MAKE this a better outcome

The PRESERVE Trial, which was recently published in the NEJM was a large study of a variety of preventive measures for contrast-induced nephropathy. It used a 2x2 factorial design to test the efficacy of NaHCO3 vs. normal saline and n-acetylcysteine vs placebo for the prevention of CIN. The trial was stopped early as there were no signals that any of the treatments were better than any other suggesting that the best treatment for CIN right now is likely the use of both saline and the smallest possible quantity of low-osmolar contrast. The overall decline in the rates of CIN over the last few years are likely more related to the change in the way that contrast is used rather than any special benefit that we were imparting using novel measures to prevent CIN.

The trial was enriched to try and increase the rate of AKI - it included patients with an eGFR between 15 and 45 (non-diabetic) increasing to 60 in diabetics. The overall mean eGFR was 50 so perhaps there were not quite enough patients with advanced disease but given that more than 5000 patients were included in the study, it is hard to really draw the conclusion that it was underpowered for subgroups. Patients with AKI were also understandably excluded and it is unclear what the risk is in this subgroup of patients.

The other thing that was really interesting about this study was the outcome used. Traditionally, studies on CIN have used AKI as the outcome. This being defined typically as some change in creatinine in either absolute or percentage terms. The current AKIN definition of stage I AKI is a 0.3mg/dl increase. This study used an increase of 0.5mg/dl. AKI of this magnitude has been shown in large studies to be associated with adverse outcomes including increased length of hospitalization and mortality but there is always a lingering question about how clinically significant it is in the long run when it tends to resolve in most patients.

Because of these concerns, there has been a recent move towards using MAKE (major adverse kidney events) as a composite outcomes in trials of kidney disease. This concept, stolen somewhat from the cardiology literature is thought to be more meaningful as it results in real, long-term harm to patients. In this study, the authors chose MAKE90 - a composite of death, need for dialysis and permament 50% increase in creatinine at 90 days as the outcome. Overall, approximately 9% of patients had AKI following contrast administration and about 4.5% had a MAKE90 event (2.5% died, 1.5% required dialysis and 1% had a permanent decline in renal function). Another thing I would take away from this study is that we should not let people tell us that CIN does not exist (which I have heard around the halls more often than I like over the last year or so).

Overall, I think this move towards MAKE as an outcome in clinical trials is a welcome one. It is a well defined outcome that demonstrates clear harm to patients. There is no reason to chuck AKI out completely - it is a perfectly good secondary outcome - but in terms of figuring out who are the patients most likely to benefit from interventions in the future, MAKE is the way to go

NephMadness 2014 Part 4 - AKI Bracket

In the AKI bracket one team caught my eye, RIPC, Remote Ischemic Preconditioning. This is a procedure of inducing transient ischemia in the arm by inflating a BP cuff for 5 minutes x2 with an interim deflation for 5 minutes. This procedure was done before coronary angiography, aneurysm repair and was shown to reduce myocardial ischemia, renal injury and contrast-induced nephropathy. These procedures have a high prevalence and there are no other therapies to reduce CNI other than fluids (see ACT trial on N-acetyl cysteine), this team is a good contender this year. Hopefully we will see more evidence for this simple procedure. However, my favorite from this group is normal saline, simple, cheap effective, global! How many times has an acute renal failure case, presented to you as a complicated mess by a resident, been solved by some salty water!! I love it! This team goes all the way to the final four for me.

Remember to fill out your NephMadness 2014 brackets! Find them at eAJKD

Contrast-Induced Nephropathy (CIN): An Update from the Cardiology Literature

Happy new year to all! I’ve been thinking about how much Nephrology content there is in non-renal journals and how much we miss by just reading the usual renal periodicals each month. You cannot be expected to read every Cardiology/Endocrinology/Rheumatology journal every month so I thought I would bring you what I’ve noticed in the non-Nephrology literature recently. Today I’m focusing on CIN. If the post proves popular, we may run a regular literature watch from non-Nephrology journals on RFN.

Since, N-acetylcysteine (NAC) has fallen out of favor, all we have in our armamentarium for CIN prevention is IV volume expansion. There were 3 recent publications in the Journal of the American College of Cardiology which you may have missed regarding novel therapies for CIN prevention. Firstly, the PRATO-ACS study which was a RCT of rosuvastatin in >500 consecutive statin-naïve patients presenting with a NSTEMI who proceed to cardiac catheterization. Greater then 50% of patients were diabetic and mean eGFR was 82mls/min. The incidence of CIN was significantly lower in the rosuvastatin group (6.7 vs 15.1%) as were other renal, cardiac events and mortality. In the same issue, a study of 3000 diabetic CKD (Stage 2 or 3) patients randomized patients to either rosuvastatin or placebo 5 days before intra-arterial contrast. The rosuvastatin group had a significantly lower incidence of CIN compared to controls (2.3% vs 3.9%). Statins are postulated to improve CIN by their anti-inflammatory and positive endothelial effects. These large scale prospective RCTs support a meta-analysis from 2012 suggesting a benefit in statin use for CIN.

A meta-analysis of trials using ascorbic acid for the prevention of CIN was also recently published in J Am Coll Card. Over 1500 patient were included from 9 RCTs and it was demonstrated that receiving ascorbic acid (mix of oral and IV) had 33% less risk of CIN compared to the control group. Six of the 9 studies had significantly p values themselves. The postulated mechaism of action of ascorbic acid is similar to the theory behind NAC, antioxidant free-radical scavenger i.e who knows?

Verdict: The 2 statin trials are good sized prospective studies in high risk populations (ACS & diabetic CKD) receiving intra-arterial contrast and do support short term statins in this circumstance. Knowledge gaps exist for lower risk patients and those already on statins (a lot of patients these days) as well as dose, duration and the benfit of other statins. The Vitamin C meta-analysis looks interesting but given the study heterogeneity and the NAC story, I remain skeptical until well powered RCTs are performed.

ACT: Now more than thespians

As a kid growing up in the San Francisco ACT brought to mind guys in tights running around and projecting their voices at the American Conservatory Theater. Now I've got to adjust my association with the acronym with the recent publication in Circulation of the Acetylcysteine for Contrast-Induced Nephropathy Trial (ACT).


ACT is the largest randomized trial to date to evaluate the use of oral acetylcysteine prophylaxis in patients at risk for contrast induced nephropathy (CIN). In this multicenter trial, 2803 patients at risk for CIN undergoing an intravascular angiographic procedure were randomized to either acetylcysteine or placebo. Roughly 35% of patients in each group had eGFRs between 30-60 ml/min by MDRD and about 5% in each group had an eGFR of less than 30 ml/min.


The acetylcysteine group received 1200 mg of oral acetylcysteine q12 hours before and after the contrast load. The use of normal saline at a rate of 1 mL/kg per hour, from 6 to 12 hours before to 6 to 12 hours after angiography, was strongly recommended but changes in the total volume or speed of administration were permitted. In the end, 94% of patients received normal saline in both the treatment and placebo groups at some dose with 47% in both groups receiving exactly the recommended dose.


There was no difference between the treatment and placebo arms in terms of the primary outcome of CIN (defined as a 25% elevation of serum creatinine above baseline between 48 and 96 hours after angiography) or the secondary composite outcome of need for dialysis or death in the intention to treat analysis. Approximately 13% of patients in each group experienced CIN with 2% and 0.3% experiencing death or need for dialysis respectively.


This looks to be the end of the acetylcysteine in CIN argument. It's hard to imagine a larger or more well performed study. There was no hint in the subgoups that patients with more severe chronic kidney disease or diabetic nephropathy might benefit. This is clear evidence that we should drop acetylcysteine from the CIN prophylaxis tool kit and focus our efforts elsewhere.


Addendum:


So, the argument is not over. As noted in the commentary, the ACT results are limited by the low resolution of the description of the fluid strategies between groups leaving the bias door open a crack. The issue of generalizability is also raised in patients with advanced chronic kidney disease (and other high risk groups) given their relatively low numbers in the study. Although the ACT subgroup analyses give no hint that NAC might be of benefit in these groups the hypothesis of benefit for them remains to be adequately tested.

Super Bowl Sunday Contrast Nephropathy Post

How common is contrast nephropathy in the general outpatient setting? When you send one of your outpatients to get a CT scan with iv contrast, what is the real risk?  Much of the randomized, controlled trial data on contrast nephropathy comes from INPATIENTS undergoing angiography or other procedures, often focusing on those with the greatest perceived risk of contrast nephropathy (e.g., diabetics, those with underlying CKD to begin with) in order to generate enough meaningful datapoints for analysis. An article by Mitchell et al in this month's CJASN reports the risk of contrast nephropathy in a mixed OUTPATIENT group--and finds that the risk of contrast nephropathy may actually be higher than what has generally been appreciated.

One big caveat, in my mind: even though the article is entitled, "Incidence of contrast-induced nephropathy after contrast-enhanced computed tomography in the outpatient setting," the phrase "outpatient setting" should be examined a little more closely. Their patient population wasn't exactly "outpatient" as we normally think of it; it was all the patients entering a large, academic emergency department who underwent CT scans with iv contrast. One could argue that by the very nature of their showing up in an emergency department (as opposed to their primary care physician office) they could be seen as "sicker" and therefore more susceptible to the effects of iv contrast use. Nonetheless, the researchers found that the incidence of contrast nephropathy (defined as an increase in serum creatinine greater than 0.5 mg/dL or greater than 25% of baseline between 2-7 days after contrast administration) was 11% (70 out of 633 patients receiving iv contrast), which is quite high. Not surprisingly, the development of CIN was also associated with an increased risk for developing severe renal failure and death.

There's still a lot of controversy about contrast nephropathy and just how significant it is--after all, the vast majority of patients see a return of their creatinine to baseline levels--but perhaps the knowledge that over 10% of patients in an emergency department show evidence of renal injury with a simple CT scan with iv contrast should make us think twice about ordering this test so frequently. 

There are bicarbonate doubters amongst you...

Final results of last week's Renal Fellow Network poll:

There seemed to be general agreement that volume expansion with isotonic saline is definitely effective, and many individuals also selected the isotonic saline + N-acetylcysteine combination.  Interestingly, only 23% advocated using a sodium bicarbonate-based regimen (6% iv bicarb alone + 17% iv bicarb + oral NAC) despite its superiority as demonstrated in the oft-cited 2004 JAMA paper in which patients were randomized to receive either an iv bicarbonate or an iv saline regimen, and this recent 2009 AJKD meta-analysis of trials involving sodium bicarbonate for the prevention of contrast nephropathy.  However it is essential to point out that this remains a controversial topic, and hopefully further RCTs will clarify the issue.  

Osmotic Nephrosis

"Osmotic nephrosis" is a term which describes a common form of renal tubular injury in response to hyperosmolar substances. It is especially relevant to iv contrast nephropathy, and was first observed in animals and human patients infused with hypertonic sucrose. The pattern of injury is shown below (images from a 2008 AJKD review by Dickenmann et al): by both light microscopy and electron microscopy, there is vacuolization and swelling, predominantly along the apical membrane of proximal tubular cells.

In addition to iv contrast nephropathy, the osmotic nephrosis pattern of injury can also occur in response to IVIG, mannitol, and hydroxyethyl-starch ("hetastarch"). It is useful a reversible injury, but not always. Vacuolated cells can be seen in other forms of renal injury (e.g., calcineurin inhibitor toxicity, renal clear cell carcinoma, foam cells) which must be differentiated from osmotic nephrosis.

Prophylactic hemodialysis for iv contrast exposure?

The issue comes up all the time: you try and set up one of your ESRD patients for a procedure requiring iv contrast, such as a cardiac catheterization or a PE-protocol CT scan. You get a call from radiology saying that they won't do the test until you, as the representative for the nephrology team, can guarantee the patient will be promptly dialyzed so that the patient will not feel the ill effects of iv contrast exposure. Does this make sense? Is there any data to guide us here? Perhaps more importantly, would it be possible to prevent contrast nephropathy from occurring at all in patients with advanced CKD but not on dialysis?

On the one hand, one might expect dialyzing away contrast to be a beneficial thing. There has been increasing data suggesting that residual renal function--even in patients on dialysis--can be a good thing. iv contrast is definitely of low enough molecular weight to allow it to be efficiently dialyzed.

HOWEVER--most of the actual data runs counter to this position. One particular study by Vogt et al performed a randomized controlled trial in which patients with a baseline Cr greater than 2.3 md/dL were randomized to receive either prophylactic hemodialysis or not immediately after iv contrast exposure. The creatinine was initially lower in the treatment group (as would be anticipated given the effect of dialysis on reducing creatinine levels) but by Day 6 there was no signficant difference between treatment and control groups--and there was even a suggestion that dialysis could be harmful. You can find studies reporting the opposite finding--that is, a potential beneficial effect of prophylactic dialysis on contrast nephropathy, such as this recent study by Lee et al--but in my opinion the suspected mechanism by which iv contrast causes renal injury would cast doubt on such a strategy. iv contrast's toxicity is thought largely due to its high osmolar content, as evidenced by the appearance of intracellular vesicles in the renal tubular epithelium exposed to iv contrast. I would imagine that this tubular damage occurs rapidly after iv contrast exposure, and removal of contrast by dialysis would likely occur after the fact.

My own approach to the dilemma above: agree to dialyze the patient in order to get the test performed promptly. Evaluate the patient after the exam and see whether or not there is any objective reason to dialyze them earlier than their usual schedule would warrant. In rare instances, large amounts of iv contrast can cause volume overload/pulmonary edema. But in the majority of these instances, my suspicion is that prophylactic dialysis either to prevent contrast nephropathy in a patient with CKD or to preserve residual renal function in a patient with ESRD is not useful.

Outside Hospital

At the risk of offending community hospitals out there...I present to you "Outside Hospital", a video I helped make during my residency in Internal Medicine parodying some of the outside hospitals from which we used to receive patient transfers from occasionally...it's all in good fun, and the first bit actually has a reference to contrast nephropathy, so I'm still on-topic. BTW, I'm the guy in the old man wig. Enjoy & have a great weekend.

Trick Question Regarding Contrast-Induced Nephropathy

Trick Question: Which would you imagine to be lower osmolarity between "low osmolar iv contrast" and "iso-osmolar iv contrast"?

The answer is, for historical reasons, "iso-osmolar contrast."

The increasing use of iv contrast agents led to the identification of contrast-induced nephropathy (CIN) as a major cause of in-hospital acute kidney injury. CIN is characterized by an increase in Cr beginning 24-48 hours following dye exposure followed by resolution (in most but not all cases) within the following 3-5 days. It has become apparent that the osmolarity of the iv contrast solution plays a major role in the nephrotoxicity of iv contrast: initial preparations were highly hyperosmolar (1400-1800 mosm/kg) and caused CIN commonly; the "2nd generation" iv contrast compounds were called "low osmolar iv contrast" based on the fact that their osmolarity was substantially lower (500-850 mosm/kg) than that of 1st generation compounds but nonetheless still hyperosmolar when compared with human serum. Currently it is felt that "iso-osmolar iv contrast" agents (e.g., iodixanol--molecular structure below) have the lower risk of CIN.