April 24, 2014

IV Contrast fact and fiction

There are times when the safety of IV contrast can be a confusing quagmire. We know that iodinated contrast for CT scans can hurt the kidneys. But is it harmful for someone who already has renal failure and is on dialysis?  What about the breastfeeding mother? Will IV contrast harm her infant? How should we pretreat patients who have had a previous reaction to IV contrast? These questions and more answered on this episode of ERCast.

Direct Download Podcast

If you want to be uber-educated, check out The American College of Radiology’s (ACR) Manual on Contrast Media. As this type of document goes, it’s actually pretty concise, and something you may want to keep as a reference.

Q: Is it safe to give iodinated contrast to a breastfeeding mother?

A: Probably.

Less than 1% of contrast is excreted in the breast milk and less than 1% of that is absorbed by the child’s gut. If the patient is concerned about even this level of exposure, she can always discard the next 24 hours of breast milk (the so-called ‘pump and dump’), but the ACR feels it’s safe for breastfeeding to continue without interruption after administration of iodinated contrast.

Q: If we give iodinated contrast to a dialysis patient, do they need to be dialyzed immediately, or can they wait until their next regularly scheduled dialysis?

A: Most patients can wait until their next regularly scheduled dialysis. If a patient has severe underlying cardiac disease and the small osmotic load of contrast will potentially send them into pulmonary edema, then urgent dialysis may be indicated. Does a potential risk warrant a default of immediate – post contrast dialysis? Probably not. I think a more reasoned approach is to assess the patient after they have received contrast. If there are signs of volume overload, then dialyze. If not, then dialysis can wait.

There is also a theoretical risk of making an oliguric patient anuric, but limited evidence to give a clear answer. Bottom line from the American College of Radiology:

“Unless an unusually large volume of contrast medium is administered or there is substantial underlying cardiac dysfunction, there is no need for urgent dialysis after intravascular iodinated contrast medium administration.” For a specialty that defines itself by beating around that bush, that is a pretty clear statement. No clinical correlation recommended, nothing in the differential diagnosis includes that this patient who you just sent home  is going to die from this incidentaloma.

The Renal Fellow Network has an excellent synopsis on dialysis post CT contrast. Worth a read


Q: What is the best way to pretreat patients who have had a previous reaction to iodinated contrast?

A: Many EDs give a two pronged pretreatment: steroids and antihistamines with a one hour delay between treatment and injection of contrast.


The steroid studies have shown that pretreatment several hours before contrast decreases the incidence of reactions. The limited data on a single steroid dose two hours before contrast injection shows no benefit. This goes along with our current thinking about steroids – they take several hours to work.


H1 blockers: Probably useful when given one hour before contrast injection.

H2 blockers:  Unclear if they have beneficial effect. There is little downside to giving an H2 blocker along with an H1 blocker, but it should not be used as a substitute.

For a deep dive review of pretreatment, check out this  2006 article from the British Medical Journal

Q: Is there a role for pre-treatment to decrease the chance of contrast induced nephrotoxicity (CIN)?

A: A bigger question is: why does CIN happen in the first place? No one really knows. There are theories ranging from  direct renal tubular toxicity to vasoconstriction. Since the cause isn’t clear, treatment is somewhat a patchwork of guesswork.

There is no evidence that gives a clear creatinine cutoff as to when we should or should not give contrast. The ACR feels that a creatinine of less than 2 mg/dL is  safe for IV contrast, but that’s a fuzzy line. Is a creatinine of 1.9 mg/dL safer than 2.1? I’ll put it this way: clinical correlation recommended.

How should we pretreat patients we’re worried about CIN? Hydration is a good bet. There is decent evidence that pretreating with IV 0.9% NS decreases the incidence of CIN. How much, how long, how fast to infuse? Unknown.

The bigger mystery lies in pretreatment with sodium bicarbonate and N-acetylcystine (NAC). These have both fallen in and out and then in and then back out of vogue. Where they are now in the sphere of medical thinking is a mystery to me.

Bicarb: there is  some evidence to say it decreases CIN and other evidence that it makes no difference.

Does it work? Conflicting evidence.

NAC: here is what the ACR has to say about NAC pretreatment:

The efficacy of N-acetylcysteine to reduce the incidence of CIN is controversial. Multiple studies and a number of meta-analyses have disagreed as to whether this agent reduces the risk of CIN. There is evidence that it reduces serum creatinine in normal volunteers without changing cystatin-C (cystatin-C is reported to be a better marker of GFR than serum creatinine). This raises the possibility that N-acetylcysteine might be simply lowering serum creatinine without actually preventing renal injury. There is insufficient evidence of its efficacy to make a definitive recommendation. N-acetylcysteine should not be considered a substitute for appropriate pre-procedural patient screening and adequate hydration.

Does it work: The jury is still out.

Q: Who should have their creatinine checked before IV contrast? 

A: Some of this is evidence based, some is OGSAR based (old guys sitting around a room). Below are the ACR consensus  recommendations taken directly from the document.

  • Age > 60
  • History of renal disease, including:
  1. Dialysis
  2. Contrast-Induced Nephrotoxicity
  3. Kidney transplant
  4. Single kidney
  5. Renal cancer
  6. Renal surgery
  • History of hypertension requiring medical therapy
  • History of diabetes mellitus
  • Metformin or metformin-containing drug combinations*

Patients who are scheduled for a routine intravascular study but do not have one of the above risk factors do not require a baseline serum creatinine determination before intravascular iodinated contrast medium administration.

*Metformin does not confer an increased risk of CIN. However, metformin can very rarely lead to lactic acidosis in patients with renal failure. Therefore, patients who develop CIN while taking metformin are susceptible to the development of lactic acidosis. To assess the risk of lactic acidosis, it is probably prudent to stratify the risk of CIN in patients taking metformin who will be exposed to intravascular iodinated contrast medium .

Nephrogenic Systemic Fibrosis

We’ve been talking about safety of iodinated contrast for CT scans, but what about gadolinium – the contrast used for MRI? It’s much less common for us to order an MRI than CT and even less common to give gadolinium, but it occasionally comes up. Is there a problem with Gadolinium and the kidneys? We used to think, and not that long ago, that gadolinium was safe for the kidneys. About as safe as injecting saline. The risk of direct nephrotoxicity is indeed extremely low – there have been some cases of gadolinium related kidney injury, but for the most part, it’s not directly nephrotoxic.  The problem is something called Nephrogenic Systemic Fibrosis or NSF. NSF is a condition of  progressive fibrosis throughout the body. It usually starts with skin thickening and pruritis but can involve several organs including the heart, lungs, esophagus, skeletal muscles. It can even be fatal. Not so good. The exact mechanism is unclear, but the primary risk factor for developing NSF is renal insufficiency. The screening for who is a risk for NSF is much like the screening for who is at risk for kidney injury from CT scans. Avoid gadolinium in patients on dialysis, acute kidney injury or chronic renal insufficiency with a depressed GFR. A GFR less than 30 scrubs the mission – do not inject.

This podcast was inspired by and partly based on a post by one of my favorite FOAMed bloggers…

Bryan Hayes on Renal Failure, Dialysis and IV contrast


Related posts:


  1. Great piece, what about gadolinium and pregnancy/ breastfeeding?

    • Rob Orman says:

      There is insufficient data to make a definitive recommendation on gadolinium in pregnancy. Here’s what the ACR has to say…

      “Because it is unclear how gadolinium-based contrast agents will affect the fetus, these agents should be administered only with extreme caution. Each case should be reviewed carefully and gadolinium-based contrast agent administered only when there is a potential overwhelming benefit to the patient or fetus that outweighs the possible risk of exposure of the fetus to free gadolinium ions.”

  2. Hi Rob
    Nice summary.
    My mate Roy (local Pharmacist) and I got into a metformin mythology / renal impairment / contrast debate. It really is a bizarre circular argument when you look at the evidence for (or lack thereof ) for metformin causing lactic acidosis. It just doesn’t seem to be true.
    See: http://broomedocs.com/2011/04/medical-myths-001-metformin-mythology/
    for reference articles, including one done at my old hospital -- Fremantle

    Still -- if you have a diabetic with renal, heart or other acute illness (eg sepsis ) then you need to stop the metformin.

  3. Mike Burke says:

    I am a Radiologist. I also deal with patients reporting a history of contrast allergy.

    IV contrast material is not just one substance however. There are several distinct products on the market by different manufacturers at any given time. These products have gone through several generations over the years. Newer products are generally safer, better tolerated and more expensive. ( Think of antibiotics as an analogy )
    Unlike with antibiotics, patients generally aren’t told which product they received, so a bad experience with one product gets generalized to an “allergy” to contrast material in general.

    Compounding this issue, there was a major change in the type of IV contrast material sometime in the 1990s. All the vendors changed from ionic ( Hyperosmolar) contrast material to nonionic ( iso-osmolar) contrast material, a totally different class of material. Reactions such as vomiting, flushing and hives were very common with the the older agents, and much less common with the newer agents.

    When a patient reports an “allergy” to contrast material, it is important to ascertain the date and a description of the reaction. A lot of what was labelled as allergy was really due to the rapid infusion of hyperosmolar contrast in the past.

  4. Johnnymac says:

    Rob, enjoyed the podcast. Hope you are well.

  5. You didn’t mention Remote Ischemic Preconditioning as a prevention for kidney injury related to IV contrast: remote ischemic preconditioning has been shown, by an undefined mechanism, to protect renal, as well as other tissues such as brain and myocardium, from ischemic injury -- hypothesized to be part of the mechanism of contrast induced AKI. Er, et al (Circulation July 17, 2012) looked at a small group and showed a reduction in AKI -- using the usual clinically unimportant criteria for AKI. Preconditioning is a simple affair: blow up a BP cuff well above systolic pressure for 4 cycles of 5 minutes on, 5 minutes off, and ending shortly before the contrast injection. AKI was reduced from 26% to 12% in the experimental group (this was a high risk group who also received NAC and saline infusion). BTW, the cuff is applied around an arm, not around the kidney. Not ready for clinical use, but pretty cool technique if it is ultimately demonstrated to work.

    In more important areas of life, we use ischemic preconditioning to prepare for a bicycle race: it’s generally called “warm-up.” We use heart rate as a marker, but induced ischemia prior to a major ischemic event (the race) seems to provide improved performance. In usual protocols, the ischemia is in the same organ (legs) as during the event. If “remote” ischemia were to work, I could simply intermittently inflate a BP cuff to painful levels, while driving to the race, and avoid all that silly stuff of pedaling like mad, while going nowhere, on a trainer before the start of the race. I think I’ll try it for tomorrow night’s time trial. This could be the start of a great journal article.

Awesome article, I know - please share your erudite thoughts...