I realize that many practicing emergency physicians aren't clamoring to use sodium polystyrene sulfonate (SPS; brand name Kayexalate) to treat hyperkalemia. The lack of efficacy for this particular treatment has been well documented time and again. Indeed, Anand Swaminathan, MD, took dead aim at it in an excellent EMN column not all that long ago, walking us through the history and demonstrating clearly that SPS is, at best, useless and, at worst, harmful. (EMN. 2015[37:1; http://bit.ly/2Y2i9jy.)
Why am I beating a dead horse? For three really good reasons:
- Emergency physicians and APPs are still asked to give this medication when we admit folks to inpatient services or it is ordered by our inpatient colleagues while patients are boarding in the ED.
- When the ED is busy, we don't always have time to engage in a stimulating repartee in which we cite all the relevant literature highlighting SPS's uselessness.
- Hot off the presses is a new study that should provide the last nail in the Kayexalate Coffin™.
The study in question was performed by the good J. Ariana Noel, MD, et al., and was released for our reading pleasure in June. (JAMA Intern Med. 2019 Jun 10. doi: 10.1001/jamainternmed.2019.0631.) They conducted a retrospective matched cohort study of patients 66 or older who received SPS between 2003 and 2015 in Ontario, Canada. They were able to identify more than 27,000 individuals who were given SPS using their fancy national and provincial health databases and then matched 20,020 of them with a cohort of 20,020 patients who did not receive it. The mean age was 78, and the population was nearly 55 percent male.
They noted that this matching was performed using a high-dimensional propensity score that draws from more than 200 variables and can potentially limit bias when compared with propensity matching techniques (though it still feels like they are overcompensating because they know in their hearts that we'll never love this methodology as much as we love our randomized controlled trials).
The primary outcome was hospitalization or an emergency department visit in the 30 days after being given SPS for a composite of adverse GI events, which the authors broadly defined as intestinal ischemia/thrombosis, GI ulceration/perforation, or resection/ostomy. I don't love the composite outcomes in general, but I think we can all agree that we don't want to cause any of our patients' bowels to die, burst, or need to be cut out of their body, so I'm going to give them a little slack here.
So what did they find? Patients who received SPS were at nearly double the risk of hospitalization from adverse GI events within 30 days of starting the medication (hazard ratio, 1.94; 95% CI, 1.10-3.41). The overall event rates were small, 0.2 percent vs. 0.1 percent, but the effect was robust and persisted across almost all of their subgroup analyses. It didn't matter what the patients' kidney functions were, what comorbidities they had, how elevated their potassium was, what medications they were taking, or even whether they lived in an urban or rural location. The risk of badness stayed consistent and significant.
The authors even got fancy and compared the risks of developing cholecystitis, diverticulitis, and appendicitis among patients who received SPS and those who did not, and found no association between these outcomes and the drug's use. This negative outcome further supports the hypothesis that the increased risk of adverse GI events measured in the primary outcome was because of a real harm caused by the SPS itself.
Some caveats are important to note. First, the statistical significance in the composite primary outcome was driven by the strong association between SPS administration and intestinal ischemia/thrombosis (HR: 4.92; 95% CI, 1.09-22.25), while no statistically significant increased risk of GI ulceration/perforation or resection/ostomy was seen. That said, thrombosing the blood supply to one's intestines and causing ischemia is a pretty nasty outcome, so putting elderly patients at nearly a fivefold risk of this badness is clearly no good.
This study is also limited by all the issues that plague retrospective cohort studies, and the authors admitted that they “lacked any pathological data that could suggest a causal association between sodium polystyrene sulfonate use and adverse GI events.” So they see a robust association between giving SPS and causing patients harm, but they don't have a great hypothesis about why this is happening.
Some argue that it's not the SPS that was the bad guy here, but the sorbitol in which it used to be administered that was to blame. The authors made a key observation that “the risk remained consistent before and after 2009, the year the FDA issued a warning about concurrent administration of sodium polystyrene sulfonate and sorbitol, suggesting that sorbitol alone was not driving the risk of GI injury.”
We already knew that SPS was ineffective, and we suspected, based on some very small studies and reports, that it could cause harm. Now we have a population-level study demonstrating that SPS has an extremely robust association with hospitalization for a nasty set of adverse events, driven in large part by a nearly fivefold risk of intestinal ischemia or thrombosis. So the next time it feels like you don't have time to argue against SPS on the merits, print a copy of this excellent study by Dr. Noel and use it as an academic (and if need be, physical) shield to protect your patients from this useless and harmful relic
Share this article on Twitter and Facebook.
Access the links in EMN by reading this on our website or in our free iPad app, both available at www.EM-News.com.
Comments? Write to us at [email protected].
Dr. Rundeis the assistant residency director and an assistant professor of emergency medicine at the University of Iowa Hospitals and Clinics, where he serves as co-director for the associate fellowship in medical education. He creates content for and is a member of the editorial board forwww.TheNNT.com, and is a content contributor forwww.MDCalc.com. Follow him on[email protected]_MC, and read his past articles athttp://bit.ly/EMN-MythsinEM.