Medicine news

Toxicology Rounds: The Truth About High-Dose Insulin : Emergency Medicine News


high dose insulin, calcium channel blocker poisoning, amlodipine


The causative agent when our poison control center is consulted for cases of calcium channel blocker poisoning (CHB) is almost always amlodipine. Quite often the clinical team will be looking for high dose insulin (HDI) as they call us. After all, HDI is often considered a universal antidote to CCB poisoning, capable of reversing all aspects of the toxicity of these drugs.

This is a serious misconception. Many doctors believe that the effects of HDI mimic those of norepinephrine, increasing the force of myocardial compression and constricting blood vessels. In other words, they think HDI is an inotrope and a vasoconstrictor.

It’s not correct ! HDI is a positive inotrope, but it is not a vasoconstrictor at all. In fact, HDI is a vasodilator. Being an inotrope and a vasodilator, HDI is more like dobutamine than norepinephrine.

HDI is not necessarily a logical first choice for amlodipine overdose as an antidote to reverse hemodynamic instability. Amlodipine is a dihydropyridine, a class of BCCs that includes the now little-used agents nifedipine and felodipine. Compared to non-dihydropyridines such as verapamil and diltiazem, amlodipine has predominantly vasodilating effects on arteries and arterioles and relatively little effect on myocardial compression or the cardiac conduction system. Of course, this selectivity is sometimes lost in overdose and dihydropyridines can impair cardiac function.

But it is striking how often these patients exhibit reflex tachycardia in response to vasodilation and a decrease in systemic vascular resistance. Most commonly, when the heart is viewed at this time with a bedside echocardiogram, it is seen beating at a rate of approximately 120 bpm, with a high ejection fraction often around 70% or more.

The question is: do you really need an inotrope in this situation, with tachycardia and increased ejection fraction, where the main problem is vasodilation? How high do we want the ejection fraction to go? 80%? 90%? 110%? And do you really want to start with HDI, which will dilate the peripheral vessels even more, when vasodilation is the main problem? Or would we be better off with a vasopressor such as norepinephrine, which will provide inotropic support in addition to being a potent vasoconstrictor?

An easy and reproducible way to help sort out what emergency physician Martin Smilkstein, MD many years ago called an infusion salad is the use of bedside ultrasound and echocardiography. (Acad Emergency Med. 1996;3[2]:99.) It’s not complicated and only takes a few seconds because the question asked is simple: are cardiac compression and ejection fraction altered, relatively normal or truly supernormal? It would make sense to start specific treatment with HDI if cardiac function is depressed, perhaps with a small dose of vasopressor to help offset the expected increase in vasodilation.

On the other hand, it does not make sense to whip the myocardium even more, and norepinephrine or another vasopressor with α1 would be a good first choice for maintaining blood pressure if the ejection fraction is above 50-60%. It might be reasonable to add HDI after the pressor to support myocardial function because the heart works harder pushing blood against increased systemic vascular resistance.

A recently published article addresses another important point regarding CCB poisoning.

The known unknown: non-cardiogenic pulmonary edema in amlodipine poisoning, a cohort study

Lindeman E, Alebring J, et al.

Clin Toxicol (Phila).


This prospective observational study from the Swedish Poisons Information Center in Stockholm investigated patients with an amlodipine overdose who required treatment with an inotrope or vasopressor. They found that 47% (9/19) of their cohort developed noncardiogenic pulmonary edema (NCPE) and concluded that NCPE was a “common finding” in amlodipine poisonings.

Not so fast. It was a very select group. A patient had to have an amlodipine concentration greater than 0.25 mcg/mL to be eligible for the study. This is about 10 to 25 times the therapeutic level. We certainly don’t notice NCPE in nearly half of the amlodipine overdose cases we see, but it’s important to recognize that this side effect can occur. The reason is interesting.

Amlodipine is a vasodilator, but it has a much greater dilating effect on pre-capillary vessels, not so much on post-capillary vessels. This is why dihydropyridines frequently cause pedal edema. Improved blood flow increases hydrostatic pressure in the alveolar capillaries of the lungs. This, in addition to loss of autoregulation, leads to NCPE and ventilation/perfusion mismatch. Too aggressive volume repletion and an excessive increase in cardiac output can be detrimental in this situation. NCPE often becomes evident in these cases even when cardiac function appears to be improving.

The authors raise the possibility that HDI treatment may increase the risk of NCPE because it may increase alveolar blood flow and pulmonary capillary hydrostatic pressure. I’m not entirely convinced, but I certainly agree with the authors that fear of a later NCPE should not prompt the clinician to avoid HDI in amlodipine toxicity when indicated. However, it would be good to know more precisely when exactly the HDI should be used.

Bedside echocardiography can provide crucial information in this regard. Fear of a later NCPE should not lead the clinician to avoid HDI in amlodipine toxicity when indicated without knowing exactly how the heart works in these cases, and we will often find ourselves stumbling in the black when we try to determine the optimal treatment.

Dr. Gussowis a volunteer attending physician at the John H. Stroger Hospital of Cook County in Chicago, assistant professor of emergency medicine at Rush Medical College, consultant at the Illinois Poison Center, and senior lecturer in emergency medicine at the University of the Illinois Medical Center in Chicago. Follow him on Twitter@poisonreview, and read his past columns on