Let’s finish off our tour of the vasopressor armamentarium by considering a set of drugs that aren’t pressors, but are often used like they were: corticosteroids.
Corticosteroids (more specifically glucocorticoids) are sophisticated hormones that play a number of diverse roles in the body. If they seem complex, you’re not missing something—they are complex.
Infectious disease folks emphasize their immunosuppression. Rheumatologists appreciate their anti-inflammatory effects. Endocrinologists remark upon their tendency to raise serum glucose by stimulating gluconeogenesis. Surgeons complain that they impair wound healing. And the primary care crowd notice that they make people feel unpleasantly antsy.
In a global sense, many of these effects can be understood through the basic truism that steroids are stress hormones. While we’re all familiar with the acute “fight-or-flight” response in which epinephrine dominates, steroids are the second wind of that process, ensuring the prolonged stress response we need to finish the fight. Consequently, they optimize the body for continued stress, yet inhibit the long-term, home-building activities—such as healing and immune response—that are unimportant while we flee from saber-tooth tigers.
Steroids are important in a number of diseases, but their role in the treatment of shock is a unique one, only really encountered in one other situation: adrenal crisis.
Consider the patient with Addison’s disease, or chronically on high-dose steroids for other reasons. In order to maintain physiologic steroid levels, they are dependent on extrinsic steroid supplementation. But then, they either miss a dose, or suffer an acute illness. Suddenly, they’re steroid deficient.
(Why does the Addison’s patient suffer adrenal insufficiency when they come down with the flu or break their leg, even if they’re dutifully taking their normal dose? Remember that this is a stress hormone. When faced with acute illness, the body must be able to upregulate cortisol production as part of its compensatory response. When you’re sick, the “normal” dose becomes insufficient.)
What happens to these people? Their bodies basically become unwound. Adrenal crisis is characterized by widespread organ dysfunction, including unexplained pain, GI symptoms, obtundation—and most importantly, severe, refractory hypotension.
Corticosteroids do not induce the compensatory shock response of vasoconstriction and cardiac stimulation. The catecholamines do that. However, steroids play a permissive or enabling role in catecholamine responsiveness. Without their presence, all the epinephrine in the world won’t be able to do its job.
The classic adrenal crisis patient, therefore, is the one who turns up in the ED altered or unresponsive, with unexplained cardiovascular collapse. Yet to your consternation, fluid resuscitation proves inadequate, and their response to pressors is inexplicably poor. You dump in the norepinephrine, add on other agents, and their pressure still doesn’t come up.
Then someone has the clever idea of trying steroids. Maybe they dug through some records and discovered that the patient has Addison’s, or has been on prolonged steroid therapy for a rheumatologic condition. Or maybe it’s just a last-ditch effort because the patient is dying. They push some IV steroids, and before long, the patient rallies and begins to revive. It is the classic essential therapy that will readily save a life if administered, and if missed leaves little room for a good outcome. If they need steroids, they need steroids.
The relative adrenal insufficiency of critical illness
So now consider the patient without Addison’s, on no chronic suppressive steroid therapy, and without any reason to have adrenal insufficiency on this particular day.
They are acutely ill, however. Most classically, they have sepsis. And they are hypotensive.
Here’s the issue: a significant portion of critically ill patients will develop a temporary relative adrenal insufficiency, even if they previously had normal adrenal function. This is especially common in the septic population, where observational studies have demonstrated depressed cortisol levels in as many as 60% of the sickest patients.
Presumably this is because these patients have adrenals that can produce enough cortisol for everyday function, but not enough to meet the increased demands of illness. Probably the pathophysiology is multifactorial. (For example, the sedative etomidate, often used for intubation, has been shown to temporarily suppress adrenal function.) But the fact is that many of our sick ICU patients will fail to mount an adequate adrenal response.
Can we figure out which ones? Sure. We can draw a random cortisol level—but these are often neither sensitive nor specific for determining who will be a “steroid responder.” Better is to perform an ACTH stimulation test, where we check a baseline cortisol, administer the pituitary hormone ACTH (aka Cosyntropin) and then redraw cortisol levels 30–60 minutes later, looking for an appropriate increase. While this can be helpful at extreme values, again the results are often indeterminate. Besides, all of these tests may be too slow in the crashing patient, and in truth none have been shown to be very predictive of response to steroid administration.
The solution: if the patient is sufficiently ill, just treat them empirically. In the shocked patient, unresponsive to fluids and on escalating doses of pressors, whilst you throw the kitchen sink at them, eventually you’ll add some steroids to that sink. “Nobody in the ICU should die with steroids,” goes the old saw.
Mostly it doesn’t work. But just like in the Addison’s patient, you’d hate to miss it.
The literature on this topic is a deep dive for another day, but overall is rather unclear. Overall, it suggests that steroids may indeed help to reverse shock in some patients, but we’ve had a hard time proving that it saves lives. Then again, saving lives is hard.
A typical dose is around 200mg of hydrocortisone per day, usually as 50mg q6h. Why hydrocortisone? Hydrocortisone is basically cortisol, our body’s chief intrinsic corticosteroid. Other steroids are available, of course. But the alternatives, such as prednisone or dexamethasone, have a stronger anti-inflammatory effect, and while this may have some relevance in sepsis, it is not our chief objective. Hydrocortisone also offers the most potent mineralocorticoid effects, which are likely valuable in shock as well. (Remember, mineralocorticoids like aldosterone are required to prevent salt-wasting and halt diuresis.)
If the patient seems to respond to the steroid trial, for instance with a reduction in their pressor requirement, then continue it until they’re out of the woods. If there is no response within a day or two, it’s probably safe to stop. Either way, a taper is probably appropriate, either brief or prolonged depending on the duration of therapy.
This is all quite separate from the patient on chronic steroid therapy (including the Addisonians). For those patients, their usual dose is their baseline, and they’ll still need a “stress dose” on top of that. Do not fear this. Every so often, a patient is admitted with an infectious illness, and some well-meaning house officer holds their long-standing steroids for fear of immunosuppression. But these people need their steroids; their chronic dose isn’t supplementation, it’s physiologic. Holding it just throws them into an adrenal crisis, which is a great way to kill them.
This completes our exploration of the world of pressors. Questions, reflections, or objections? Add them to the comments or send us an email!