discuss these factors more in depth here.

 

Adenosine acts as a vasodilator peripherally, whereas intrarenally, it actions include vasoconstriction and are associated with a reduction in renal blood flow and, therefore, glomerular filtration rate.  Hypertonic tubular fluid delivered to the macula densa influences as increase in adenosine release by the proximal tubular cells, endothelial cells and vascular smooth muscle cells thereby causing intrarenal vasoconstriction.  This adenosine-induced vasoconstriction is antagonized by theophylline leading to studies as a possible renoprotective agent which will be discussed later.

 

Calcium also appears to have a major role in radiocontrast-induced vasoconstriction.  A study by Bakris and Burnett (10) analyzed vasoconstriction and the effects of calcium entry of the renal vasculature in dogs after contrast infusion.  Administration intrarenal verapamil, diltiazem or a calcium chelator was shown to markedly decrease the magnitude and duration of contrast-induced intrarenal vasoconstriction compared to controls.  It has also been noted elsewhere that cytosolic calcium overload causes cell blebbing and death (21).

 

Nitric oxide and vasodilatory prostaglandins, namely PGE1 and PGI2, have vasodilatory effects that may be beneficial in preventing RIN.  Attempts to augment PGE1 have failed to date.  On the other hand, PGE2 increases causing intrarenal vasoconstriction; no studies have directly addressed blockade of PGE2.  Endothelin, a peptide produced in renal endothelial and mesangial cells, has been shown to cause an intense and long-lasting vasoconstriction of renal vasculature after contrast administration.  Radiocontrast stimulates the release of this vasoactive peptide and has been studied for therapy intervention.

 

Tubular obstruction occurs by two proposed mechanisms: radiocontrast administration leads to cast formation from aggregation of Tamm-Horsfall proteins and increased urate excretion with tubular precipitation.

 

Risk Factors

 

Several factors have been identified that place patients at risk for development of RIN.  Common risk factors include preexisting renal insufficiency, diabetes with known concurrent renal insufficiency, intravascular volume depletion, type and dose of contrast used and older age.  These will be expanded upon here.

 

Normal renal function

It is currently believed individuals with normal renal function are not at risk of development of RIN, with retrospective studies estimating incidence of RIN at 0.1%. 

 

Diabetes

Diabetes is frequently quoted as an independent risk factor of RIN, however a study by Parfrey et al (12) found diabetic patients with normal renal function are not at increased risk.  This prospective study compared six groups of patients: diabetics with normal renal function, nondiabetics with renal insufficiency and diabetics with renal insufficiency with controls undergoing non-contrasted computed tomography and experimental group receiving contrast.   The incidence of RIN (3.4%) in diabetic patients with normal renal function was similar to the control diabetic group (1.5%).  It was concluded the risk of acute renal insufficiency attributable to contrast was minimal in patients with diabetes with normal renal function and nondiabetic patients with chronic renal insufficiency.  Although not demonstrated in this study, multiple other studies quoted in this paper found increased incidence of RIN in patients with moderate to severe renal insufficiency. 

 

Preexisting renal insufficiency

This is considered to be the most important risk factor for the development of RIN.  As one’s sCr increases, the incidence of RIN increases.  Barrett et al (13) found the incidence of RIN at 6% with a concurrent baseline sCr 1.4 mg/dL with an increase to 16.7% with baseline sCr 2.25 mg/dL (Table 2).  This has been duplicated in other studies (Table 2).  The probability of requiring dialysis after the development of RIN in this group ranges from 0.04% to 48% as the precontrast creatinine clearance decrease from 50 to 10 mL / min (Table 3) (5).    

 

                             


Diabetes with concurrent renal insufficiency

This cohort of patients is at the greatest risk of development of RIN and severe renal dysfunction often requiring acute or chronic hemodialysis.  Probability of requiring dialysis increases from 0.2% to 84% as the precontrast creatinine clearance decreases from 50 to 10 ml / min (Table 4) (5).  Wexler et al (24) also found risk in diabetic patients with renal insufficiency to be 3.7 times higher than those with renal insufficiency alone (Table 5).  It can therefore be ascertained that patients with preexisting renal insufficiency and diabetes with presumed micro and macrovascular disease are much higher at risk of developing RIN. 

 

           

Contrast

There are four primary types of contrast media used in routine practice today: nonionic low-osmolal, ionic low-osmolal, nonionic iso-osmolal, and ionic high-osmolal contrast.  First generation ionic high-osmolal monomers are extremely hyperosmolal at 1500-1800 mosmol/kg and were found to be associated with higher rates of nephrotoxicity.  It was initially believed the use of second generation nonionic low-osmolality contrast would be associated with a decline in RIN in all patient populations.  Second generation agents such as Optiray used at Baptist Hospital have measured osmolality of 600-850 mosmol/kg, but are quite hyperosmolal when compared to plasma (300mosm).  In studies of patients with moderate renal insufficiency, sCr between 1.4 and 2.4 mg/dL, nonionic, low osmolality agents have been associated with a lower incidence of RIN.  Wexler et al (24) found the incidence of RIN in ionic contrast (meglumine / sodium diatrizoate) to be more than double that of nonionic, low-osmolal contrast (iohexol) (Table 5).  These effects were found primarily in patients with renal insufficiency and diabetes with concurrent renal insufficiency.  Other subsequent prospective studies compared nephrotoxicity incidence between the two contrasts with varying results.  As with Wexler (24), most failed to find a significant difference in development of RIN in patients who are not risk (3,22,23).  A more recent randomized controlled trial by Rudnick et al (22) confirmed the findings of Wexler et al by comparing the incidence of RIN between high- and low-osmolal contrasts.  In patients using low-osmolal contrast, they found a significant decrease (15%) in contrast nephropathy in high-risk patients with known renal insufficiency (sCr >1.5 mg/dL) and diabetes mellitus, a small decrease (3%) in patients with renal insufficiency only and no difference in non azotemic patients.  Low-osmolal agents have now become standard of care in most institutions and should always be used in high-risk patients. 

 

Newer generation nonionic contrast agents are available with iso-osmolal concentration at 290 mosmol/kg and may decrease risk of RIN.  However, a recent study demonstrated a decrease in medullary oxygen level in both low- and iso-osmolar agents, which indicate that both low- and iso-osmolal agents are able of potentiating ischemic insult and, therefore, RIN.  Other studies note a lower peak increase in the plasma sCr within the first three days after contrast administration (0.13 versus 0.55 mg/dL) in those receiving iso-osmolal contrast.  Of note, since Baptist is part of a committed agreement with Premier Purchase Group, the hospital must use Optiray contrast and is unable to provide any nonionic iso-osmolar contrast due contract limitations.

 

The volume of contrast also plays a role in RIN.  It is generally believed volumes less than 70 mL confer less risk, and larger volumes are associated with higher incidence of RIN, although there are exceptions to this statement.  A frequently quoted study by Cigarroa et al (25) compared patients receiving a maximal precalculated ionic high-osmolal contrast volume to those without volume limitations.  Patients had baseline sCr >1.8 mg/dL and RIN was defined as absolute increase sCr >1.0 mg/dL.  Only 2% of patients to which the formula was applied developed RIN, while 26% of those without volume limitations developed nephropathy.

 

Other factors

Several other factors influence the development of RIN.  Advanced age is associated with a dominance of renal vasoconstrictive forces instead of vasodilatory forces.  This may be related to decrease prostaglandin E2, a renal vasodilator, in healthy older people, rendering the kidney vulnerable to contrast induced ischemia (21).  Intravascular volume depletion, NYHA class III and IV congestive heart failure and hepatic cirrhosis are also associated with increased risk.  It is currently debated whether multiple myeloma is an independent risk factor for development of RIN.  Nonetheless, multiple myeloma is frequently associated with intravascular volume depletion, renal insufficiency with which risk of RIN development increases.  Metformin alone is not nephrotoxic but has long been associated with lactic acidosis in the setting of renal insufficiency.  Although uncommon, metformin associated lactic acidosis carries a mortality of 50%.  In light of these findings, it is generally recommended to discontinue metformin 48 hours prior to a procedure using intravenous contrast media and withhold treatment until resolution of RIN, normalization of sCr or 48-72 hours postprocedure (21).  This practice is currently being debated.

 

Treatment history

 

Clinical trials abound testing multiple modalities of RIN prophylaxis, most having sound theoretical benefit.  These will be briefly reviewed here.

 

Calcium channel antagonists

As described above, calcium plays a significant role in intrarenal vasoconstriction making calcium channel blockers an attractive RIN prophylactic agent.  Theoretically, calcium channel blockers would antagonize afferent arteriole constriction thereby maintaining GFR.  As a class, calcium channel blockers have been shown to retard the decline in GFR and the duration of intrarenal vasoconstriction after contrast exposure.  Russo et al (32) examined the protective role of nifedipine in a randomized, double-blind, placebo-controlled study.  Nifedipine was found to prevent the anticipated deterioration in renal hemodynamics induced by high-osmolal contrast.  Patients who received placebo had a significant decrease in renal blood flow as measured by enzymuria (inulin and para-aminohippurate).  However, this study only took measurements up to two hours after contrast administration, only included low-risk patients (none had renal insufficiency or diabetes) and did not mention fluid hydration rates.  Serum creatinine measurements were not followed up to 72 hours after contrast; the study failed to analyze the “gold standard,” if you will, of measured sCr as most studies mentioned in this paper.  Two other studies that showed “benefit” from calcium channel blockade used similar protocols to the Russo study, measuring only urine enzymes, not serum creatinine.  It should be noted, though, that the above studies did show a significant preservation of GFR and amelioration of enzymuria, suggestive of protection against RIN.

 

A more recent study by Spangberg-Viklund et al (33) randomized 27 patients with normal to moderate renal insufficiency (15 diabetics and 12 non-diabetics) to receive either oral felodipine XR or placebo in addition to at 2 L / 24 h of intravenous hydration.  Patients in the experimental arm had a significant increase in sCr from baseline, while the patients receiving placebo did not.  However, the others do contend felodipine may have mild renoprotective benefit in those with more advanced renal insufficiency after subgroup analysis.

 

Given the contradictory findings, an ongoing multicenter, international, randomized, double-blind, placebo-controlled trial testing the effect of the calcium channel antagonist amlodipine, started 7 days before and continued 2 days after the injection of non-ionic radiocontrast media has been initiated (34).  Chronic renal failure patients (calculated creatinine clearance between 10 and 60 ml/min) with scheduled intravascular radiographic investigation will be included. They will receive standard hydration with 0.45% saline infusion.  The total number of patients to enter the study will be 290. Results are expected for 2004. 

 

Unfortunately, firm conclusions cannot be drawn from the published trials, as study populations are small, measured outcomes are not standardized, as well as study protocols for hydration, dosage and duration of medication administration.  Large, randomized, prospective trials, like the one mentioned above, are needed to further delineate the role of calcium channel blockade in RIN.

 

Mannitol

Theoretically, mannitol is believed to increase tubular flow rates and reduce the time of contrast exposure in addition to increasing atrial natriuretic peptide production thereby increasing GFR.  Early studies (27) showed benefit with a relative risk reduction of 71% in patients receiving mannitol immediately after contrast infusion.  The control group received no therapy including hydration, limiting the validity of this study.  A follow-up study to the Anto trial addressed this deficit and found no significant difference between the group receiving mannitol and the control group receiving saline hydration.  However, a study analyzing the benefit of mannitol helped develop a significant change in the strategy to prevent RIN.  This randomized, prospective trial by Solomon et al (28) found patients (baseline sCr of >1.6 mg/dL) receiving 1 mL/kg/h IV 0.45% saline had a significantly reduced incidence of RIN (11%) compared to those receiving a combination of saline and mannitol (28%) or saline and furosemide (40%), establishing the renoprotective benefit of hydration.  Given these results, it would be wise to avoid mannitol prior to or after contrasted studies.

 

Theophylline/Aminophylline

Theophylline acts as an adenosine antagonist and theoretically would block the effect of adenosine thereby decreasing intrarenal vasoconstriction.  Since theophylline is simple to use, inexpensive and rapid-acting, it would be an attractive prophylactic agent.  A number of studies have addressed theophylline using varying doses, dosage forms and yielded conflicting results.  Four randomized, prospective studies found a significant lack of renal function deterioration in low-risk patients undergoing cardiac catheterization receiving hydration and theophylline to hydration alone.  A randomized, controlled, double-blinded study by Huber et al (29) compared hydration versus hydration and theophylline in high-risk patients and found similar results.  Only 4% developed RIN in the theophylline group compared to 16% in the hydration alone group.  However, other studies have not found any benefit in patients using theophylline for RIN prophylaxis.  As there were inconsistencies in dosing, administration regimens and definition of RIN in the aforementioned studies, solid conclusions cannot be made.  Although the above studies suggest benefit, theophylline is currently not recommended as a prophylactic agent for RIN until larger prospective trials further investigate its use.

 

ACEi

Many studies have identified ACE inhibitors as a risk factor for RIN (30).  However, a study by Gupta et al (31) challenged this concept with the belief that contrast related renal blood flow reduction was related to afferent arteriole constriction via the renin-angiotensin system.  Patients were randomized to receive captopril and hydration versus hydration alone beginning at least three hours prior to cardiac catheterization and continued for at least six hours after.  Only 6% of the experimental group compared to 29% (p<0.02) developed RIN.  Again, more studies are needed to evaluate ACEi use in RIN prevention as this was a small trial (n=71), and other trials have exhibited unfavorable associations between ACEi and RIN.

 

Dopamine

“Renal-dose” dopamine has been discussed in the literature at length.  Low-dose dopamine (0.5 to 3 mcg/kg/min) predominately activates DA-1 dopamine receptors which increases renal blood flow by intrarenal vasodilatation.  Activation of the DA-1 receptor results in an increase in natriuresis and renal blood flow, whereas DA-2 activity results in vasoconstriction intrarenally.  The goal of dopamine therapy is to maximize DA-1 receptor effect while minimizing DA-2 receptor activity.

 

A study by Kapoor et al (31) compared dopamine infusion to placebo in diabetic patients undergoing coronary angiography.  Fifty percent of the control compared to 0% of the dopamine group had elevation of sCr >25%.  Another study comparing saline infusion to dopamine infusion in patients with chronic renal insufficiency in patients undergoing abdominal arteriography or aortography of the lower extremities found a significant reduction in RIN in the experimental group.  However, these studies did not use standardized saline infusions and had varying contrast volumes.

 

A randomized, prospective study by Abizaid et al (32) studied patients with chronic renal insufficiency, some with diabetes, who underwent cardiac catheterization.  In this study, all groups received fluid hydration with 0.45% saline at 1mL/kg/h beginning 12 hours prior to the procedure and continuing 12 hours after.  The two experimental groups received dopamine (2.5 mcg/kg/min) plus saline or aminophylline (4 mg/kg followed by a drip of 0.4 mg/kg/hour) plus saline.  Fifty percent of patients receiving dopamine and saline infusion developed RIN, compared to only 30% of those receiving saline alone.  This difference was not significant in either the dopamine or aminophylline group.  The authors actually found dopamine to have a deleterious effect on the severity of renal failure, prolonging the course.  There were several limitations to the current study. A very sensitive definition of contrast-induced ARF, the number of patients enrolled was small, which may have precluded detecting a difference among the groups and there was a lack of information regarding fluid balance and body weight; patients treated with dopamine or aminophylline may have had a negative fluid balance and a contribution of prerenal azotemia to the elevated creatinine.  Subsequent studies have also lacked evidence showing a protective effect of dopamine for prevention of RIN.            

Since selective activation of DA-1 receptors cannot be reliably achieved, there have been reports of “spillover stimulation” of alpha and beta adrenergic receptor contributing arrythmias and other complications of acute renal failure.  Also, many studies have found an increased incidence of nephrotoxity in diabetic patients.  Again, there is conflicting evidence regarding the efficacy of dopamine in prophylaxis against RIN, and its use cannot be supported at this time.

 

Endothelin receptor antagonists

Endothelin-1, a potent vasoconstrictor, is thought to play a role in the development of RIN.  Animal studies support the protective role of endothelin receptor blockade after contrast administration.  A multicenter, double-blind, prospective study of 158 human patients comparing an endothelin antagonist to placebo failed to confirm these findings, however.  All patients received hydration (1 ml/kg/h) prior to after contrast exposure.  The incidence of RIN was 27% higher with endothelin blockade (p<0.05) (31).

 

Fenoldopam

The dopamine-1 receptor agonist (fenoldopam) is a vasodilator derived by modification the structure of dopamine making it a pure DA-1 agonist without DA-2, alpha-adrenergic, beta adrenergic stimulation as seen with dopamine.  Fenoldopam induces renal vasodilatation thereby decreasing renal vascular resistance, increasing medullary blood flow, glomerular filtration rate and urinary sodium and water excretion (23).  A recent literature review by Asif (21) analyzed five studies including two case series, two RCTs and one retrospective analysis.  The data were inconclusive and did not support for the use of fenoldopam.  Reasons for this included lack of a large RCT, lack of a significant difference in treatment and control groups, associated hypotensive risk with infusion and suboptimal infusion and dosing regimens.  A large RCT could address these studies’ shortfalls.

 

Hydration

Hydration with 0.45% normal saline or normal saline has been shown to more effective than placebo in multiple trials.  Most studies have used an algorithm including 0.45% NS for pre-contrast hydration.  A prospective, randomized, controlled, open-label study by Mueller et al (37) compared the incidence of contrast media–associated nephrotoxicity with isotonic or half-isotonic hydration.  Patients scheduled for elective or emergency coronary angioplasty were randomly assigned to receive isotonic (0.9% saline) or half-isotonic (0.45% sodium chloride plus 5% glucose) hydration beginning the morning of the procedure for elective interventions and immediately before emergency interventions.  RIN was defined as an increase in serum creatinine of at least 0.5 mg/dL within 48 hours. Secondary end points were cardiac and peripheral vascular complications.

 

A total of 1620 patients were assigned to receive isotonic (n = 809) or half-isotonic (n = 811) hydration. Primary end point analysis was possible in 1383 patients. Baseline characteristics were well matched. Contrast media–associated nephropathy was significantly reduced with isotonic (0.7%, 95% confidence interval, 0.1%-1.4%) vs. half-isotonic (2.0%, 95% confidence interval, 1.0%-3.1%) hydration (P = .04). Three predefined subgroups benefited in particular from isotonic hydration: women, persons with diabetes, and patients receiving 250 mL or more of contrast.

 

The authors found isotonic hydration to be superior to half-isotonic hydration in the prevention of contrast media-associated nephropathy.  However, this data should be taken with a grain of salt as glucose was included in the 0.45% NS algorithm.  This is important as many of the aforementioned studies do not have glucose included in their study protocols.  Also, as shown in stroke studies, glucose exposure is often detrimental and expands the ischemic penumbra.  This may relate to contrast-induced ischemia in the renal medulla and progression of nephropathy.  A study comparing normal saline to 0.45% NS would best investigate a difference in hydration techniques.

 

Merten et al (38) showed a significant decline in RIN in patients receiving a sodium bicarbonate infusion compared to those receiving saline alone.  This prospective, single-center RCT of 119 patients with stable serum creatinine levels of at least 1.1 mg/dL were randomized to receive either NS or sodium bicarbonate as a bolus 3 mL/kg/h for one hour before and 1 mL/kg/h for six hours after the contrasted procedure.  RIN was defined as an increase of 25% or more in serum creatinine within 48 hours.  Baseline sCr in the study group was 1.89 mg/dL, 1.71 mg/dL for the control group; this difference was not significant.  There were no other differences in patient baseline characteristics.  A nonionic, low-osmolal was used in this study. 

 

RIN occurred in 8 patients receiving NS and only 1 (1.7%) receiving sodium bicarbonate infusion.  This difference was significant and led to a follow-up registry of 191 patients in which RIN occurred in only 1.6% of those receiving sodium bicarbonate, confirming the RCT results.  It is believed sodium bicarbonate produces its effect by increasing medullary pH thereby protecting it from oxidant injury.  This effect is also noted when acetazolamide is used in rats receiving contrast, and in those undergoing hemofiltration which leads to serum alkalinization.  There are limits to this study, however.  The patient sample size is small, and the study has to be reproduced outside the original study center.  Power was achieved in this study. 

 

N-acetylcysteine

Recently, Tepel et al (26) showed that the antioxidant N-acetylcysteine attenuates RIN in patients with CRI undergoing contrasted computed tomography with a relative risk of 0.11, establishing N-acetylcysteine as a potent prophylactic agent against RIN.  N-acetylcysteine has three mechanisms by which it prevents RIN.  NAC may combine with nitric oxide, leading to more stable storage forms of NO.  This increases direct vasodilating properties of NO on the kidney, prevents tolerance to the vasodilatory effects of nitrates, contributing to improved renal hemodynamics and decreased medullary hypoxia.  It has the ability to scavenge oxygen-free radicals, possibly preventing direct oxidative tissue damage that occurs in patients receiving contrast.  As a precursor of glutathione synthesis, N-acetylcysteine significantly increases intracellular redox potential and, likely, improves the reductive status of critical regulatory protein thiol groups. 

 

N-acetylcysteine: The Data

 

For

MacNeill et al (36) performed a double-blind, placebo-controlled RCT of 43 patients with baseline sCr >1.5 mg/dL undergoing cardiac catheterization.  All patients were prehydrated with 1ml/kg/h of 0.45% NS, 21 received five doses of 600 mg of N-acetylcysteine with two doses given before the catheterization, 22 received placebo.  An increase of sCr of 25% or greater from baseline defined RIN.  The primary protocol difference in this study included the first dose administered at time of randomization and the second at four hours.  No significant differences existed between the control and experimental groups; 20 diabetics were included in the study.                                                                                   

 

 

 

 

 

 

 

 

 

 

          Figure 6

 

The authors found a significant reduction in RIN in patients receiving N-acetylcysteine with only one patient diagnosed in the experimental arm compared to seven in the placebo group (p<0.05).  Those receiving placebo had continued increase in sCr measurements after 24 hours (Figure 6).  This study also helps establish merit in pretreating patients undergoing day-case and semiemergent procedures.  Limitations include a small study population size as enrollment was terminated when statistical significance was achieved.  Creatinine levels peaked at four to five days post exposure, therefore sCr levels at 72 hours are believed to fail to detect 10% of cases.

 

A recent meta-analysis by Alonso et al (2) published in January, 2004, reviewed the results of eight double-blinded and unblinded randomized controlled trials using N-acetylcysteine for the prevention of RIN in humans older than 18 with renal insufficiency (baseline creatinine ranging from 1.3 to 2.8 mg/dL).  The authors used an Ovid multidatabase search of MEDLINE, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and HealthSTAR to identify candidate articles using the search terms “acetylcysteine, Parvolex, Mucomyst, radiocontrast nephropathy” or “contrast-induced renal failure”.  The authors also searched proceedings of the American Society of Nephrology (1999-2003), American College of Cardiology (1999-2003), American Society of Hypertension (2000-2003) and Transcatheter Cardiovascular Therapeutics (2001-2002).  Over 6,000 selections were listed, twenty-five articles initially met inclusion criteria.  Of the initial 25 articles, only eight met the information requirements of the authors. 

 

Intravenous fluid administration was considered standard therapy in all studies, but varying algorithms were used between the studies.  Two studies used NS, while the other six used .45% NS with varying pre and post procedure timing.  However, all patients received 1 ml/kg/h, and most received hydration 12 hours before and after contrast administration.  Mean radiocontrast volume ranged from 75 to 230 mL, with seven studies using low-osmolal contrast and one iso-osmolal.  N-acetylcysteine was administered orally in seven studies.  Five studies used 600 mg po bid for four doses, one 400 mg po bid for four doses and another 1,200 mg prior to contrast exposure.  The last study administered IV N-acetylcysteine at 150 mg/kg 30 minutes prior to exposure followed by 50 mg/kg over four hours.

 

The primary analysis included eight full-text published RCTs, totaling 885 patients (2).  Overall, there were 35 cases of RIN in the N-acetylcysteine group compare to 82 in the control group receiving hydration alone.  Overall relative risk associated with N-acetylcysteine use was 0.41 (CI 0.22 to 0.79, p<0.01) (Figure 8).  Estimated NNT was eight to prevent one case of RIN.  Subgroup analyses showed significant benefit in patients with creatinine <1.9 mg/dL receiving >140 mL contrast volume. 

 

This study analyzed data of eight RCTs, only three of which did not support the prophylactic use of N-acetylcysteine.  However, in one of the three studies, N-7 acetylcysteine was administered one hour prior to the contrast load, possibly contributing to the lack of efficacy (2).  Also, CRI is associated with depleted glutathione peroxidase; the antioxidant effect may require a longer duration of N-acetylcysteine exposure.  It is also important to note that in the study by Baker et al (1), IV administration of N-acetylcysteine began 30 minutes prior to the contrast study.  This contradicts Alonso’s assumption of a prolonged pre-exposure window proposed as the weakness in the Durham et al (17) RCT.  It may also open a window for the use of IV N-acetylcysteine. 

                               

 

 

 

 

 

 

 

                                                Table 7

 

Alonso supports the use of N-acetylcysteine for prevention of RIN.  These data suggest that N-acetylcysteine prevents RIN in subjects with mild to moderate CRI, typically receiving a greater contrast load.  However, it is difficult to interpret the subgroup analyses due to power limitations, as they may not represent true differences.  Also, the cutoff value 140 mL of contrast was arbitrarily chosen as N-acetylcysteine was found beneficial in only one study (Tepel) receiving less than this volume. 

 

In another meta-analysis by Birck et al (39) which combined the effect sizes of six randomized controlled trials that used N-acetylcysteine for prevention of contrast nephropathy in chronic renal insufficiency, a significant 56% relative risk reduction in patients given N-acetylcysteine.  A major difference in this trial lies in the inclusion of patients receiving contrast media for cardiovascular interventions only; thereby excluding the Tepel study.  This study also excluded the study by Baker which used an intravenous regimen.  Study design details differed between the analyzed trials particularly the degree of chronic renal insufficiency before the procedure and the amount of radiocontrast media given.  Since both variables have been reported as independent predictors of contrast nephropathy these discrepancies might have contributed to the heterogeneous results.  However, meta-regression showed no significant relation between these covariates and the relative risk of contrast nephropathy as a dependent variable.  

 

Image

 

 

 

 

 

 

 

 

 

 

 

 

                       Figure 8

 

Again, this meta-analysis showed a significant benefit of N-acetylcysteine treatment in prevention of contrast nephropathy in patients with chronic renal insufficiency. Whether the observed reduction in relative risk of an arbitrarily defined increase in serum creatinine will confer benefit in clinical practice remains controversial. Nevertheless, the reported association of contrast nephropathy with increased morbidity, mortality, and lengthened hospital stay might justify use of N-acetylcysteine for prophylaxis of contrast nephropathy since it is cheap, easy to use, and has a favorable side-effect profile. Trials designed to investigate the course of serum creatinine after radiocontrast media application and to elucidate the effect of N-acetylcysteine on hard clinical endpoints are warranted.

 

Both the meta-analyses by Birck and Alonso et al have several limitations that should be taken into account. First, Birck et al acknowledged the possible presence of bias, particularly the absence of small trials with negative results.  They could not rule out that publication bias might lead to an overestimation of the true treatment effect.  Second, all included studies used the surrogate endpoint of contrast nephropathy as a primary outcome. Contrast nephropathy was defined as an increase of serum creatinine of more than 25% from baseline values, which is, especially in patients with pre-existing renal insufficiency, a minor deterioration of renal function. Even in the high-risk cohorts included, contrast nephropathy was almost always transient and only rarely needed dialysis-indicated by an overall incidence of dialysis dependency of 0.7% (seven of 805).  A heterogeneous patient population is noted in both paper, and Alonso included different modalities of contrasted studies.  Alonso also noted differences in baseline sCr, doses of N-acetylcysteine, hydration protocols and even definition of RIN.  However, the authors derived their definition of RIN from the different definitions used in the individual studies.  This analysis was limited to summary measures of unadjusted RIN rates.  A pooled analysis of individual patient data would be necessary to assess variables and risk factors for RIN.

 

Despite the reported association of contrast nephropathy with impaired outcomes, particularly in patients with advanced renal failure or requiring dialysis after administration of radiocontrast media, no trial was designed to investigate the effect of N-acetylcysteine on hard clinical endpoints such as in-hospital morbidity, mortality, or dialysis dependency. Only the trial implemented by Kay and colleagues also investigated the effect of N-acetylcysteine on the secondary endpoint of length of hospital stay and found a significant reduction of a half day in patients given this drug. Thus, the clinical relevance of the renoprotective effect of N-acetylcysteine remains somewhat debatable whereas periprocedural use of adequate hydration regimens is of proven benefit in preventing contrast nephropathy as emphasized by multiple studies.

 

Against

A recently completed RCT by Fung et al (19) looked at 91 patients with chronic renal disease graded moderate to severe with a baseline creatinine level of 1.69 to 4.52 mg/dL undergoing coronary procedures.  Patients were randomly assigned to receive oral N-acetylcysteine in three 400 mg doses on the day before and also on the day of the procedure (n=46) or intravenous hydration (n=45).  Although a different regimen than most studies, cumulative doses were the same at 1200 mg/day.  Both groups were administered intravenous fluid hydration with normal saline at 100 cc/hour from 12 hours before until 12 hours after the procedure, with exclusions including clinical heart failure (6 in the experimental and 7 in the control groups).  A nonionic low-osmolality contrast iopromide was used for all procedures.  Power was calculated on an incidence of RIN in 21% of controls, 2% of treatment groups from previous analysis, therefore requiring the attained cohort number to be 90.  The definition of RIN was an increase in serum creatinine concentration of 0.5 mg/dL or reduction of 25% or greater of the baseline value.

 

There were no significant differences between the two groups in baseline characteristics, the group receiving N-acetylcysteine had marginally more patients with hypertensive nephrosclerosis, renal artery stenosis, greater estimated GFR at baseline, larger dose of contrast administered and who were on aspirin therapy.  Also fewer of the experimental group were on an angiotensin receptor antagonist.  No patients were lost to follow-up during the study.  Six (three with diabetes) patients developed RIN in the control group versus eight (two with diabetes) in the NAC group (p=0.8).  There was no significant difference in subgroup analyses comparing presence or absence of diabetes, estimated GFR mL/min or less or those to be greater than 30 mL/min.  No adverse events were encountered in either group.  The study was grossly underpowered as the original power was estimated on the risk reduction by Tepel (26).  This study enrolled only 91 patients whereas 300 would be required to provide 80% power.

 

Three recent studies by Durham et al (17), Oldemeyer et al (11) and Goldenberg et al (14) again failed reduce the incidence of RIN after cardiac angiography in patients with mild to moderate renal insufficiency.  These were randomized, placebo controlled, double-blind trials with all patients receiving IV hydration and the study group receiving 1,200 mg of N-acetylcysteine before the procedure with repeated doses after.  Protocols varied between the studies with Durham giving 1,200 mg an hour prior to and then three hours after the procedure, Oldemeyer giving 1,500 mg the evening before the procedure followed by three doses every 12 hours and Goldenberg giving 1,200 mg bid the day before and the day of the study.  All patients had CRI with baseline sCr ~1.6 mg/dL in two of the three trials with more severe dysfunction noted in the trial by Durham (sCr ~2.2 mg/dL).  None of the three trials showed a significant difference between the treatment and control groups, indicating lack of preventive benefit.  The study by Durham varied the most in study design from the majority of other studies, making its validity questionable.  Of the four aforementioned trials, both studies by Oldemeyer and Durham are included in the following meta-analysis.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 9

 

A third meta-analysis by Pannu et al (40) was not as optimistic as the prior two.  This study included 15 RCTs with a total of 1776 test subjects, considerably more than Birck et al study.  This meta-analysis included several recently completed studies in which the effect of N-acetylcysteine was nonsignificant, including the largest study to date which included 397 patients (40).  All of the studies were randomized, but only six were double blind.  The Q statistic (Q=26.3, p=0.02) in this paper suggested significant heterogeneity across the studies.  A random effects analysis suggested the proportion of patients with DM, volume of contrast, baseline serum creatinine were not responsible for the heterogeneity.  (Of note, heterogeneity was noted in the aforementioned studies.) 

 

Sensitivity analyses showed that the protective effect of N-acetylcysteine was observed in one prespecified subgroup of trials: RCTs which completed peer review.  In a secondary analysis, the authors excluded studies which were not peer reviewed; N-acetylcysteine appeared to be more effective in this group as stated above.  With this, the authors also noted a publication bias, reflected by an absence of smaller published trials, which show no protective benefit of N-acetylcysteine.  Also, no study reported long-term effects including adverse renal outcomes or economic impact.

 

As demonstrated in Figure 8, a trend indicating benefit of N-acetylcysteine, although insignificant (CI 0.43, 1.00) is present.  However, if the studies by Fung and Goldenberg were included in this analysis, this trend would be further skewed to the right.  And if the study by Tepel were removed, again a shift to the right would be expected.

 

Conclusions

           

Radiocontrast-induced nephropathy is strongly associated with increased morbidity, mortality, substantial health care costs and prolonged hospital stay.  Several studies have identified factors associated with the development of radiocontrast nephropathy, including diabetes mellitus, high doses of contrast medium, volume depletion, coadministration of nephrotoxic medications, and preexisting chronic kidney disease.  Multiple studies addressing a wide range of prevention modalities have yielded at best variable results.  Preventative therapies have included a variety of intravenous fluids, osmotic and loop diuretics, dopamine and other vasodilators (CCBs), adenosine antagonists, agents with antioxidant properties such as N-acetylcysteine.  With the exception of N-acetylcysteine and various hydration regimens, most of the above have not been efficacious in reducing the risk of RIN. 

 

To date, successful strategies in preventing RIN have included prehydration and the use of low-osmolality contrast agents.  It is proposed iso-osmolal contrast may provide further attenuation, but this is unclear at this time.

 

A recent editorial in Lancet (20) suggested the use of N-acetylcysteine in all patients at risk for contrast nephropathy, which the author personally observed being practiced by many of his colleagues.  Much attention has been given to NAC as prophylactic agent for RIN since the study by Tepel et al (26) in 2000. In this study, however, the sample size was small enough that the significant difference depended entirely on the large effect size, with only 10 patients experiencing the outcome.  Variable and conflicting results have littered the pathway to clear benefit, making several authors cast doubt on the benefit of N-acetylcysteine, as individual studies and several meta-analyses have reached conflicting conclusions.  This obscures the otherwise clear picture painted in 2000.

 

Merten et al (38) looked at another viable option for prevention of RIN through the use of a bicarbonate infusion.  This study provided an appropriate power calculation with adequate patient enrollment.  It is also significant to note when the study monitor identified a large beneficial effect of bicarbonate, the authors patiently enrolled another 191 patients in a registry phase during which open-label bicarbonate-based fluids of only slightly different composition were administered, instead of halting the trial. During this phase, only 3 patients (1.7%) developed radiocontrast nephropathy, similar to the randomized trial. The authors demonstrated that the intervention yielded a measurable physiologic effect, an increase in urine pH significantly higher in the patients receiving bicarbonate.  Although the actual mechanism of benefit is unknown, the authors' proposal of a reduction in oxidative injury is reasonable and not inconsistent with a potential benefit afforded by other agents, such as N-acetylcysteine.

 

The use of bicarbonate-based solutions for volume expansion in patients with prerenal azotemia should be evaluated prospectively. This may also be of benefit in volume resuscitation in volume-contracted patients, as saline often worsen a metabolic acidosis.  In addition, increased attention should be paid to correction of metabolic acidosis among patients with acute renal failure receiving hemodialysis or hemofiltration if acidosis is found to impede functional recovery.

 

Taking the above into account, the following conclusions can be made.  The most important risk factor for the development RIN involves the presence of CRI, especially in diabetics.  These patients must be approached cautiously and treated aggressively prior to performing contrasted studies. 

 

All patients undergoing contrasted studies should be volume replete with IV fluids prior to any contrasted study.  As of the data to date, it would be prudent to hydrate with saline crystalloids (1 mL/kg/h for at least 12 hours before and after the procedure) or bicarbonate infusions using appropriate protocols.  To evaluate hydration status, physical exam is of utmost importance and an orthostatic evaluation should be included.  Preparation should also include discontinuation of nephrotoxic medications, especially NSAIDs and diuretics.  Hold metformin for risk of lactic acidosis for 48 hours and restart when sCr has stabilized after the 72 hour window.  Although recommended previously, the cessation of ACEi therapy may not be necessary.  It is prudent to delay nonemergent studies in the setting of sepsis, decompensated CHF or cirrhosis until the patient has stabilized.

 

Even though the data for N-acetylcysteine is controversial, two doses of 600 mg orally given the day of or before the procedure followed by a repeated two doses should be entertained.  This rule should apply to those with a baseline sCr >1.5 mg/dL, although this number is arbitrary and the patient’s clinical scenario should be assessed fully.  Although there are no conclusive studies indicating the use of N-acetylcysteine in diabetics with normal renal function, it may be prudent to do so if a large contrast volume is anticipated, the patient is elderly or cannot be volume resuscitated appropriately.  As elderly are at higher risk for RIN, N-acetylcysteine could be considered in this population, also, understanding there is not data to back this recommendation.

 

Although typically out of the internist’s hands, contrast volume limitation is important as is the use of nonionic, low-osmolal agents.  Limit contrasted studies to intervals no less than three days apart, preferably longer.

 

Expectant management involving maintenance of a normotensive, volume-replete status after the procedure is performed is imperative.

 

Bibliography

  1. Baker CS, Wragg A, Kumar S, De Palma R, Baker LR, Knight CJ. A rapid protocol for the prevention of contrast-induced renal dysfunction: The RAPPID Study. Journal of Amer Coll of Card 2003;47:2114-2118.

  2. Alonso A, Lau J, Jaber BL, Weintraub A, Sarnak M. Prevention of radiocontrast nephropathy with N-Acetylcysteine in patients with chronic kidney disease: a meta-analysis of randomized, controlled trials. Amer Journal of Kidney Dis 2004;43:1-9.

  3. Asif A, Epstein M. Prevention of radiocontrast-induced nephropathy.  Amer Journal of Kidney Dis 2004;44:12-24.

  4. Powe NR, Steinberg EP, Erickson JE, et al. Contrast medium-induced adverse reactions: economic outcome.  Radiology 1988;169:163-168.

  5. McCullough PA, Wolyn R, et al. Acute renal failure after coronary intervention: incidence, risk factors and relationship to mortality.  Amer Journ of Med 1997;55:254-257.

  6. Gruberg L, Mehran R, Dangas G, et al. Acute renal failure requiring dialysis after percutaneous coronary interventions.  Catheter Cardiovasc Inter 2001;52:409-416.

  7. Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure in mortality.  A cohort analysis. JAMA 1996;275:1488-1494.

  8. Solomon R. Contrast medium-induced acute renal failure. Kidney Int 1998;53:230-242.

  9. Heyman SN, Reichman J, Brezis M. Pathophysiology of radiocontrast nephropathy: a role for medullary hypoxia.  Invest Radiol 1999;34:685-691.

  10. Bakris GL, Burnett JC, Jr. A role of calcium in radiocontrast-induced reduction in renal hemodynamics.  Kidney Int 1985;27:465-468.

  11. Oldemeyer JB, Biddle WP, Wurdeman RL, Moss AN, Cichowski E, Hilleman DE. Acetylcysteine in the prevention of contrast-induced nephropathy after coronary angiography. Amer Heart Journ 2003;146:1-6.

  12. Parfrey PS, Griffiths SM, Barrett BJ, Paul MD, Genge M, Withers J, Farid N, Macmanamon PJ. Contrast material-induced renal failure in patients with diabetes mellitus, renal insufficiency, or both. A prospective, controlled study. NEJM 1989;320:143-149.

  13. Barret BJ, Parfrey PS, Vavasour HM et al. Contrast nephropathy in patients with impaired renal function: high versus low osmolar media. Kidney Int 1992;41:1274-1279.

  14. Goldenberg I, Shechter M, Matetzky S, Jonas M, Adam M, Pres H, Elian D, Agranat O, Schwammentahal E, Guetta V. Oral acetylcysteine as an adjunct to saline hydration for the prevention of contrast-induced nephropathy following coronary angiography. European Heart Journ 2004;25:212-218.

  15. Kefer JM, Hanet CE, Boitte S, Wilmotte L, De Kock M. Acetylcysteine, coronary procedure and prevention of contrast-induced worsening of renal function: which benefit for which patient? Acta Cardiol 2003;58:555-560.

  16. Diaz-Sandoval LJ, Kosowsky BD, Losordo DW. Acetylcysteine to prevent angiography-related renal tissue injury (The APART Trial). Amer Journal of Card 2002;89:356-358.

  17. Durham JD, Caputo C, Dokko J, Zaharakis T, Pahlavan M, Keltz J, Dutka P, Marzo K, Maesaka JK, Fishbane S. A randomized controlled trial of N-acetylcysteine to prevent contrast nephropathy in cardiac angiography. Kidney Int 2002;62:2201-2207.

  18. Brigouri C, Manganelli F, Scarpato P, et al. Acetylcysteine and contrast agent-associated nephrotoxicity. JACC 2002;40:298-303.

  19. Fung JWH, Szeto CC, Chan WMC, Kum LCC, Chan AKY, Wong JTH, Wu EB, Yip GWK, Chan JYS, Yu CM, Woo KS, Sanderson JE. Effect of N-acetylcysteine for prevention of contrast nephropathy in patients with moderate to severe renal insufficiency: a randomized trial. Amer Journal of Kidney Dis 2004;43:801-808.

  20. Kellum JA. A drug to prevent renal failure? Lancet 2003;362:589-590.

  21. Asif A, Preston RA, Roth D. Radiocontrast-induced nephropathy.  Amer Journal of Ther 2003;10:137-147.

  22. Rudnick MR, Goldfarb S, Wexler L, et al. Nephrotoxicity of ionic and nonionic contrast media in 1196 patients: a randomized trial. Kidney Int 1995;47:254-261.

  23. Singer I, Epstein M. In depth review: potential of dopamine A-1 agonist in the management of acute renal failure.  Amer Journal of Kidney Dis 1998;31:743-755.

  24. Wexler L, Cohen MB et al. Multicenter trial of ionic and nonionic contrast media: nephrotoxicity following cardiac angiography. Radiology 1982;181:292-299.

  25. Cigarroa RG, Lange RA, Williams RH, Hillis LD. Dosing of contrast material to prevent contrast nephropathy in patients with renal disease. Amer Journal Medicine 1989;86:649-652.

  26. Tepel M, van der Giet M, Schwartfel C, Laufer U, Liermann D, Zidek W. Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. NEJM 2000;343:180-184.

  27. Anto HR, Chous S-Y, Porush JC, et al. Infusion intravenous pyelography and renal function: effects of hypertonic mannitol in patients with chronic renal insufficiency. Arch Int Med 1981;141:1652-1656.

  28. Solomon R, Werner C, Mann D, et al. Effects of saline, mannitol and furosemide on acute decreases in renal function induced by radiocontrast agents. NEJM;1994:1416-1420.

  29. Huber W, Jeschke B, Page M, et al. Reduced incidence of radiocontrast-induced nephropathy in ICU patients under theophylline prophylaxis: a prospective comparison to series of patients at similar risk. Int Care Med 2001;27:1200-1209.

  30. Cox CD, Tsikouris JP. Preventing contrast nephropathy: what is the best strategy? A review of the literature. Therapeutic Review 2004;44:327-337.

  31. Gupta RK, Kapoor A, Tewari S, Sinha N, Sharma RK. Captopril for prevention of contrast-induced nephropathy in diabetic patients: a randomized study. Indian Heart Jour 1999;51:521-526.

  32. Wang A, Holsclaw T, Bashore TM, et al. Exacerbation of radiocontrast nephrotoxicity by endothelin receptor antagonism.  Kidney Int 2000;57:1675-1680.

  33. Russo D, Testa A, Della Volpe L, Sansone G. Randomized prospective study on renal effects of two different contrast media in humans: protective role of a calcium channel blocker. Nephron 1990;55:254-257.

  34. Spangberg-Viklund B, Berglund J, Nidonoff T, Nyberg P, Skau T, Larsson R. Does prophylactic treatment with felodopine, a calcium antagonist, prevent low-osmolar contrast induced renal dysfunction in hydrated diabetic and nondiabetic patients with normal or moderately reduced renal function? Scand Jour Urol Nephrol 1996;30:63-68.

  35. Esnault VL. Radiocontrast media-induced nephrotoxicity in patients with renal failure: rationale for a new double-blind, prospective, randomized trial testing calcium channel antagonists. Nephrol Dial Transplant. 2002;17:1362-1364.

  36. MacNeill BD, Harding SA, Bazari H, Patton KK, Colon-Hernadez P, deJoseph D, Jang I. Prophylaxis of contrast-induced nephropathy in patients undergoing coronary angiography. Cath and Cardiovasc Int 2003;60:458-461.

  37. Mueller C, Buerkle G, Buettner HJ, Petersen J, Perruchoud AP, Eriksson U, Marsch S, Roskamm H. Prevention of contrast media-associated nephropathy: randomized comparison of two hydration regimens in 1620 patients undergoing coronary angioplasty. Arch Int Med 2002;162:329-336.

  38. Merten GJ, Burgess WP, Gray LV, Holleman JH, Roush TS, Kowalchuk GJ, Bersin RM, Moore AV et al. Prevention of contrast-induced nephropathy with sodium bicarbonate. JAMA 2004;291:2328-2334.

  39. Birck R, Krzossok S, Markowetz F, Schnulle P, van der Woude FJ, Braun C. Acetylcysteine for prevention of contrast nephropathy: meta-analysis. Lancet 2003;362:562-603.

  40. Pannu N, Manns B, Lee H, Tonelli M. Systematic review of the impact of N-acetylcysteine on contrast nephropathy. Kidney Int 2004;65:1366-1374.