Contrast Induced Nephropathy - Cleveland Clinic Journal of Medicine-2006-Rudnick-75-80

CLEVELAND CL IN IC JOURNAL OF MEDICINE VOLUME 73 • NUMBER 1 JANUARY 2006 75

MICHAEL R. RUDNICK, MD*

Associate Professor of Medicine, University ofPennsylvania School of Medicine, Philadelphia

AARON KESSELHEIM, MDBrigham and Women’s Hospital, Boston

STANLEY GOLDFARB, MD†

Professor of Medicine, University ofPennsylvania School of Medicine, Philadelphia

Contrast-induced nephropathy:How it develops, how to prevent it

REVIEW

■ ABSTRACT

No current treatment can reverse or ameliorate contrast-induced nephropathy once it occurs, but prophylaxis ispossible. Many preventive measures have failed to showbenefits in well-designed, prospective, randomized,double-blinded trials. This review will focus only on theprophylactic strategies that have possible or proven value.

■ KEY POINTS

The risk of contrast-induced nephropathy is directlyproportional to the severity of preexisting renal insufficiency.

Hydration with normal saline solution is the most widelyaccepted preventive intervention.

N-acetylcysteine may be effective, but studies have givenconflicting results.

Sodium bicarbonate may be of value, but larger multicenterstudies are needed to determine its true effectiveness.

Newer contrast agents that are nonionic and of lowerosmolality than older agents are less nephrotoxic but canstill cause nephropathy.

Due to the logistical effort and high cost associated withhemofiltration, larger randomized trials should beperformed before this technique can be recommended asstandard prophylaxis against contrast-inducednephropathy in high-risk patients.

Theophylline cannot yet be recommended as standardprophylaxis against contrast-induced nephropathy.

UCH REMAINS TO BE DETERMINED aboutcontrast-induced nephropathy, ie, the

acute renal failure that sometimes developsafter giving iodinated radiocontrast agents.For example:• What causes it? The short answer seems tobe renal ischemia, but via what pathways? Arecontrast agents directly nephrotoxic?• How can it be prevented, short of notusing contrast? Many agents that looked goodin theory have proved useless. Hydrationseems to be a good principle, but what is thebest prescription? Must it be intravenous, orwill oral hydration suffice? Is sodium bicar-bonate better than sodium chloride as anintravenous hydration solution?• Is the latest iso-osmolar agent better thanthe low-osmolar agents currently in use?

This review examines the multipledimensions of contrast-induced nephropathy.We will discuss the evidence for using variousstrategies for prophylaxis—hydration, N-acetylcysteine, sodium bicarbonate, the-ophylline, and hemofiltration—and then giveour recommendations.

■ STILL COMMON

Contrast-induced nephropathy continues tobe a common form of hospital-acquired acuterenal failure.1 Although its incidence is lowin patients with normal renal function, it canbe much higher in those with severe renalinsufficiency at baseline.

M

*Dr. Rudnick has indicated that he has received grant or research supportfrom, serves as a consultant for, and is on the speakers’ bureau of the GEHealthcare corporation.†Dr. Goldfarb has indicated that he is on the speakers’ bureau of the GEHealthcare corporation.

on January 28, 2014. For personal use only. All other uses require permission.www.ccjm.orgDownloaded from

http://www.ccjm.org/


76 CLEVELAND CL IN IC JOURNAL OF MEDICINE VOLUME 73 • NUMBER 1 JANUARY 2006

Moreover, an enormous number ofpatients receive contrast agents. For example,in 2000, approximately 1,318,000 diagnosticcardiac catheterizations and 561,000 percuta-neous transluminal coronary angioplasty pro-cedures were performed, which are just two ofthe many procedures in which contrast isused.2

■ TYPES OF CONTRAST MEDIA

The earliest contrast agents were ionic, con-taining a sodium atom that dissociated fromthe molecule in aqueous solution. Each mole-cule of the agent carried three iodine atoms.Therefore, these agents required two osmoti-cally active particles to deliver three iodineatoms, and they had extremely high osmolali-ties (about 2,000 mOsm/L). These agents,termed high-osmolar or ionic, were the predom-inant ones used until the 1980s (FIGURE 1).

The next generation, introduced in the1980s and still the predominant contrastmedia in use, are nonionic.3 Since they there-fore need only one osmotically active particleto deliver three iodine atoms, their osmolality

is only about 600 to 900 mOsm/L, and theyare termed low-osmolar.

Both types of agents are monomers, withone benzene ring and three iodine atoms.Dimer molecules consisting of two joined ben-zene rings contain a total of six iodine atomsper molecule. There is one ionic dimer,ioxaglate, which has a 6:2 or 3:1 ratio ofiodine atoms to osmotically active particlesand has an osmolality of 600 mOsm/L, similarto other low-osmolar contrast agents.

The newest contrast agent, iodixanol, is anonionic dimer. The chemical structure of thisagent allows six iodine atoms to be attached toone osmotically active particle, resulting in anosmolality of 300 mOsm/L, which is iso-osmo-lar with normal plasma.

■ DOES CONTRAST NEPHROPATHYINCREASE MORTALITY?

Patients undergoing percutaneous coronaryinterventions have a higher mortality rate ifnephropathy develops.4,5 The risk of dying isgreatest in patients who require dialytic sup-port because of the nephropathy. For example,

The newestcontrast agentis a nonionicdimer and isiso-osmolar

CONTRAST-INDUCED NEPHROPATHY RUDNICK AND COLLEAGUES

Contrast agentsMONOMERS DIMERS

FIGURE 1

IonicI I I I I I

I I I I I I

I I

I I

I

I

- - -R- - -

- - -R- - -

COO–Na+ COO–Na+

Iothalamate*

Diatrizoate*

Metrizoate*

Ioxaglate†

Iohexol†Ioversol†Ioparnidol†

Iodixanol‡

Nonionic

*High-osmolar agents†Low-osmolar agents‡Iso-osmolar agent

ADAPTED FROM RUDNICK MR. THE ROLE OF OSMOLALITY IN CONTRAST-ASSOCIATED NEPHROTOXICITY. APPLICATIONS IN IMAGING—CARDIAC INTERVENTIONS. SCOTCH PLAINS, NY: ANDERSON PUBLISHING, 2003.




McCullough et al5 found that in-hospitalmortality rates were 1.1% for patients with nocontrast-induced nephropathy compared with7.1% for those with nephropathy alone, andup to 35.7% for those with nephropathyrequiring dialysis.

In this and other studies, patients inwhom nephropathy developed had a higherprevalence of preexisting conditions andperiprocedural complications than those inwhom it did not develop.

The comorbidities complicate the analysis,as one cannot determine with certaintywhether contrast-induced nephropathy con-tributes directly to mortality in this population,whether this complication simply selects out asubgroup of patients at significantly greater riskof dying, or if both possibilities are correct.Multivariate regression analyses demonstratedthat contrast-induced nephropathy was anindependent predictor of death, but this type ofanalysis does not prove a cause-and-effect rela-tionship.

The question of whether contrast-induced nephropathy directly contributes tomortality is further confounded by recentstudies demonstrating an increased risk ofdeath in cardiac patients with preexistingrenal insufficiency undergoing coronaryrevascularization.6,7 Since most patientswho develop contrast-induced nephropathyhave preexisting renal insufficiency, the spe-cific contribution of contrast-inducednephropathy alone to increased mortality isunclear.

■ FEW PATIENTS NEED DIALYSIS

In most cases of contrast-induced nephropa-thy, serum creatinine begins to rise within 24to 48 hours after exposure, reaches a peakwithin 3 to 5 days, and then returns to base-line levels within 7 to 10 days.8 In more severecases, the creatinine concentration may notpeak until 5 to 10 days, and the increase maybe associated with oliguria.8,9

Fortunately, few patients need acutehemodialysis. Diaabetic patients who takeinsulin and have advanced renal insufficiencyare more susceptible to prolonged acute renalfailure, often with oliguria or the need forhemodialysis.

Findings on urinalysis in patients withcontrast-induced nephropathy are similar tothose in patients with other causes of acutetubular necrosis. Typical findings are coarsegranular casts, renal tubular epithelial cells,and amorphous debris.

■ RISK FACTORS

Preexisting renal insufficiency is the sin-gle greatest risk factor.8,9 In one comprehen-sive review, an estimated 60% of patients withcontrast-induced nephropathy had preexist-ing renal insufficiency.9

The more severe the baseline renal insuf-ficiency, the greater the risk.8,9 Although therisk of contrast-induced nephropathy for agiven serum creatinine value can vary widely,one can roughly estimate the percent risk bymultiplying the serum creatinine concentra-tion in milligrams per deciliter by 10.

Diabetes mellitus is often cited as a riskfactor for contrast-induced nephropathy,9,10

but the risk ascribed to it is probably due tocoexisting renal insufficiency, usually diabeticnephropathy, rather than to the diabetes perse.9,10 In recent prospective studies, the inci-dence in patients with diabetes and normalrenal function was similar to that in nondia-betic patients with normal renal function.10,11

On the other hand, patients with diabetesand preexisting renal insufficiency have agreater risk for contrast-induced nephropathythan nondiabetic patients with similar levels ofpreexisting renal insufficiency.10,11 Moreover,when patients in this high-risk group developnephropathy, they more often develop oliguriaand need dialysis.12 As with patients withoutdiabetes, the risk of contrast-induced nephrop-athy is directly proportional to the severity ofpreexisting renal insufficiency.

Volume of contrast media. Some studiesfound a correlation between the volume ofcontrast given and the risk of nephropa-thy,11–13 whereas other studies did not.14

Cigarroa et al13 used a predetermined for-mula based on body weight and baseline renalfunction to limit the volume of contrast mediain patients undergoing coronary angiography.The limit was 5 mL of contrast per kg of bodyweight up to a maximum of 300 mL, dividedby the serum creatinine concentration in mil-


To calculatethe risk ofcontrastnephropathy,multiply thecreatinine levelby 10





ligrams per deciliter. Nephropathy developedin 21% of the patients in whom the total vol-ume of contrast exceeded the formula amount,compared with 2% (P < .001) of patients inwhom the contrast volume fell within the pre-scribed limit.

Multiple myeloma has traditionally beenconsidered a risk factor for contrast-inducednephropathy.9,10 However, McCarthy andBecker15 reviewed several retrospective stud-ies of contrast use in patients with myelomaand found an incidence of nephropathy ofonly 0.6% to 1.25%, indicating that this groupis not at increased risk with modern contrastagents, provided that volume expansion isachieved at the time of exposure.

Even though multiple myeloma shouldnot be an absolute contraindication for con-trast use, clinical prudence warrants perform-ing radiologic studies with contrast only ifnecessary and avoiding dehydration in thesepatients.

■ HOW DO CONTRAST AGENTSCAUSE NEPHROPATHY?

The primary pathways by which contrastagents cause nephropathy are by renalischemia (by reducing blood flow or increasingoxygen demand) and, possibly, by direct toxic-ity to tubular epithelial cells.

Renal ischemiaAfter contrast is injected, renal blood flowtransiently increases and then decreases over alonger time, suggesting that renal ischemia is amajor factor in the pathogenesis of contrast-induced nephropathy.16

In experimental studies of contrast-inducednephropathy, the kidneys show pathologicischemic changes—necrosis of the medullarythick ascending limbs as well as tubular collapseand casts—primarily in the outer medullaryarea of the kidney.17 Moreover, contrast agentscause a marked decrease in medullary oxygena-tion that can be directly measured with oxygenmicroelectrodes.18

Based on these observations, the followingmechanism for acute renal failure induced bycontrast agents has been proposed.19,20

Even under normal conditions, the renalmedulla is poorly oxygenated, making it par-

ticularly susceptible to hypoxic injury. Theoxygen tension in the medulla is 10 to 20 mmHg compared with 50 mm Hg in the cortex.Reasons for the low oxygen tension are coun-tercurrent exchange of oxygen between thevasa recta and oxygen use by active transportof sodium in the ascending limb of the loop ofHenle.19

Contrast agents reduce the oxygen ten-sion in both the cortex and the medulla.18

This effect may be due to increased work ofactive transport in response to an osmoticdiuresis from hyperosmolar agents, as well asthe release of vasoconstrictive compoundssuch as endothelin (see below). Furthermore,blockade of vasodilatory compounds such asnitrous oxide and prostaglandins appears tomarkedly exacerbate contrast-induced medul-lary hypoxic injury.19

VasoconstrictionMany substances may mediate renal vasocon-striction and subsequent hypoxic injury. Ofnote, adrenergic stimulation and activation ofthe renin-angiotensin system do not seem tobe involved in contrast-induced vasoconstric-tion.16,17 Prostaglandins with vasodilatoryproperties may counter the vasoconstrictioninduced by contrast media, since pretreatmentwith indomethacin is necessary to induceexperimental contrast-induced renal injury.18

Endothelin. Multiple experimental obser-vations suggest that endothelin, a potent renalvasoconstrictor, may play a critical role incontrast-mediated vasoconstriction.8

These observations led to a clinical trial inwhich patients with chronic kidney diseaseundergoing cardiac angiography were random-ized to receive either the endothelin receptorantagonist SB 290670 or placebo.21 Surprisingly,the incidence of contrast-induced nephropathywas higher in the treatment group (56%) than inthe placebo group (29%; P = .002).

Adenosine. The role of adenosine in thepathogenesis of contrast-induced nephropathyis described in detail in an excellent review byPflueger et al.22 Adenosine causes vasodilata-tion through A2 stimulation of the efferentarteriole and medullary capillaries, and it alsocauses vasoconstriction through A1 stimula-tion of the afferent arterioles. However, renalvasoconstriction dominates, explaining why

Even undernormalconditions,the renalmedulla ispoorlyoxygenated





intrarenal adenosine infusion results in adecrease in renal blood flow.22

In experimental studies, theophylline, anonselective adenosine receptor antagonist,inhibited contrast-media induced renal vaso-constriction.22

Role of osmolalitySeveral clinical and experimental observa-tions suggest that the hyperosmolality of con-trast media may play a role in the pathogene-sis of contrast-induced nephropathy. Clinicalstudies demonstrated that low-osmolar con-trast agents cause less nephrotoxicity thanhigh osmolar agents.11,23 Furthermore, in onestudy,24 the incidence of contrast-inducednephropathy was lower with an iso-osmolarcontrast agent than with a low-osmolar agent.

In experimental studies, hypertonic solu-tions of saline or mannitol reduce theglomerular filtration rate and renal blood flowand increase enzymuria similarly to high-osmolar contrast media but with a lesser mag-nitude.16,25 A theory to account for thesenonspecific adverse effects is that hyperosmo-lality activates tubuloglomerular feedback orcauses an increase in tubular hydrostatic pres-sures, either of which could lead to a decreasein glomerular filtration. In addition, theosmotic diuresis produced by contrast mediamay result in increased active transport ofsodium in the thick ascending limb and alsoin vasoconstriction, and both of these couldlead to worsened medullary hypoxemia.18–20

On the other hand, most studies in ani-mals specifically comparing iso-osmolar con-trast agents (iodixanol and iotrolan) withhigh-osmolar and low-osmolar contrast agentshave not demonstrated any lower rate of renalabnormalities with the iso-osmolar agents.26,27

The reason may be that the iso-osmolar agentsare more viscous, which could increase redblood cell aggregation and decrease renalblood flow, offsetting any reduction inmedullary hypoxemia from their lower osmo-lality.

Reactive oxygen speciesReactive oxygen species formed as a result ofpostischemic oxidative stress can lead to acuterenal failure through their direct effects onrenal endothelial cells, which include apop-

totic cell death. Adenosine’s role in thepathogenesis of contrast-induced nephropa-thy may be due to this molecule’s ability toincrease generation of oxygen-free radicals.28

The possible benefit of N-acetylcysteine andsodium bicarbonate in preventing contrast-induced nephropathy (see below) is hypothe-sized to be due to the ability of these com-pounds to mitigate oxidative injury.

Direct cellular toxicityA number of experimental observations sug-gest that contrast agents are directly toxic tokidney cells, causing proximal cell vacuoliza-tion, interstitial inflammation, cellular necro-sis, and enzymuria.8,17 Furthermore, suspen-sions of proximal tubular segments exposed tocontrast media showed abnormalities in sever-al markers of cellular injury, that were poten-tiated by hypoxia and were more pronouncedwith high-osmolar agents than with low-osmolar agents.29

■ PREVENTIVE MEASURES:MANY TRIED, FEW SUCCEEDED

Currently, there is no treatment to reverse orameliorate contrast-induced nephropathyonce it occurs, but prophylaxis is possible.

Many preventive measures have beentried that may interfere with one or more ofthe currently accepted pathogenetic mecha-nisms for contrast-induced nephropathy(TABLE 1). However, many of these measureslater failed to show benefits in well-designed,prospective, randomized, double-blinded tri-als. Among this group are diuretics,30 manni-tol,30 dopamine,31 atrial natriuretic pep-tide,32 endothelin receptor antagonists,21

and fenoldopam.33

This review will focus only on the strate-gies that have possible or proven prophylacticvalue.

Hydration is indicated,but what kind, how much?Hydration is the primary intervention for pre-venting contrast nephropathy.34

The theoretical rationale for hydration isthat it should decrease the activity of therenin-angiotensin system, reduce the levels ofother vasoconstrictive hormones such as


More contrastnephropathyoccurred withsaline plusmannitol orfurosemidethan withsaline alone





endothelin, increase sodium diuresis, decreasetubuloglomerular feedback, prevent tubularobstruction, protect against reactive oxygenspecies, and dilute the contrast media in thetubule, thus decreasing any direct nephrotox-ic effect of the contrast agent on the tubularepithelium.34

Several studies in animals demonstratedhydration with saline infusions to be beneficialin preventing contrast-induced nephropathy.35

Early clinical studies used historical con-trols for comparison and also suggested thathydration is beneficial. Subsequently, intra-venous hydration became the standardmethod to prevent contrast-induced neph-ropathy.8,9

There have been a few prospective ran-domized studies comparing saline alone vsother therapies as prophylactic strate-gies.21,30–33

Solomon et al30 randomized patients withchronic kidney disease undergoing cardiacangiography to receive either saline alone,

saline and mannitol, or saline and furosemide.All three groups received 0.45% saline intra-venously at 1 mL/kg/hour for 12 hours beforeand 12 hours after receiving contrast. Neph-ropathy occurred in 11% of patients receivingsaline alone vs 28% who received saline andmannitol and 40% who received saline andfurosemide.

Different regimens of saline hydrationhave been used, but no one regimen hasdemonstrated clear superiority.

Trivedi et al36 prospectively randomizedpatients undergoing cardiac angiography toreceive either intravenous saline for 12 hoursboth before and after catheterization or oralfluids only, taken as desired. Contrast-inducednephropathy occurred in 3.7% of those whoreceived intravenous saline vs 34.6% of thosewho received only oral fluids.

In contrast, the Preparation for Angiog-raphy in Renal Dysfunction (PREPARED)trial37 showed that, in patients with chronickidney disease undergoing coronary angiogra-phy, hydration on an outpatient basis beforecatheterization, coupled with a brief period ofintravenous hydration, was equivalent toovernight intravenous hydration.

Bader et al38 randomized patients under-going computed tomography or digital angiog-raphy to receive either 2,000 mL of intra-venous fluid over 24 hours (12 hours beforeand 12 hours after contrast) or 300 mL ofintravenous fluid during the radiologic proce-dure. The glomerular filtration rate fell by 18.3mL/minute in the continuous infusion groupcompared with a 34.6 mL/minute fall in thebolus infusion group (P < .05), suggesting thatslow hydration is superior to bolus expansionduring the procedure.

The Prevention of Radiocontrast InducedNephropathy Clinical Evaluation (PRINCE)study39 tested the hypothesis that forceddiuresis with maintenance of intravascularvolume would result in less contrast-inducedrenal injury. Although no difference in theincidence of contrast-induced nephropathywas observed between patients who under-went forced diuresis and those who did not,the incidence in participants with urine flowrates greater than 150 mL/hour was 21.6% vs45.9% in those with lower urine flow rates (P= .03).

Strategies for preventingcontrast-induced nephropathy

Strategies that do not workMannitolFurosemideDopamineAtrial natriuretic factorFenoldopamHemodialysis

Strategies that may workCalcium channel blockersTheophyllineIso-osmolar contrast mediaN-acetylcysteineHemofiltrationSodium bicarbonateAscorbic acidProstaglandins

Currently recommended strategiesEmploy noniodinated contrast studiesAvoid nonsteroidal anti-inflammatory drugsProvide adequate time between contrast proceduresMinimize contrast volumeParenteral hydrationLow-osmolar or iso-osmolar contrast media

T A B L E 1





Mueller et al40 compared the use of iso-tonic (0.9%) saline (n = 685) vs half-isotonic(0.45%) saline (n = 698) in patients undergo-ing coronary angioplasty. Both groupsreceived about 2,000 mL of intravenous fluid.The incidence of contrast-induced nephropa-thy was significantly lower with isotonicsaline (0.7%) than with half-isotonic saline(2%, P = .04).

Comment. These experimental and clini-cal studies support the use of intravenoushydration to prevent contrast-inducednephropathy, especially in patients withazotemia at high risk. As yet, no sufficientlypowered, controlled, prospective trials haveexamined the minimally effective length oftime, optimal rate, and fluid composition ofintravenous hydration required before andafter contrast administration in high-riskazotemic patients.

N-acetylcysteine: Data are conflictingN-acetylcysteine has been shown in experi-ments in animals to ameliorate renal injuriesfrom ischemia and nephrotoxins.41 Potentialmechanisms include antioxidation (eitherdirectly as a free radical oxygen scavenger orindirectly through glutathione production),41

preventing apoptotic cell death mediated bythe generation of oxygen free radicals, andvasodilation.42

Tepel et al43 first found N-acetylcysteinebeneficial in a study of 83 patients withchronic renal failure (32.5% had diabeticnephropathy) undergoing computed tomogra-phy with contrast. Patients received intra-venous saline for 12 hours before and afterreceiving the contrast and were prospectivelyrandomized to receive either N-acetylcysteine600 mg by mouth twice daily 1 day before andon the day of the study (total of four dosesover 2 days) or placebo. Contrast-inducednephropathy occurred in 2% of the N-acetyl-cysteine group vs 21% of the placebo group (P= .01; relative risk 0.1).

Several subsequent studies confirmed thevalue of N-acetylcysteine in preventing con-trast-induced nephropathy, all in patientsundergoing cardiac angiography.41 Based onthese data, N-acetylcysteine became widelyaccepted as a prophylactic therapy.

Because N-acetylcysteine undergoes sig-

nificant first-pass metabolism and its oraladministration poses logistical problems, Bakeret al44 evaluated its intravenous use. acetylcys-teine was given intravenously at a dose of 150mg/kg in 500 mL of normal saline solutionover 30 minutes immediately before contrastexposure and then 50 mg/kg in 500 mL of nor-mal saline solution over the next 4 hours.Contrast-induced nephropathy occurred in5% of the N-acetylcysteine group comparedwith 21% of the saline-alone group (relativerisk 0.28, P = .045). These results suggest thatprolonged use of N-acetylcysteine before con-trast exposure may not be necessary.

On the other hand, many other studiesdid not demonstrate a prophylactic value forN-acetylcysteine.41 For example, Durham etal45 studied 79 patients with chronic kidneydisease who underwent diagnostic cardiaccatheterization, percutaneous coronary inter-vention, or both. The patients were randomlyassigned to receive oral acetylcysteine orplacebo. All patients received hydration with0.45% saline for up to 12 hours before andafter catheterization. There was no significantdifference in the incidence of contrast-induced nephropathy between the two groups:26.3% in the acetylcysteine group and 22% inthe control group.

Nallamothu et al46 performed a meta-analysis of 20 studies involving 2,195 patientsand calculated that the relative risk of con-trast nephropathy in patients who received N-acetylcysteine was 0.73 (95% confidenceinterval 0.52–1.0; P = .08). Pannu et al47 per-formed another meta-analysis of 15 studiesinvolving 1,776 patients and calculated therelative risk at 0.65 (95% confidence interval0.43–1.00). Both groups of investigators werecautious in their conclusions, pointing outthat the individual studies showed substantialheterogeneity in design and results and callingfor definitive studies. Some of the variablesthat differed among the studies published todate include the severity of baseline renalinsufficiency, the percentage of diabeticpatients, the type and amount of contrastused, the amount and timing of N-acetylcys-teine administration, and the amount ofhydration.

Conclusions. Although N-acetylcysteineis safe, easy to use, and inexpensive, its value


Studies ofN-acetylcys-teine differedin design andresults





in preventing contrast-induced nephropathyremains controversial.

Theophylline:Not recommended at this timeSeveral reports suggest that theophylline, anadenosine antagonist, prevents contrast-induced nephropathy.22,48

Erley et al48 randomized 39 patients whoreceived contrast media to receive either intra-venous theophylline or placebo. Although nopatient in either group developed contrast-induced nephropathy, the glomerular filtrationrate decreased in the placebo group from 88mL/minute at baseline to 75 mL/minute 4hours after contrast administration; it remainedunchanged in the theophylline group.

In several other placebo-controlled stud-ies, theophylline (given orally or intravenous-ly) prevented contrast-induced falls in creati-nine clearance, but all the studies were in low-risk patients, and contrast-induced nephropa-thy was not seen in any groups.

Theophylline has potential risks, includ-ing ventricular arrhythmias, seizures, andshock—all of which may be potentiated by avariety of other drugs.22

Conclusions. The data regarding theo-phylline are mixed. Favorable studies werelimited by small numbers, absence of high-riskpatients, and a failure to demonstrate differ-ences in the incidence of contrast-inducednephropathy. Therefore, theophylline cannotbe recommended as standard prophylaxisagainst contrast-induced nephropathy at thistime.

Low-osmolar agentsare better than high osmolar agentsIntroduced in the 1980s, nonionic low-osmo-lar contrast agents have replaced ionic high-osmolar agents as the standard intravascularcontrast media because of their lower inci-dence of adverse effects.3

Studies in animals have demonstratedthat, compared with high-osmolar agents,low-osmolar agents result in less nephrotoxic-ity but still can cause nephropathy.8 Initialclinical studies comparing high-osmolar vslow-osmolar contrast agents failed to demon-strate a difference between these two types ofagents but were underpowered in respect to

high-risk azotemic patients.8In 1995, Rudnick et al11 performed a

prospective, randomized, double-blind studycomparing the high-osmolar contrast agent dia-trizoate with the low-osmolar contrast agentiohexol in 1,196 patients, including 509azotemic patients, of whom 213 had diabetes.In patients without azotemia, the incidence ofcontrast-induced nephropathy was negligiblewith either agent, regardless of whether or notthe patient had diabetes. However, in patientswith azotemia but without diabetes, the inci-dence of contrast-induced nephropathy was 7%with the high-osmolar agent vs 4% with thelow-osmolar agent. In patients with azotemiaand diabetes, the differences were even morepronounced: 27% with the high- osmolar agentand 12% with the low-osmolar agent.

A subsequent meta-analysis indicated thatlow-osmolar agents reduced the incidence ofcontrast-induced nephropathy by 50%.23

Are iso-osmolar agentsbetter than low-osmolar agents?The iso-osmolar contrast agents have under-gone experimental and clinical studies com-paring their nephrotoxicity with that of thecurrently popular low-osmolar agents. As dis-cussed above, these third-generation agentshave not demonstrated less nephrotoxicitythan low-osmolar agents in studies in animals.Furthermore, only a few clinical studies havecompared the incidence of contrast-inducednephropathy with the two types of agents.24,49

Aspelin et al24 performed a prospective,randomized, double-blind, multicenter trial in129 patients with azotemia and diabetes.Patients were randomly assigned to receiveeither iohexol (a low-osmolar agent) or iodix-anol (an iso-osmolar agent). The incidence ofcontrast-induced nephropathy was 3% withthe iso-osmolar agent vs 26% with the low-osmolar agent (odds ratio 0.09, P = .002).

Larger randomized trials will be needed toverify these encouraging results, especiallywith comparisons to other low-osmolar con-trast agents.

Hemodialysis and hemofiltrationNumerous studies have demonstrated that 2 to3 hours of hemodialysis effectively removes60% to 90% of contrast medium.50 Several


Low-osmolaragents havebecome thestandard




studies explored the prophylactic value ofhemodialysis in high-risk patients, but mostfailed to demonstrate a reduced incidence ofcontrast-induced nephropathy.50

On the other hand, Marenzi et al51 recent-ly found that hemofiltration significantlyreduced contrast-induced nephropathy inpatients at high risk. In this study, patientswith chronic kidney disease undergoing coro-nary angiography were randomized to undergoeither hemofiltration in an intensive care unitor parenteral saline hydration. Hemofiltrationwas started 4 to 6 hours before contrast admin-istration, stopped for coronary angiography,then resumed for an additional 18 to 24 hours.Isotonic saline was used as replacement fluidand was given at a rate of 1 L per hour, whichmatched the ultrafiltration rate so that no netfluid loss resulted. In the control group, iso-tonic saline was given at 1 mL/kg/hour for 6 to8 hours before and 24 hours after angiography.

The incidence of contrast-inducednephropathy was 5% in the hemofiltrationgroup compared with 50% in the controlgroup (P < .001). The in-hospital mortalityrate was 2% in the hemofiltration group com-pared with 14% in the control group (P = .02).

Despite these impressive results, the con-clusions of this study should be viewed withsome caution. Removal of creatinine byhemofiltration per se could result in a lowerincidence of contrast-induced nephropathy,although this alone would not account for dif-ferences in mortality. Moreover, the mortalityrate in the control group was inordinately high,suggesting that it was not a good representativecohort. Both groups received an extraordinaryvolume of contrast (approximately 250 mL) forpatients with moderately severe chronic kidneydisease (their baseline mean creatinine con-centration was 3.0 mg/dL).

Conclusions. Given these reservations,due to the logistical effort and high cost asso-ciated with hemofiltration, larger randomizedtrials should be performed before this tech-nique can be recommended as standard pro-phylaxis against contrast-induced nephropa-thy in high-risk patients.

Somewhat related is the not-infrequentclinical question of when to perform the nexthemodialysis treatment in a patient undergo-ing chronic hemodialysis who receives

intravascular contrast media. Although thisquestion has not been extensively investigat-ed in clinical trials, there is evidence thatmost patients can safely wait 24 to 36 hoursafter contrast exposure until their nexthemodialysis treatment.

Sodium bicarbonate: Data are preliminaryMerten et al,52 in a randomized controlledtrial at a single center, compared hydrationwith sodium bicarbonate vs sodium chlorideto prevent contrast-induced nephropathy inazotemic patients receiving low-osmolarcontrast agents. Both infusions contained154 mEq of either sodium chloride or sodiumbicarbonate in 1 L of 5% dextrose and water.A close approximation of the sodium bicar-bonate solution can be achieved by adding 3ampules (150 mEq) of sodium bicarbonate to1 L of 5% dextrose in water: the final sodiumbicarbonate concentration is 130 mEq/L.The infusion rate for either fluid was 3mL/kg/hour for 1 hour before contrastadministration, followed by 1 mL/kg/hourduring contrast administration and then for6 hours afterward.

Contrast-induced nephropathy occurredin 1.7% of patients who received sodiumbicarbonate compared with 13.6% of patientswho received sodium chloride (P = .02).

The benefit of sodium bicarbonate in pre-venting contrast-induced nephropathy isprobably not simply due to volume expansion,which was similar between treatment groups.The authors postulate instead that sodiumbicarbonate may reduce the formation of oxy-gen free radicals (a pH-dependent reaction),previously reported to play a pathogenetic rolein contrast-induced nephropathy.52

Conclusions. We agree with the authorsthat sodium bicarbonate infusion may provide asimple, safe, and inexpensive method to preventcontrast-induced nephropathy but the results ofthis study need to be confirmed in a larger, mul-ticenter, prospective randomized trial.

■ CURRENT RECOMMENDATIONS

The use of intravascular contrast in patients atrisk for contrast-induced nephropathy should beconsidered only when alternative imaging teststhat do not use iodinated contrast cannot pro-


Start salinehydration 2–4hours beforethe procedureand continue4–6 hours after




■ REFERENCES1. Hou SH, Bushinsky DA, Wish JB, Cohen JJ, Harrington JT. Hospital-

acquired renal insufficiency: a prospective study. Am J Med 1983;74:243–248.

2. American Heart Association. Heart Disease and Stroke Statistics—2003 Update, pp 38–39.

3. McClennan BL. Low-osmolality contrast media: premises and promis-es. Radiology 1987; 162:1–8.

4. Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure onmortality. A cohort analysis. JAMA 1996; 275:1489–1494.

5. McCullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. Acuterenal failure after coronary intervention: incidence, risk factors, andrelationship to mortality. Am J Med 1997; 103:368–375.

6. Szczech LA, Reddan DN, Owen WF Jr, et al. Differential survival aftercoronary revascularization procedures among patients with renalinsufficiency. Kidney Int 2001; 60:292–299.

7. Reinecke H, Trey T, Matzkies F, Fobker M, Breithardt G, Schaefer RM.Grade of chronic renal failure, and acute and long-term outcomeafter percutaneous coronary interventions. Kidney Int 2003;63:696–701.

8. Rudnick MR, Berns JS, Cohen RM, Goldfarb S. Contrast media-associ-ated nephrotoxicity. Semin Nephrol 1997; 17:15–26.

9. Berns AS. Nephrotoxicity of contrast media. Kidney Int 1989;36:730–740.

10. Rudnick MR, Berns JS, Cohen RM, Goldfarb S. Nephrotoxic risks ofrenal angiography: contrast media-associated nephrotoxicity andatheroembolism—a critical review. Am J Kidney Dis 1994;24:713–727.

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

12. Manske CL, Sprafka JM, Strony JT, Wang Y. Contrast nephropathy inazotemic diabetic patients undergoing coronary angiography. Am JMed 1990; 89:615–620.

13. Cigarroa RG, Lange RA, Williams RH, Hillis LD. Dosing of contrastmaterial to prevent contrast nephropathy in patients with renal dis-ease. Am J Med 1989; 86(6 Pt 1):649–652.

14. Barrett BJ, Parfrey PS, Vavasour HM, et al. Contrast nephropathy inpatients with impaired renal function: high versus low osmolarmedia. Kidney Int 1992; 41:1274–1279.

15. McCarthy CS, Becker JA. Multiple myeloma and contrast media.Radiology 1992; 183:519–522.

16. Katzberg RW, Morris TW, Burgener FA, Kamm DE, Fischer HW. Renal

renin and hemodynamic responses to selective renal artery catheteri-zation and angiography. Invest Radiol 1977; 12:381–388.

17. Heyman SN, Brezis M, Reubinoff CA, et al. Acute renal failure withselective medullary injury in the rat. J Clin Invest 1988; 82:401–412.

18. Heyman SN, Brezis M, Epstein FH, Spokes K, Silva P, Rosen S. Earlyrenal medullary hypoxic injury from radiocontrast andindomethacin. Kidney Int 1991; 40:632–642.

19. Heyman SN, Rosen S, Brezis M. Radiocontrast nephropathy: a para-digm for the synergism between toxic and hypoxic insults in the kid-ney. Exp Nephrol 1994; 2:153–157.

20. Brezis M, Rosen S. Hypoxia of the renal medulla—its implications fordisease. N Engl J Med 1995; 332:647–655.

21. Wang A, Holcslaw T, Bashore TM, et al. Exacerbation of radiocon-trast nephrotoxicity by endothelin receptor antagonism. Kidney Int2000; 57:1675–1680.

22. Pflueger A, Larson TS, Nath KA, King BF, Gross JM, Knox FG. Role ofadenosine in contrast media-induced acute renal failure in diabetesmellitus. Mayo Clin Proc 2000; 75:1275–1283.

23. Barrett BJ, Carlisle EJ. Meta-analysis of the relative nephrotoxicity ofhigh- and low-osmolality iodinated contrast media. Radiology 1993;188:171–178.

24. Aspelin P, Aubry P, Fransson SG, Strasser R, Willenbrock R, Berg KJ;Nephrotoxicity in High-Risk Patients Study of Iso-Osmolar and Low-Osmolar Non-Ionic Contrast Media Study Investigators. Nephrotoxiceffects in high-risk patients undergoing angiography. N Engl J Med2003; 348:491–499.

25. Talner LB, Rushmer HN, Coel MN. The effect of renal artery injectionof contrast material on urinary enzyme excretion. Invest Radiol 1972;7:311–322.

26. Liss P, Nygren A, Erikson U, Ulfendahl HR. Injection of low and iso-osmolar contrast medium decreases oxygen tension in the renalmedulla. Kidney Int 1998; 53:698–702.

27. Lancelot E, Idee JM, Lacledere C, Santus R, Corot C. Effects of twodimeric iodinated contrast media on renal medullary blood perfu-sion and oxygenation in dogs. Invest Radiol 2002; 37:368–375.

28. Bakris GL, Lass N, Gaber AO, Jones JD, Burnett JC Jr. Radiocontrastmedium-induced declines in renal function: a role for oxygen freeradicals. Am J Physiol 1990;258(1 Pt 2):F111–F120.

29. Messana JM, Cieslinski DA, Nguyen VD, Humes HD. Comparison ofthe toxicity of the radiocontrast agents, iopamidol and diatrizoate,to rabbit renal proximal tubule cells in vitro. J Pharmacol Exp Ther1988; 244:1139–1144.


vide the necessary clinical information. In manycases, ultrasonography, nuclear medicine, mag-netic resonance, and unenhanced computedtomography can provide sufficient data withoutexposing the patient to iodinated contrast mediaand the risk of contrast-induced nephropathy.

When exposure to iodinated contrastmedia is unavoidable, we recommend the fol-lowing approach.• Nonsteroidal anti-inflammatory drugsshould be discontinued before contrast expo-sure.• Patients should receive hydration withintravenous normal saline starting 2 to 4 hoursbefore receiving the contrast, during the radio-graphic procedure, and continuing 4 to 6 hoursafterward. The duration of the saline infusionshould be longer with more severe chronic kid-

ney disease or if underlying diabetic nephropa-thy is present.• The radiologist or cardiologist should usethe smallest volume of contrast needed toobtain the critical imaging.• Based on currently available data, theremay be an advantage in using iso-osmolar con-trast media.• Hypotension in the peri-imaging periodshould be avoided if possible.• N-acetylcysteine and bicarbonate hydra-tion can be used since they are safe and inex-pensive, although their use is somewhat con-troversial.• Serum creatinine should be measuredapproximately 48 hours after contrast exposureto determine if contrast-induced nephropathyhas occurred.






30. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, manni-tol, and furosemide to prevent acute decreases in renal functioninduced by radiocontrast agents. N Engl J Med 1994; 331:1416–1420.

31. Weisberg LS, Kurnik PB, Kurnik BR. Risk of radiocontrast nephropathy inpatients with and without diabetes mellitus. Kidney Int 1994;45:259–265.

32. Kurnik BR, Allgren RL, Genter FC, Solomon RJ, Bates ER, Weisberg LS.Prospective study of atrial natriuretic peptide for the prevention ofradiocontrast-induced nephropathy. Am J Kidney Dis 1998; 31:674–680.

33. Stone GW, McCullough PA, Tumlin JA, et al, for the CONTRASTInvestigators. Fenoldopam mesylate for the prevention of contrast-induced nephropathy: a randomized controlled trial. JAMA 2003;290:2284–2291.

34. Erley CM. Does hydration prevent radiocontrast-induced acute renal fail-ure? Nephrol Dial Transplant 1999; 14:1064–1066.

35. Vari RC, Natarajan LA, Whitescarver SA, Jackson BA, Ott CE. Induction,prevention and mechanisms of contrast media-induced acute renal fail-ure. Kidney Int 1988; 33:699–707.

36. Trivedi HS, Moore H, Nasr H, et al. A randomized prospective trial toassess the role of saline hydration on the development of contrastnephrotoxicity. Nephron Clin Pract 2003; 93:C29–C34.

37. Taylor AJ, Hotchkiss D, Morse RW, McCabe J. PREPARED: Preparation forAngiography in Renal Dysfunction: a randomized trial of inpatient vsoutpatient hydration protocols for cardiac catheterization in mild-to-moderate renal dysfunction. Chest 1998; 114:1570–1574.

38. Bader BD, Berger ED, Heede MB, et al. What is the best hydration regi-men to prevent contrast media-induced nephrotoxicity? Clin Nephrol2004; 62:1–7.

39. Stevens MA, McCullough PA, Tobin KJ, et al. A prospective randomizedtrial of prevention measures in patients at high risk for contrastnephropathy: results of the P.R.I.N.C.E. Study. Prevention ofRadiocontrast Induced Nephropathy Clinical Evaluation. J Am CollCardiol 1999; 33:403–411.

40. Mueller C, Buerkle G, Buettner HJ, et al. Prevention of contrast media-associated nephropathy: randomized comparison of 2 hydration regi-mens in 1,620 patients undergoing coronary angioplasty. Arch InternMed 2002; 162:329–336.

41. Fishbane S, Durham JH, Marzo K, Rudnick M. N-Acetylcysteine in theprevention of radiocontrast-induced nephropathy. J Am Soc Nephrol2004; 15:251–260.

42. Heyman SN, Goldfarb M, Shina A, Karmeli F, Rosen S.N-Acetylcysteine ameliorates renal microcirculation: studies in rats.Kidney Int 2003; 63:634–641.

43. Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann D, Zidek W.Prevention of radiographic-contrast-agent-induced reductions in renalfunction by acetylcysteine. N Engl J Med 2000; 343:180–184.

44. Baker CS, Wragg A, Kumar S, De Palma R, Baker LR, Knight CJ. A rapidprotocol for the prevention of contrast-induced renal dysfunction: theRAPPID study. J Am Coll Cardiol 2003; 41:2114–2118.

45. Durham JD, Caputo C, Dokko J, et al. A randomized controlled trial ofacetylcysteine to prevent contrast nephropathy in cardiac angiography.Kidney Int 2002; 62:2202–2207.

46. Nallamothu BK, Shojania KG, Saint S, et al. Is acetylcysteine effective inpreventing contrast-related nephropathy? A meta-analysis. Am J Med2004; 117:938–947.

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

48. Erley CM, Duda SH, Schlepckow S, et al. Adenosine antagonist theo-phylline prevents the reduction of glomerular filtration after contrastmedia application. Kidney Int 1994; 45:1425–1431.

49. Chalmers N, Jackson RW. Comparison of iodixanol and iohexol in renalimpairment. Br J Radiol 1999; 72:701–703.

50. Sterner G, Frennby B, Kurkus K, Nyman U. Does post-angiographichemodialysis reduce the risk of contrast medium nephropathy? Scand JUrol Nephrol 2000; 34:323–326.

51. Marenzi G, Marana I, Lauri G, et al. The prevention of radiocontrast-agent-induced nephropathy by hemofiltration. N Engl J Med 2003;349:1333–1340.

52. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast-inducednephropathy with sodium bicarbonate: a randomized controlled trial.JAMA 2004; 291:2328–2334.

ADDRESS: Michael Rudnick, MD, Presbyterian Medical Center, MedicalOffice Bldg., Suite 240, 39th and Market Streets, Philadelphia, PA 19104;e-mail [email protected].

For sales inquiries, [email protected] call 877-MED-POEM

www.infopoems.com




Contrast Induced Nephropathy - Cleveland Clinic Journal of Medicine-2006-Rudnick-75-80

Documents

Transcript of Contrast Induced Nephropathy - Cleveland Clinic Journal of Medicine-2006-Rudnick-75-80