Best of 2020: Steroid Poker Save
Editor's note: This blog originally appeared January 14, 2020, and is being shared again as a "Best of 2020" piece. Enjoy!
It began as many cases do: an ill patient, in the ICU, with signs and symptoms across several body systems, yet no clear unifying diagnosis on admission. With things stabilizing, the internal medicine hospital team on which I was serving as hospitalist that week, assumed care of the patient.
As the case unfolded – pulmonary infiltrates that could be hemorrhagic, renal dysfunction with proteinuria – rheumatic diseases rose in the differential. When serologic studies and other data suggested granulomatous polyangiitis(GPA) rather than glomerular basement membrane (GBM) disease or other possibilities such as infection, it seemed the right time to act. And that is when a game of what I call “steroid poker” began.
The medical student on the case offered the opening ‘bid’ of 40 mg of prednisone a day, a choice perhaps a bit on the lower side. This was probably due in part to a recent diatribe I had gone on about the side effects of steroids, and also due to the fact that that was the highest dose of prednisone he had ever seen a patient receive so far in his young career.
The intern suggested upping the ante to 60 mg. We discussed competing aspects of the case: the diagnosis did seem to be GPA, a serious condition that merited serious treatment. However, the patient also had diabetes, and overall seemed to have stabilized and even begun to improve. After discussing the relevant data, we decided to go with “high dose steroids” – meaning 1 mg/kg/day prednisone equivalent – and even though the patient weighed 71 kg, we rounded up our ante to 80 mg/d, and commenced therapy.
Then, other players placed their bets in the game of steroid poker. Renal scoffed at our “low dose” and suggested 500 mg IV on 2 successive days Pulmonary then also suggested “pulse steroids”, but their definition called for 1000 mg IV twice. The day after I left the service, a rheumatology consultant new to the case recommended 1 gm, but for 3 days. All the while, some of the various players wrote notes in the chart that would surely arouse any plaintiff’s attorney within a 20 mile radius – for example suggesting that the particular dose of steroid they recommended was “the dose typically required in cases like this”. Strong statements, with not a smidgeon of supporting medical evidence. Strong recommendations, from practitioners who had surely seen more patients named Wegener than patients who actually had GPA.
And “steroid poker” is hardly limited to patients with vasculitis. How much prednisone should be given to a gout patient in the Emergency Department? If you say 10 mg, I say 15mg; if you say 15mg, I say 20mg, and so on. How about PMR? A flare of SLE? Most rheumatologic diseases seem to lend themselves to “steroid poker”.
What is it about this particular class of medications that causes otherwise thoughtful, sensible, physicians to abandon all semblance of evidence-based practice, and pick doses based on urban legend and hearsay? One seldom encounters such behavior with other medications. I cannot recall hearing people say things such as “you want to give 10 mg of coumadin? Too low… I say we give 25 mg!”.
Eminence based medicine seems to be much more typical when it comes to steroids. Perhaps it relates to the history, the tradition and the mystique that have been associated with steroids. Steroids have been around so long that factors impacting their use have largely been forgotten. Why is 125 mg of IV solumedrol a commonly used dose, instead of, say, a rounder number like 100 mg? Because decades ago, the manufacturer of IV solumedrol decided to package it into vials containing 125 mg – why waste that last 25 mg! Where did the dose of “pulse steroids” originate?
Again, decades ago, experiments on dog models of shock lung suggested a possible benefit of 30 mg/kg of prednisone equivalent. Translated into a ‘typical’ 70 kg human, that ended up as about 2 grams, which came to be used as 1 gram on 2 successive days. As with many aspects of steroid dosing, the idea that “if 2 is good, 3 is better” prevailed by the time 3 days of 1 gram of steroids was shown to have benefit in SLE nephritis – one of the few bits of actual medical evidence on such high doses of steroids.
One aspect of super-high dose steroids that seems to have been mostly forgotten is the possibility of severe reactions, even death, associated with pulse steroid administration. Possibly related to acute electrolyte shifts, this and other side effects tempered the excitement about using pulse steroids years back when such therapy was used for patients with almost all rheumatologic diseases. Indeed, it is the potential toxicities of steroids that we all must keep in mind as we try to balance risk and benefit with these storied agents.
Remember this the next time you walk past a conversation about how to treat a patient with some acute rheumatologic disorder – try to avoid the urge to up the ante in “steroid poker”.