Share this article on:

Complications of Radiographic Contrast Material: Contrast-Induced Nephrotoxicity

Roberts, James R. MD

InFocus: Part 3 in a Series

A compendium of Dr. James Roberts' InFocus columns is available in book form. The 302-page volume, InFocus: Roberts' Practical Guide to Common Medical Emergencies, is available from Lippincott Williams & Wilkins for $59.95 by calling (800)638–3030.

Author Credentials and Financial Disclosure: James R. Roberts, MD, is the Chairman of the Department of Emergency Medicine and the Director of the Division of Toxicology at Mercy Health Systems, and a Professor of Emergency Medicine and Toxicology at the Drexel University College of Medicine, both in Philadelphia, PA. Dr. Roberts has disclosed that he has no significant relationships with or financial interests in any commercial companies that pertain to this educational activity.

Release Date: August 2004



Learning Objectives: After reading this article, the physician should be able to:

  1. Identify patients at high risk for dye-induced nephrotoxicity.
  2. Discuss the incidence and pathophysiology for dye-induced nephrotoxicity.
  3. Describe prevention strategies that may limit this complication.

Emergency physicians are often called upon to order numerous studies using iodinated intravenous contrast material. We are all aware of the occasional anaphylactoid reactions that occur, a nightmare for the radiology department, the patient, and the ordering physician. There are, however, other clinical considerations surrounding the use of contrast materials, including the effect on renal function, and the inadvertent extravasation of dye into the soft tissues. The emergency physician also must be aware of special situations that are potential harbingers of dye problems, such as metformin use and multiple myeloma.

It is not uncommon to get a call from the radiology department saying that they are sending over a patient (always a bad way to start a conversation) who is having an allergic reaction, or an individual who had his IV infiltrate during a dye injection. The worst scenario is a code called in radiology, occasionally a disaster with regard to logistics and outcome. Radiologists are the true the experts in this field, and they often totally handle minor reactions or extravasations. The majority of EPs probably think that IV dyes are now totally safe, and don't give the process of an intravenous dye injection a second thought, but the burden of evaluating and treating an aliquot of adverse reactions occasionally winds up in the ED's mix of the sick and injured.

Fortunately, the advent and now routine use of newer and less toxic contrast agents has markedly reduced the incidence of serious reactions, including anaphylactoid reactions. Local tissue reactions, once a formidable problem, have all but disappeared. Although the newer low osmolality nonionic contrast agents are more expensive than their predecessors, they are now quite affordable, costing only $10 to $20 per test. Most radiology departments have changed their protocols to use the less problematic dyes exclusively.

Previous columns have discussed the scenarios of extravasation of IV dye and acute anaphylactoid reactions. Fortunately, most extravasations, even large ones, are minimally problematic, and one should avoid overtreatment or unnecessary anxiety-producing remarks, conclusions, or interventions. The idiosyncratic acute anaphylactoid reactions, although very rare, can be fatal, and they are treated like anaphylaxis. This month's column discusses the omnipresent renal toxicity associated with IV dyes. Most emergency physicians know that contrast agents can somehow injure the kidneys, but the practicing clinicians who order these tests usually have only a rudimentary knowledge of the specifics.

Although numerous reviews seem to take the same bottom line, a detailed review of the literature demonstrates that many dogmatic statements about contrast nephropathy are simply not supported by the facts. I thought I knew enough about contrast dye and the kidneys, but now I am not so sure.

Preventing Contrast Nephropathy: What is the Best Strategy? A Review of the Literature Cox C, Tsikouris JP J Clin Pharmacol 2004;44:327

This is an up-to-date extensive review and discussion of the issues of contrast agent-induced nephropathy. This erudite and detailed review discusses commonly held views about the pathogenesis of contrast nephropathy, risk factors for its development, and types of contrast agents that are available. It also analyzes various preventive options. Contrast nephropathy is generally considered to be third leading cause of acute renal failure in hospitalized patients; all clinicians who order IV contrast studies must be aware of the potential renal complications from IV dye. High-risk patients may have as high as a 50 percent incidence of contrast nephropathy, although the general overall complication rate is actually quite low, about one to six percent.

The generally accepted definition of contrast nephropathy is a clinically significant change in renal function following dye administration. Most authors consider an absolute rise in serum creatinine of 0.5 mg/dL, or a 25 percent increase from baseline, as the laboratory definition of contrast nephropathy. Others use a 25 percent to 50 percent decrease in GFR as the definition. The rise in creatinine occurs within 24 to 48 hours following contrast administration, and usually peaks in three to five days. In most cases, the renal dysfunction is transitory, and returns to normal within seven to 10 days. Although the vast majority of cases are clinically silent and reversible and do not result in oliguria or the need for dialysis, about a third of patients have some permanent decline in renal function secondary to contrast nephropathy. Those who experience permanent renal dysfunction have an increased overall mortality rate attributed to this complication.

Patients at greatest risk for long-term renal complications are those with underlying renal insufficiency prior to the procedure. Diabetics with renal disease are at very high risk. The more diagnostic procedures performed, the greater the chance for contrast nephropathy.

Pathogenesis and Risk Factors: The generally accepted pathophysiology of contrast nephropathy is direct renal tubular epithelial cell toxicity, coupled with contrast-induced renal medullary ischemia. IV dyes produce adverse alterations in renal hemodynamics, the end result being prolonged periods of vasoconstriction. Vasoconstriction induces an overall decreased renal blood flow, particularly in the renal medulla. Alterations in the local renal metabolism of nitric oxide, adenosine, endothelin, angiotensin, and prostaglandin also play some yet undefined role in this complication.

Patients with normal renal function prior to contrast administration have a very low risk of developing renal failure. As the pretest renal function declines, the risk of contrast nephropathy increases. Diabetes is another risk factor, but there are conflicting results regarding this disease as an independent risk factor. It is now generally believed that the incidence of contrast nephropathy in diabetic patients with normal renal function is similar to that reported in the general population. Dehydration, congestive heart failure, concurrent nephrotoxic drug use, and other contributing factors are also culprits.

The Role of Contrast Material: The osmolality of the contrast agent is thought to be an important variable in the development of nephrotoxicity. Some studies show that low osmolar and iso-osmolar contrast agents will reduce the risk of contrast nephropathy when compared with high osmolar compounds. Other studies find no such correlation. The increased osmolality theoretically slows elimination of the contrast material, exposes the kidneys to the contrast agent for longer periods of time, and increases the ischemia in the medullary region. This is theoretical pathophysiology, however, with no convincing evidence to support this contention. Most experts recommend low osmolality dyes for all procedures.

In addition, the volume of contrast material used in the radiographic study correlates with the incidence of nephrotoxicity in some studies but not others. Diagnostic studies, such as coronary angiography, use a smaller volume of dye (about 100 ml) than those procedures that include cardiac interventions (approximately 250–300 ml for angioplasty). Efforts should be made to reduce the amount of contrast material infused.

Prevention Strategies: The majority of well patients tolerate a dye study very well, but a number of prophylactic interventions have been suggested to reduce the incidence of contrast nephropathy in high-risk populations. Strategies have included hydration, n-acetylcysteine (NAC), and a number of other theoretically beneficial agents. While some studies show benefit, the data are confusing and contraindicatory, and there is no clear superiority of any single prophylactic intervention. Interventions with the best clinical evidence to advocate their use include generous fluid administration, pre-test NAC, and possibly pre-test alkalinization.

The administration of saline — before, during, and after the administration of contrast material — will likely decrease the incidence of contrast nephropathy. This is especially true in dehydrated patients. The exact volume of fluid that is required is unknown, but “adequate hydration” is strongly endorsed. No specific protocol has been widely adopted, but the administration of saline at a rate of 1–1.5 ml/kg/h beginning six to 12 hours prior to the procedure and continuing for 12 hours following contrast administration is one suggested protocol. Adding mannitol or Lasix to saline hydration is not advocated, and may actually increase the incidence of contrast nephropathy.

NAC may prevent contrast nephropathy by two mechanisms. NAC is thought to have a direct vasodilating effect on the microvascular function of the kidney, contributing to better renal hemodynamics. Because NAC is a free radical scavenger, it may decrease oxidative tissue damage. The true benefit of pre-treating patients with NAC is controversial, and data are conflicting. Not all clinicians have adopted its routine use. NAC is generally administered orally at least 24 hours prior to the procedure. A regimen of 600 mg twice a day on the day before, and again on the day of contrast administration has been one suggested protocol. Some studies have shown a significant decrease in contrast nephropathy with the oral NAC regimen. Others have found absolutely no benefit. The amelioration of contrast nephropathy is probably greatest in patients who have the best underlying renal function, and for those who have received lower doses of the contrast dye. When high-dose contrast infusions have been studied, especially in patients with significant renal dysfunction, the benefit of NAC is less convincing. Pretreatment with NAC should always be coupled with adequate hydration.

Other Interventions: The dopamine agonist fenoldopam has been investigated as a vasodilating agent to decrease renal ischemia during dye studies. Early studies were quite optimistic, but recent prospective randomized trials have produced conflicting and certainly less impressive results. Most studies fail to show any benefit of fenoldopam. Currently the use of fenoldopam is not recommended as a routine intervention to reduce contrast nephropathy.

Other theoretically helpful agents have been studied, but none has been found to be effective enough in reducing or preventing contrast nephropathy to support its routine use. Theophylline, calcium channel blockers, diuretics, atrial natriuretic peptide, endothelin antagonists, ACE inhibitors, and prostaglandins all have attractive pharmacological properties that could render them beneficial, but trials have not been convincing. Although mannitol has been used to lessen renal dysfunction in a number of clinical situations, it has no beneficial effect in preventing contrast nephropathy. In fact, mannitol may be disadvantageous. Furosemide (Lasix) is of no benefit, and also may be harmful. (N Engl J Med 1994;331[21]:146, and Kidney Inter 1994;45:259.)

Recommendations: The authors conclude that all patients undergoing IV contrast injections should be assessed for the risk of subsequent contrast-induced nephropathy. Patients with underlying renal insufficiency and those with a history of renal problems secondary to diabetes are the highest risk populations. These authors believe that such patients should receive the low osmolarity contrast dyes, and the total dose of contrast material should be minimized. All patients should be adequately hydrated prior to the administration of dye. There is some potential benefit of pre-treatment with NAC, yet to be fully determined. There is no clinical evidence to support the use of other previously suggested renal protective agents.

Comment: Most issues surrounding contrast-induced nephropathy are relegated to hospitalized patients, and EPs likely underestimate the incidence of dye complications. I honestly cannot recall a single patient in 32 years of practice that had clinically important contrast nephropathy induced by a study I ordered in the ED. The literature suggests that I just don't get good enough follow-up, and statistically I must have injured some kidneys. Most EPs probably have a similar positive experience. Once in the hospital, high-risk patients are easily identified, and one has the luxury of renal function testing, alternative studies, pretreatment, and hydration. ED patients requiring emergency contrast studies are likely at the greatest risk for the development of contrast nephropathy. We order these studies on a regular basis, and generally do not give a second thought to complications that might occur two or three days later, but all of us have had the annoying phone call from radiology when the technician refuses to do the scan until the serum creatinine has been measured. This turns out to be a stellar precaution.

Most patients at high risk can usually be identified from clinical history, but subclinical diabetes, asymptomatic chronic hypertension, and occult renal failure are rampant in the undifferentiated ED patient. It seems prudent to adhere to the concept of a routine creatinine determination prior to contrast material administration in almost all patients. Generally the 30 to 60 minutes required for a creatinine determination is of no great consequence. The occasional patient, such as one with aortic dissection, may have to undergo emergency angiography with a realization that the test may further injure his kidneys. When an alternative study such as MRI, V/Q scan, or ultrasound can be performed, the dye question becomes moot. Certainly patients with a history of hypertension, diabetes, or multiple medical problems always should have their creatinine checked prior to a contrast study. Some hospitals will forgo creatinine testing in young healthy patients with no pre-existing medical problems. In today's medical legal environment, however, such radiologists are few and far between.

There appears to be no way to guarantee that IV dye will not injure kidneys, and often one has to take that risk:benefit chance given the clinical scenario. It would be nice to have all patients well hydrated prior to dye administration. Individuals requiring such emergency studies, however, commonly have underlying medical problems that not only compromise renal function but also cause dehydration. Patients on diuretics, those with persistent nausea/vomiting or blood loss, and individuals who are otherwise dehydrated from lack of fluid intake, pose the greatest risk for contrast nephropathy. Empiric administration of IV fluids prior to a dye study in the ED patient, especially for those with unknown renal function or vague past medical history, seems like a prudent strategy if the cardiovascular/respiratory systems can otherwise tolerate a fluid load.

The issue of mannitol and Lasix is an interesting one. One might intuit that diuresis would limit renal damage from dye, but just the opposite is true. Mannitol might be reasonable to clear myoglobin, but it should not be used after dye administration. Of course, protocols helpful for allergic reactions, such as steroids and antihistamines, have no protective role for contrast nephropathy.

Although NAC is a suggested prophylactic intervention in hospitalized patients, the protocol requires 24 hours, and is impractical for ED patients. I have never seen NAC given for renal protection for an ED dye study. Perhaps NAC could be started in the ED, but I have not seen this as common practice either. In fact, many radiologists or nephrologists don't ascribe any protective effect to NAC, and many don't use it. This is confusing to me because a recent meta-analysis (Lancet 2003;362:598) seems to support its use to reduce contrast nephropathy in patients with chronic renal insufficiency. It's interesting to speculate on the value of IV versus oral NAC, and a recently FDA approved IV NAC is now available (Acetadote). I have not seen any convincing data on the ED administration of NAC for emergency studies, and I know of no authors who advocate its use.

Because contrast nephropathy is the third leading cause of hospital-acquired renal failure, all physicians must be cognizant of the potential for their investigative activities to result in increased length of stay, increased morbidity, and even mortality. Fortunately, most patients will recover from dye-induced transient renal dysfunction, and the problem is often a subclinical laboratory aberration. But the kidney insult can be severe enough to necessitate long-term dialysis or even lifetime dialysis in a few individuals.

A risk factor for renal insufficiency that is often not well appreciated by clinicians is the long-term use of NSAIDs. Many patients with chronic painful conditions such as arthritis will pop NSAIDs like candy. Such patients can develop renal failure solely from OTC NSAID use. We have all seen the healthy 20- or 30-year-old who develops acute renal failure secondary to excessive ibuprofen use for a chronic toothache, headache, or back pain. I don't think the public is very aware of this potential complication from these omnipresent OTC medicines. Those with pre-existing renal dysfunction can have it exacerbated by the relatively short-term use of NSAIDs. One should always be hesitant to prescribe long-term potent NSAIDs to any elderly patient, individuals with hypertension or diabetes, or those with chronic medical illnesses before checking their creatinine.

An NSAID prescription often has a very long half-life, much longer than anticipated by the prescribing EP. You may not think five to seven days of naprosyn, Indocin, or Motrin is a big deal, but often the primary physician continues your good thought for weeks to months after the ED visit when you only intended short-term use. Also, we all know patients who are unknowingly taking two or three NSAIDS prescribed by different doctors. There are some scattered case reports of permanent renal failure after relatively small dose short-term NSAID use, such as the potent prostaglandin-inhibitor, Ketoralac. Given this potential rare downside, I generally personally opt for real pain medicines, otherwise known as narcotics. I don't have a problem with a dose or two of Ketoralac in the ED in selected patients, but its routine use as an analgesic is not my clinical style.

Fenoldopam initially appeared to have great potential as a renal protective agent, but recent studies have failed to live up to that expectation.

Are Screening Creatinine Levels Necessary Prior to Outpatient CT Examinations? Tippins RB, et al Radiology 2000;216:481

This retrospective analysis of the records of 2,034 outpatients evaluated whose serum creatinine was measured prior to IV contrast enhanced CT scanning. The creatinine was normal (less than 1.4 mg/dL) in 92 percent of patients. The creatinine was considered borderline (1.5 to 1.9 mg/dL) in five percent and abnormal (greater than 2 mg/dL) in only three percent. The baseline creatinine was not related to age. Almost every patient with an abnormal serum creatinine had one or more risk factors for nephrotoxicity that had been previously identified. The most common was prior renal insufficiency. The rest of the patients had diabetes. In only two of 2,034 patients did these authors find a creatinine that was in the abnormal range (one was 2.0 and the other 2.2 mg/dL) without identifiable historical risk factors.

The authors conclude that questioning patients prior to IV contrast administration for the presence of prior renal disease or diabetes would identify essentially all patients who required a pre-test measurement of serum creatinine. The selective screening based on the presence or absence of these two risk factors would be associated with substantial cost savings.

Comment: The authors of this study reported that only one in 1,000 patients who are sent for an outpatient IV radiographic contrast study will have an unexpectedly high serum creatinine that puts them at high risk for contrast nephropathy. I think their conclusion was that in the absence of a history of renal disease or diabetes, a serum creatinine could be safely eschewed. They did not mention asking patients about hypertension, the use of potentially nephrotoxic drugs, nor assessing for the presence of proteinuria.

It only takes one or two shifts in the ED to realize that many patients are totally unaware of their medical problems, and some have significant renal failure and are totally clueless about its presence. All they know is that they are seeing a doctor and are taking medication for something. In addition, numerous undiagnosed diabetics present to EDs requiring IV contrast material. Many will have clandestine renal insufficiency. I would take exception with the authors' conclusions. Although the chance of having a significant nephrotoxic reaction to IV dye was certainly low in this outpatient population, it was not zero.

Any busy ED performs more than a thousand contrast enhanced CT scans per year. The risk of at least one case per year of significant nephrotoxicity, some of which are not reversible, is not a strong argument for avoiding a simple blood test. Most patients will not understand such a statistical argument when it comes to their damaged kidneys. Because it's only a test and a very sensitive screening one at that, it seems reasonable just to order the serum creatinine prior to injecting the dye. This is especially true in patients presenting to the emergency department. They are statistically more likely to be unaware of their occult medical problems or have more underlying disease than the general population sent for an elective outpatient CT scan.

This article also references what is considered a creatinine that is too high for contrast administration. These authors consider a creatinine of 1.9 mg/dL to be high borderline, with levels of 2 mg/dL or higher consider distinctly abnormal. Individual radiologists vary, and many of them will perform a contrast study in patients with these borderline creatinine levels. Pretest hydration, limiting the volume of dye, and opting for the lower osmolality agents are suggested in these patients, but the latter recommendations have little scientific basis. Curiously, the specific pre-test creatinine level in patients with renal disease correlates poorly with the POS test level. It appears that any abnormal kidney function is a risk factor.

Certainly the ED often can hydrate patients prior to obtaining a study and help reduce contrast nephropathy. I used to argue with the radiologists when they demanded a creatinine level on all patients before performing a contrast study. I have given up this argument. Most hospitals use a serum creatinine of 1.8 to 1.9 mg/dL as the upper limit for performing an elective dye study. In some cases, such as the hypertensive diabetic who may have an aortic dissection, an elevated creatinine is to be expected, yet the study must still be performed. An aortogram can be done with only 30 to 50 ml of dye, and you just have to take the risk.

Finally, one should be aware of renal failure in patients with multiple myeloma who receive IV dye. The proposed mechanism of dye-induced renal failure is precipitation of myeloma proteins in the tubules. Adequate hydration is suggested in these patients, even though the serum creatinine is not alarming. (Am J Roentgenol 1971;133583.)

Back to Top | Article Outline

Risks for Contrast-Induced Nephropathy

  • ▪ Pre-existing renal insufficiency manifested as elevated serum creatinine1
  • ▪ Diabetes2
  • ▪ High osmolality dyes and high dose of contrast material (greater than 100–150 ml)3
  • ▪ Congestive heart failure
  • ▪ Dehydration
  • ▪ Concurrent administration of nephrotoxic drugs4
  • ▪ NSAIDs
  • ▪ Diuretics
  • ▪ Amphotericin B
  • ▪ Aminoglycosides
  • ▪ ACE inhibitors
  • ▪ Multiple myeloma5
Back to Top | Article Outline

Clinical Characteristics of Contrast-Induced Nephrotoxicity

  • ▪ Third leading cause of iatrogenic acute renal failure.
  • ▪ Associated with increased in-hospital mortality.1
  • ▪ First manifested as a rise in serum creatinine within 48 hours of dye administration.
  • ▪ Nephrotoxicity is defined as a rise in serum creatinine of 0.5 mg/dL or a 25 percent increase from baseline.
  • ▪ Main pathophysiology is probably renal medullary ischemia secondary to contrast-induced vasoconstriction.2
  • ▪ Incidence may be as high as 50 percent in high-risk patients, but overall incidence is one to six percent in hospitalized patients.
  • ▪ Renal dysfunction is usually reversible, but up to 30 percent of cases result in some permanent decline in renal function. Rare cases require permanent dialysis.


Back to Top | Article Outline

Strategies to Prevent Contrast-Induced Nephrotoxicity

  • ▪ Periprocedural hydration: IV saline 1–1.5 ml/kg/hr, infused for 6 to 12 hours before and 12 hours after procedure
  • ▪ Use low-osmolality nonionic contrast agent:1 Iso-osmolar to 600–700 oSm/kg
  • ▪ Limit amount of contrast agent1: 100–150 ml of contrast agent
  • ▪ N-Acetylcysteine (NAC)2
  • ▪ Conflicting data on efficacy
  • ▪ Requires two-day preparation
  • ▪ 600 mg orally BID on day prior to procedure and on day of procedure (total four doses)

1The pretest creatinine level is the most sensitive predictor of dye nephropathy.
Cited Here...

2Diabetes is not an independent risk factor. If renal function is normal, the incidence of contrast nephropathy is similar to the general population.
Cited Here...

3Some studies dispute these risks.
Cited Here...

4Drugs decrease renal blood flow or are additive to dye toxicity.
Cited Here...

5May be prevented with hydration.
Cited Here...

1Increases risk of sepsis, bleeding, and respiratory failure.
Cited Here...

2Direct renal cytotoxicity also may occur.
Cited Here...

1Some studies do not substantiate this contention.
Cited Here...

2Combined with periprocedure hydration. NAC use not standard prior to ED studies.
Cited Here...

© 2004 Lippincott Williams & Wilkins, Inc.