Introduction with two cases
Case #1
The first case of steroid use for refractory hypoglycemia at Genius General occurred several years ago. A patient developed numerous episodes of hypoglycemia requiring large volumes of IV dextrose. In efforts to avoid recurrent hypoglycemia, 125 mg IV methylprednisolone was given. Hypoglycemia resolved immediately, but the patient subsequently developed moderate hyperglycemia in the 300-400 mg/dL range (without diabetic ketoacidosis). The patient did fine clinically. However, this case suggested that steroid with a shorter half-life (e.g. IV hydrocortisone) might be preferable to facilitate titration and avoid prolonged hyperglycemia.
Case #2
A 60-year-old woman with type-II diabetes was brought to the hospital following a suicide attempt with glargine insulin. Before arrival, the patient was conscious and treated with oral carbohydrate. Initially she received aggressive IV dextrose (several ampules of D50W plus an infusion of D10W at 200 ml/hr). However, her glucose remained below 30 mg/dL. Fortunately, she remained only mildly symptomatic with a glucose in the 20-30 mg/dL range (perhaps due to adequate intracellular glucose).
Based on the failure of IV dextrose, 100 mg IV hydrocortisone Q6hr was initiated. Immediately after starting steroid, her glucose rose to a safe level. The D10W infusion was reduced from 200 ml/hr to 100 ml/hr. Over the next two days, steroid and D10W infusions were gradually weaned off. Her recovery was unremarkable, without recurrence of hypoglycemia.
She did receive one dose of IV glucagon along with the first dose of hydrocortisone. This could muddy the waters a bit. However, given the short duration of glucagon (typically lasting 15-20 minutes), it is extremely doubtful that a single dose could explain her sustained improvement.
Textbook treatment for insulin poisoning
Severe insulin overdose usually occurs as a suicide attempt, but can also result from medication error. Conventional therapy focuses on giving tons of intravenous glucose. This generally works, but it can get messy:
- In severe cases, central line placement is needed to administer D20W or D50W infusion.
- Breakthrough hypoglycemia can occur, with a risk of seizure and brain injury.
- Large-volume infusion and insulin effects can cause volume overload and electrolyte derangement (e.g. hypokalemia, hyponatremia, hypomagnesemia, hypophosphatemia).
- Hepatic injury may occur, as hepatocytes are overloaded with the influx of glucose (Warriner 2012).
Some case reports have described surgically excising the skin around the site of glargine injection to reduce the duration of hypoglycemia. Although potentially effective, this requires an emergent surgical procedure to remove normal tissue.
Theory behind steroid therapy
Basic physiology
Let's start by considering a COPD patient with borderline glycemic control who is treated with prednisone. Prednisone causes insulin resistance and hyperglycemia, which in turn requires insulin. Overall, insulin must be balanced with steroid:
This balance is common knowledge among anyone prescribing steroid. Endocrinologists take this a step further, exploring how to balance steroid and insulin more precisely. For example, the pharmacokinetics of prednisone are similar to the pharmacokinetics of NPH insulin, such that administering both simultaneously might be a sensible strategy to maintain balance (Grommesh 2016).
Use of steroid as an antidote to insulin poisoning involves using the same exact physiology in reverse:
Pharmacokinetics
Although any steroid would work for this scenario, a short-acting IV steroid has the best pharmacokinetics (e.g. IV hydrocortisone). This acts rapidly, with a biological effect of roughly 8-12 hours. A reasonable starting dose might be ~100 mg IV, with subsequent dose and interval titrated to effect based on glucose trends (e.g. steroid should be weaned off as the insulin overdose resolves).
Occasional patients present with hypoglycemia, improve, and then refuse admission to the hospital. Suicidal patients may be held against their will, but patients with unintentional insulin overdose may leave against medical advice. For a patient with moderate hypoglycemia leaving against medical advice, a longer-acting steroid could be useful (e.g. a single dose of dexamethasone would reduce insulin sensitivity for over a day).
Risk-benefit estimation
Risks of low-dose steroid
Low-dose steroid is generally a fairly safe therapy (e.g. 100 mg hydrocortisone q6hr, which is equivalent to 100 mg prednisone daily). The most common side-effect of steroid is hyperglycemia, which isn't a problem here. The risk of increased infection at this dose range is nonexistent or minimal, particularly for treatment lasting only a couple of days. The most significant side-effect might be delirium, which remains rare.
Particularly among patients with type-I diabetes, over-aggressive use of steroid could theoretically precipitate diabetic ketoacidosis (by tipping the balance too far). This might be avoided by using the following measures:
- DKA would be expected to occur if a high enough dose of steroid was used such that intravenous dextrose could be stopped entirely (suggesting absence of biological insulin activity). In order to avoid DKA, steroid monotherapy is not recommended. The goal of steroid isn't to avoid IV glucose entirely, but rather to reduce the amount of IV glucose which is required to a manageable amount. For example, steroid might allow the dose of IV glucose to be decreased to ~100 ml/hr of D10W, an amount which can easily be given via peripheral IV without much risk of volume overload.
- Among patient with Type-I diabetes, insulin treatment must be resumed as soon as the insulin overdose resolves (this is true regardless of whether steroid is used).
Benefit of low-dose steroid
Since steroid causes insulin resistance, it has the capacity to function as a true antidote. Steroid has the greatest potential benefit for patients who respond poorly to IV glucose (e.g. with refractory hypoglycemia or the requirement for massive quantities of glucose). For such patients, steroid offers:
- Avoidance of the need to place a central line to infuse D20W or D50W.
- Avoidance of fluid overload due to large-volume dextrose infusion.
- Avoidance of repeated hypoglycemic episodes which may cause seizure and brain injury.
A significant amount of literature supports the use of octreotide for sulfonylurea-induced hypoglycemia. Although octreotide is clearly a first-line agent in this situation, steroid could be used if octreotide isn't available. One advantage of steroid is that it is universally available and rapidly accessible at any medical center.
Literature review
There is only one report of steroid use for insulin poisoning (Tariq 2018). These authors describe a woman who presented after overdose of nearly 10,000 units of insulin glargine. Despite ICU admission and IV dextrose along with glucagon and octreotide, she continued having episodes of hypoglycemia (1). Steroid was started on ICU day #2 and seemed to correlate with improvement:
Like many toxicology subjects, refractory insulin poisoning is extraordinarily rare. Most insulin overdoses will respond well to conventional treatment with intravenous dextrose. Thus, it is unlikely that a strong evidence basis for steroid in refractory insulin poisoning will emerge in the immediate future.
Conceptual framework
Below is a general concept of how to approach insulin poisoning. There is nearly no direct evidence to back this up, so it's merely a general idea of how this might be approached. This will obviously need to be adjusted on a patient-by-patient basis. Nonetheless, this framework may be helpful if you're admitting several patients simultaneously and need a quick plan (2).
- Severe insulin poisoning is a rare toxologic presentation following suicide attempt or insulin dosing error.
- Most cases of insulin poisoning will respond to conventional therapy with IV dextrose, but severe cases can be refractory.
- Steroid reduces insulin-sensitivity. The need to balance steroid dose versus insulin dose is widely recognized when administering high-dose steroid.
- IV hydrocortisone may be a useful antidote for severe insulin poisoning which is refractory to IV dextrose or requiring cumbersome quantities of IV dextrose.
- The goal of steroid in this context isn't to avoid IV dextrose entirely, but merely to reduce the amount of dextrose which is required to a manageable amount.
Notes
- Octreotide is occasionally used in cases of glargine overdose. The concept is that hyperglycemia caused by IV dextrose administration can stimulate endogenous insulin release, increasing the risk of rebound hypoglycemia. Octreotide is intended here to impair endogenous insulin release. Whether this works is debatable. In this case, it didn't.
- As Louis Pasteur would say, chance favors the prepared mind. I find it useful to create treatment algorithms on a quiet Sunday afternoon with a pot of coffee. A few years later when I encounter this situation in the middle of a very busy ICU with multiple crashing patients, I have a template to work off. This cognitively unloads me, allowing stabilization of several patients simultaneously. Once things have settled down, I can re-evaluate the patient and make adjustments.
Image Credits: Seesaw
- Pulmcrit wee: The cutoff razor - April 15, 2024
- PulmCrit Blogitorial – Use of ECGs for management of (sub)massive PE - March 24, 2024
- PulmCrit Wee: Propofol induced eyelid opening apraxia – the struggle is real - March 20, 2024
as is often the case, discussing this very rare potential situation allows a consideration of the broader picture and other thoughts (octreotide, central lines, avoiding such, which dexrose (including ben and jerry’s), which steroid, why.
thank you, josh
Interesting, but before everyone starts giving steroids to everyone with insulin overdose, please remember we have an antidote that is 99.99% effective and is dextrose is still required despite steroid administration. The term “cumbersome quantities of IV dextrose” is not one I generally use but when you get to 1000ml/hr of D25 that does seem like a lot. My takehome: Steroids may be a useful adjuvant therapy in rare cases as shown in two case reports. Nice explanation, just begging people not to over generalize.
Agree, IV dextrose is certainly 1st line therapy, along with feeding the patient if possible.
One question is what to do when a patient fails to respond to D10W. Your options are placement of a central line to give more concentrated dextrose versus hydrocortisone. It’s possible that hydrocortisone could be safer, easier, and cheaper than placing a central line (noting especially that we’re only usually talking about 1-2 days of hydrocortisone therapy). More evidence would be needed to know for sure unfortunately.
Agreed, but I’ve rarely seen an intensivist try to warn of the dangers of central lines. In the setting of coingestion or hypotension, a central line might be indicated for other reasons. The cost argument seem irrelevant when the patient is taking up an ICU bed for frequent checks. Would you still advocate steroids in a patient with central access? Additionally, medicine has a femme fatale attraction to steroids as a solution to a range of problems and so I tend to be cautious. An interesting opportunity for further research.
Touche, central lines are very safe especially with ultrasound. I was just trying to contextualize the side effects of hydrocortisone a bit. The best argument against a central line might be a humanistic one – for an anxious, depressed patient who just tried to commit suicide, if I can spare them a painful scary procedure then that’s probably a reasonable goal. I don’t think it’s possible to make any general recommendation pro/con with regards to steroid in a patient with central line already. If patient is doing well on a D20 gtt then fine – just keep it running, no… Read more »
Just wanted to add in here, because recently had to give hydrocortisone for a pt with very refractory hypoglycemia. Pt was in heart failure so avoiding any extra fluids if possible. Also the hospital was running out of Amps of D50 and glucagon.
Left pondering the utility of this in severe CCB/BB OD’s when using HIET and our dextrose requirements start creeping up. Thanks for the thought provoking post as always!
I don’t think there is any role for steroid in severe CCB/BBl OD that is being treated with high-dose insulin. The steroid would cause insulin-resistance, which would make the high-dose insulin less effective — thus potentially exacerbating the CCB/Bbl poisoning.
First of all, Thank you for this site. Is a very nice place to hide myself.
Do you think there’s any place to glucagon in the severe insulin poisoned in the ICU or just an option to extrahospitalary world ?