A Brave New Kayexalate Free World

A guest post by Rory Spiegel (@EMNerd_) who blogs on nihilism and the art of doing nothing at emnerd.com.

There may soon come a time when we witness the death of the much maligned cation-exchange resin, Kayexalate  Unfortunately not for the reasons we hoped. We will not see the use of sodium polystyrene sulfonate fade from use in the modern Emergency Department because of our tireless efforts to remind our Internal Medicine colleagues of its lack of efficacy. Rather its clinically irrelevant place in the management of acute hyperkalemia will now be replaced by a brand new shiny cation-exchange resin that our Nephrologist consultants can use to delay the 3 am dialysis treatment our patient with a potassium of 9 mmol/L desperately requires.

Enter sodium zirconium cyclosilicate (ZS-9), a highly selective cation-exchanger that entraps potassium in the intestinal tract in exchange for sodium and hydrogen. Makers of this medication, ZS Pharma, claim it entraps 10 times as much potassium as the tried and (not-so-)true Kayexalate  A recent article published in the NEJM examines its efficacy in patients presenting with hyperkalemia. Authors, Packham et al, randomized 753 patients presenting with mild hyperkalemia (5.0 to 6.5 mmol/L) to either 1.25g, 2.5g, 5g, 10g, or placebo every 8 hours for 48 hours. Only the patients who responded to ZS-9 during the initial phase and were normokalemic after 48 hours were then randomized to either once daily ZS-9 at the original dose they were randomized to or placebo.  Unfortunately none of the patients who truly concern us were included in this trial. Authors excluded patients if they were receiving dialysis, had diabetic ketoacidosis, had a potassium level of more than 6.5 mmol/L, or a cardiac arrhythmia that required immediate treatment. These are often the patients in which we are asked to perform a trial of Kayexalate therapy in place of definitive dialysis.

For the initial phase of the trial, the authors found a statistically significant difference in their primary endpoint, the between-group difference in the exponential rate of change in the mean serum potassium level during the first 48 hours of treatment, between patients receiving the 2.5g, 5g, and 10g dose when compared to placebo.  At 48 hours, the absolute mean reductions in the 2.5g, 5g, and 10g group were 0.46 mmol/L, 0.54 mmol/L, and 0.73 mmol/L respectively. These differences were statistically significant when compared with a mean reduction of 0.25 mmol/L that was seen in the placebo group. The overall reduction in potassium seemed to be mildly correlated with the extent of hyperkalemia at presentation, but authors only presented the results of this analysis in the group who was administered the 10g dose of ZS-9 (1.1 mmol/L > 5.5 mmol/L, 1.0 mmol/L 5.4 to 5.5 mmol/L, and 0.6 mmol/L < 5.3 mmol/L or less). Additionally patients who received the 5g and 10g doses of ZS-9 during the 15-day maintenance phase, had significantly fewer repeat episodes of hyperkalemia. A second study just published in JAMA by Kosiborod et al also examining the utility of ZS-9 in the acute management of mild hyperkalemia (5.0 to 6.5 mmol/L) confirms these findings. Though in this trial, patients in the first 48 hours were not randomized, but rather all were given a 10g dose every 8 hours, the mean absolute change in serum potassium was comparable to the change observed in the 10g group in the Packham et al trial( −0.7 mmol/L at 24 hours and −1.1 mmol/L at 48 hours). Likewise the severity dependent response was also observed in this second trial.

Although statistically a success, ZS-9 adds very little to the acute management of clinically relevant hyperkalemia. Even the high doses of ZS-9 reduced the potassium level on average by 0.73 mmol/L at 48-hours, nowhere near the efficacy that would allow us to comfortably hold dialysis overnight in the acutely hyperkalemic patient. Interestingly these results are not dissimilar to what Scherr et al discovered in their 1961 investigation into the effects of Kayexalate on serum potassium. In this non-randomized, non-blinded trial the cation-exchange resin lowered patients potassium by a mean of 1.0 mmol/L over the first 24-hours.  Furthermore unlike the Scherr cohort neither of these studies examined oliguric patients or those with a history of ESRD on dialysis. The very patients which most frequently require emergent dialysis for acutely elevated levels of serum potassium.

Neither of these trials possessed the statistical power to definitively assess safety. Though no obvious concerns were demonstrated in this cohort, the rates of intestinal necrosis observed in patients given Kayexalate are far too infrequent to detect if ZS-9 causes similar effects with such a small sample size. While none of the patients in either of these trial experienced a fatal arrhythmia related to their hyperkalemia, the authors’ inclusion and exclusion criteria insured these types of events would be highly unlikely. On a side note, ED nurses will be happy to see that rate of diarrhea following the administration of ZS-9 at 1.9% is far less than what is commonly seen in patients given Kayexalate.

The editorial published alongside the Packham et al paper in the NEJM is entitled, “A New Era in the Management of Hypoerkalemia”. Though ZS-9 may play a role in the long-term management of patients at risk for hyperkalemia, for the acute management of hyperkalemia it seems we will still be arguing with our consultants over the administration of an ineffective exchange resin as a replacement for the definitive dialysis they require. A new era indeed…

“Sodium Zirconium Cyclosilicate in Hyperkalemia” http://www.nejm.org/doi/full/10.1056/NEJMoa1411487

“Effect of Sodium Zirconium Cyclosilicate on Potassium Lowering for 28 Days Among Outpatients With Hyperkalemia The HARMONIZE Randomized Clinical Trial” http://jama.jamanetwork.com/article.aspx?articleid=1936753