Let’s take a trip way back to 2016. You’re me, and you’re in paramedic school, in the thick of the cardiology section. You’re learning about hypotension and the various ways we can fix it. Hypovolemic hypotension gets 20 cc/kg of normal saline, hemorrhage or not. Sepsis gets 20 cc/kg of normal saline, acidosis or not. Cardiogenic shock gets…you guessed it, normal saline. If you have given this magical elixir of life (heavy sarcasm, for those who can’t see me rolling my eyes) and they’re still hypotensive, you can bust out the pressor. Note that I didn’t say pressors because way back in the Stone Age, we only had one choice: dopamine. After countless hours of practicing “dopamine clocks” to guesstimate drip rates, since IV pumps hadn’t been invented yet, we were turned loose to save the world, armed with nothing but liters upon liters of saline and a single bag of dopamine.
Fast forward a few years, maybe a critical care class or a few Studio modules, and now we’ve learned that there’s a whole world of vasopressors, inotropes, inopressors, and inodilators out there. But the question remains—how do we choose which drug to use for which patient? In this article, I will walk you through my mental schema for pressor and/or inotrope selection in cardiogenic shock. Some of this comes down to stylistic preference, and there are a few “Scientific Wild Ass Guesses” (SWAGs), as Tyler would say, that we have to make in the prehospital environment in the absence of advanced hemodynamic monitoring. However, with a solid understanding of your patient’s pathophysiology and a robust assessment, you can make the right call (most) every time.
In the EMS setting, the most common cardiogenic shock patient is most likely a STEMI. The ultimate goal is to optimize coronary perfusion pressure (CPP)—in other words, the amount of blood flow into the coronary arteries.
You may remember that CPP is the aortic diastolic blood pressure minus left ventricular end-diastolic pressure (LVEDP). What does this mean? Aortic diastolic blood pressure is easy enough. It’s the pressure in the aorta during diastole when the heart feeds itself. LVEDP is a little more complex. LVEDP (measured in the ICU using a Swan-Ganz or pulmonary artery catheter) is the pressure inside the left ventricle at the end of diastole, when the LV is at its fullest. Why do we care about this number though? When the heart is full, it puts pressure on the myocardium, compressing the coronary microvasculature. When it gets too high, it can impair blood flow into the coronaries.
In cardiogenic shock stemming from LV failure (lateral, anterior, septal, or any combination thereof), the LVEDP climbs as the heart isn’t able to eject blood, causing decreased cardiac output and potential hypotension. As a result of the increased LVEDP, CPP falls.
In right-sided shock (inferior STEMIs with +RVI), the LV is underfed due to a poorly functioning RV that can’t pump blood across the pulmonary vasculature. Cardiac output decreases, and the aortic diastolic pressure falls with it. Even though the LVEDP remains low, this decreases the CPP.
So how do we treat it? Some old-school instructors will tell you that you should give IV fluids to increase the preload on the heart (measured by central venous pressure, or CVP) to take advantage of the Frank-Starling effect. However, we have to remember that adding too much preload can be deleterious and worsen contractility. Since these patients are already in acute heart failure courtesy of the STEMI, we can safely assume that we’re past the tipping point of the Frank-Starling curve, and adding more fluid probably won’t help. We can quickly confirm this with a POCUS IVC view, or we can look for JVD and other signs of venous overload.
Instead, we will start by bringing up the diastolic pressure with norepinephrine (Levophed). Norepinephrine is a primarily alpha-1 heavy drug, with some limited beta-1 effects. In other words, it will provide mostly afterload, raising systemic vascular resistance, (SVR), with some mild inotropy (contractility). This will shunt blood back into the coronary arteries, which may be all we need to improve cardiac output. We can check again with POCUS to assess cardiac wall motion, or we can look at capillary refill, mucous membrane color, or any perfusion surrogate of your choosing.
If you’ve gotten the MAP up to a comfortable range (somewhere around a MAP of 80-90 with an NIBP cuff, or a DBP of 70-80 if you are checking a manual or have an arterial line), we need to consider adding an inotrope of some flavor. If you only carry one inotrope, the choice is easy. Low-dose epinephrine is a strong beta-1 drug with minimal alpha-1 effects if you stay below ~0.08 mcg/kg/min. This will aid in contractility and bring your cardiac output up. Often, this works almost too well, and you’ll have to titrate the norepinephrine down a bit to prevent the afterload from getting too high and working against you.
If you also carry inodilators like dobutamine and milrinone, you can get a little fancy. Dobutamine is a synthetic catecholamine, similar to epinephrine, but without the alpha-1 effects. It helps the heart beat harder, and also vasodilates the peripheral vasculature to ease afterload. If we’re already on norepinephrine to raise our SVR, this might not be the best choice for patients with left sided problems. In patients with right sided STEMIs, though, inodilators are a great choice, as they relieve the pulmonary vascular resistance (PVR), allowing the much weaker RV to feed the LV.
Unfortunately, dobutamine doesn’t have an aggressive effect on the PVR, especially when compared to milrinone, a phosphodiesterase type 3 (PDE 3) inhibitor. Put not so simply, milrinone inhibits PDE 3 leading to increased cellular levels of cyclic adenosine monophosphate (cAMP), which unregulates the release of calcium during depolarization. This leads to greatly increased cardiac contractility. Additionally, milrinone acts on the pulmonary vasculature to prevent the breakdown of cyclic guanylyl monophosphate (cGMP), which produces profound vasodilation.
However, milrinone is slow to work, with an onset between five and fifteen minutes. This is less than ideal in a crashing patient, so you’ll have to make the judgment call if you have the time to wait, or should select another inotrope instead.
Okay. That’s all a lot of words, so I’m going to use an analogy to tie it all together. Let’s say your car isn’t functioning properly. It’s kind of running, but not very fast. First, you make sure you’ve got enough gas in the tank and that you’re not running on fumes. This is our norepinephrine to raise DBP. When that doesn’t fix the problem, and you want to go real fast in a hurry, you open your nitrous tank and bleed it into your engine to increase the horsepower. This is our inotrope. If THAT doesn’t work, you roll down to your local tuner shop and get a full body kit installed along with extra nitrous to reduce wind resistance. This is our inodilator.
Put gas in the tank, crank the horsepower, and reduce drag.
NB: if you have vasopressin available to you, you can consider using it as a first-line agent instead of norepinephrine in right-sided problems. Since vasopressor acts on the V1 receptors, notably absent in the pulmonary vasculature, it will raise SVR without a significant jump in PVR.
For further reading, consult The Vasopressor & Inotrope Handbook by Eddy Gutierrez, MD
Bloom, J. E., Chan, W., Kaye, D. M., & Stub, D. (2023). State of Shock: Contemporary Vasopressor and Inotrope Use in Cardiogenic Shock. Journal of the American Heart Association, 12(15), e029787. https://doi.org/10.1161/JAHA.123.029787
Levy, B., Buzon, J., & Kimmoun, A. (2019). Inotropes and vasopressors use in cardiogenic shock: when, which and how much?. Current opinion in critical care, 25(4), 384–390. https://doi.org/10.1097/MCC.0000000000000632
Shankar, A., Gurumurthy, G., Sridharan, L., Gupta, D., Nicholson, W. J., Jaber, W. A., & Vallabhajosyula, S. (2022). A Clinical Update on Vasoactive Medication in the Management of Cardiogenic Shock. Clinical Medicine Insights. Cardiology, 16, 11795468221075064. https://doi.org/10.1177/11795468221075064
