We need to talk about activated charcoal. Why? Because we sometimes give it when we probably shouldn’t, and we sometimes don’t when we probably should. I’ll explain what I mean by this. But first, some background.
What is activated charcoal?
Activated charcoal (AC) is a gritty black substance that, when administered to a poisoned patient, gives the treating physician a sense of having done something useful. It is made by the controlled pyrolysis of carbon-based compounds such as peat, coconut shells, bamboo and other types of wood. The process of “activation” (by physical or chemical methods that will interest exactly none of you) results in a product of extremely small particle size and exceptional porosity. If you were to examine its appearance using transmission electron microscopy, you’d see a mix of pentagons and hexagons, something like this:1,2
The result of activation is a compound with an enormous surface area, ranging from 500 to 3000 m2 per gram, depending on the product. To contextualize this, an American football field has a surface area of a little over 4400 m2, so giving a standard 50 grams of charcoal is like instilling into the patient a dozen or so gritty, black football fields.
Why do we use it?
We exploit the enormous surface area of AC to adsorb poisons (“ay-dee sorb”, as the pedants like to say), generally in one of two settings. Most commonly, we give a large single dose (50 to 100 grams) shortly after an overdose, in the hope that AC will bind the poison and minimize its systemic absorption. Less commonly, we give multiple smaller doses (perhaps 12.5 to 25 grams every few hours), with the intent of enhancing the elimination of a limited number of drugs (principally dapsone, carbamazepine, methylxanthines, and phenobarbital.) So while single-dose activated charcoal (SDAC) is a means of GI decontamination (just like gastric-almost-no-one-does-that-anymore-lavage and whole bowel-they-should-have-called-it-GoHeavily®-irrigation), multi-dose activated charcoal (MDAC) is more akin to dialysis, with the gut as the dialysis membrane. This is nicely demonstrated in this Reagan-era study in which volunteers were given IV aminophylline with and without charcoal (140 grams in divided doses over 12 hours). Take a look at the figure below to see how MDAC hastened theophylline elimination.3
This remainder of the post will focus on SDAC.
Where’s the evidence?
Evidence? What do you think this is, cardiology?
Like most interventions in toxicology, we give SDAC because we think it might help, not because we know it will. I’ve written about this in more detail before4, but here’s a summary of what we know.
In vitro and animal studies make clear that AC binds a wide range of drugs and other poisons. (Note that there are some poisons we don’t give SDAC for — metals, hydrocarbons, caustics, and toxic alcohols, for example — either because they don’t adsorb well to charcoal or because the prospect of SDAC-related vomiting will worsen matters). More than a hundred volunteer studies have been published, most of them using a crossover design. When at least 50 grams of SDAC is given, the average reduction in drug absorption is 47.3% at 30 minutes, 40.1% at 60 minutes and 16.5% at 120 minutes. Don’t generalize those numbers to your patients too freely, though, because these studies involve healthy, cooperative people who took a single drug in a sub-toxic dose in a controlled setting. In other words, not patients.
As for timing, here’s what volunteer studies tell us about the reduction in area under the curve (AUC) vs. time to SDAC (see figure below).5 Setting aside their limitations, you get a pretty clear sense that the intervention is time-sensitive. This is not surprising of course, but more on that later.
Two randomized controlled trials of SDAC in acutely poisoned patients have been published, with neither showing a benefit.6,7 However, I wouldn’t be too quick to infer that charcoal has no role in the management of poisoning based upon these studies, because neither enrolled the sort of subject we’d most expect to benefit: a patient who presents shortly after an ingestion that is likely to result in serious toxicity or death. I doubt we’ll ever see such a study, chiefly because a control group in such patients would be hard to justify.
Evidence for the real-world utility of SDAC comes from a series of population modeling studies from Down Under by @Geoff_Isbister, @SDuffull and colleagues. Some of these focus on absorption, with SDAC given within 4 hours reducing citalopram absorption by 22%,8 quetiapine by 35%,9 sertraline by 27%,10 escitalopram by 31%,11 and venlafaxine by 29%.12 Others, however, examine more clinically relevant outcomes. For example, patients given SDAC after acetaminophen overdose were less likely to require N-acetylcysteine,13 and those given SDAC after citalopram overdose were 60% less likely to display high‐risk ECG abnormalities.14 These studies underscore a point I never grow tired of making: there’s a lot more to “evidence” than randomized trials. While not perfect, these modeling studies represent the best available evidence about the utility of SDAC after acute poisoning.
What’s the downside of giving charcoal?
Fair question. Charcoal is relatively safe, especially if the patient can swallow safely or a nasogastric tube is in place. Aspiration and its sequelae are the main concerns.15 While these are thankfully uncommon, the possibility needs to figure into decisions about whether to give SDAC or not. And that brings us to what we need to talk about.
Should I give activated charcoal or not?
As with all clinical questions, the answer is “It depends.” Different toxicologists might give you different answers for a given patient (although disagreement is less likely with SDAC than with other forms of GI decontamination.16) But let’s step back and ask ourselves what it is we’re really trying to accomplish when we give a patient charcoal.
Yes, we’re trying to reduce systemic absorption. But what we’re really trying to do is favorably influence the patient’s outcome – maybe save them an intubation, reduce their need for antidote therapy, shorten their hospital stay, or even (dare I say it?) prevent mortality. With almost no evidence to show that SDAC actually does these things, we’re relegated to relying on common sense.
Keeping in mind that most poisoned patients will do just fine with supportive care alone, let’s consider some factors that common sense dictates are likely to cumulatively increase the appropriateness of giving SDAC:
- Recency of ingestion: Preventing absorption is intuitively a time-sensitive intervention: charcoal is more likely to make a difference when given at one hour than at six. But timing is only one consideration, and the idea that there’s some magical 1- or 2-hour window beyond which charcoal should not be given is the product of not thinking about the issue quite enough. This is important because only about 16% of acutely poisoned patients present to hospital within one hour, and most present later than two hours.17
The flip side of this point is that patients shouldn’t automatically get SDAC just because they were good enough to present early. I make this point because I often get called about patients who’ve been given charcoal, even though by history a good outcome without charcoal is a virtual certainty. Which brings me to my next point . . . - Likelihood of a bad outcome: You’re not going to give SDAC to the patient who swallows 100 omeprazole tablets, even if he downs them right in front of you. The same goes for the toddler who took a few of Grandma’s alendronate. A good outcome is preordained in these patients, and charcoal stands no chance of improving things.
On the other hand, when you’re dealing with poisons that tend to kill people, your inclination to give charcoal ASAP should go way, way up. This is true even at 3 or 4 hours after ingestion, maybe even later. Some examples of these include calcium channel blockers, colchicine, bupropion, ASA and cyclic antidepressants. And obviously the larger the ingestion, the more consideration you should give to SDAC. - Lack of a specific antidote: This is a variation on the preceding concept. All other factors being equal, the case for SDAC will be weaker in a patient with an acute overdose of acetaminophen, opioids or sulfonylureas because these drugs have other antidotes that are reasonably safe and effective, making a bad outcome less likely. Conversely, a lack of other effective therapies (and here I’ll invoke ASA, bupropion, cyclic antidepressants, and colchicine again) makes SDAC an even more important consideration.
- Modified-release formulations: The inclination to give charcoal should also go up, at least a little, with modified release formulations. Yes, 6000 mg of verapamil is bad. But SDAC is more likely to make a dent in the ingestion if it’s 50 tablets of verapamil 120 mg SR than 75 tablets of verapamil 80 mg IR. Why? Because, all other factors being equal, more of the former is expected to remain in the lumen at a given time post-overdose. True, there’s not a lot of data to support this claim. But a volunteer study reinforces the concept (as well as point #1 above): SDAC given 4 hours after a sub-toxic dose (2.9 grams) of enteric-coated ASA reduced absorption by 57%.18 Not too shabby, but then again, this was a controlled, artificial situation involving volunteers rather than patients.
- Expectation of slowed gastric emptying: When an overdose involves drugs that slow gastric emptying (for example, opioids and drugs with antimuscarinic properties like diphenhydramine, paroxetine, cyclobenzaprine, and many antipsychotic drugs), SDAC becomes intuitively more appealing. In theory, at least, more drug in the lumen means more drug available for adsorption.
- Alert, cooperative patient: This gets at the safety angle and our overall goal of favourably influencing the patient’s outcome. All other factors being equal, an alert, cooperative patient is going to present a lower risk of charcoal aspiration than a somnolent, combative patient. Moreover, to the extent that somnolence and combativeness reflect the absorption of whatever was ingested, this is a consideration that should diminish, at least somewhat, the enthusiasm to give charcoal.
The bottom line
The decision to give charcoal should begin with the realization that most poisoned patients will do fine without it. The subset of patients most likely to benefit from SDAC is defined by several factors, of which time since ingestion is only one. Consequently, basing the decision on timing alone—or even timing as the primary consideration—risks giving SDAC to patients unlikely to benefit and withholding it from patients who might. Decisions about GI decontamination don’t fit nicely into an algorithm. If, after weighing the factors above, you still aren’t sure whether to give charcoal or not, a toxicologist is only a phone call away.
Charcoal Latte by Nathan Dumlao- 1.Harris PJF, Liu Z, Suenaga K. Imaging the atomic structure of activated carbon. J Phys: Condens Matter. August 2008:362201. doi:10.1088/0953-8984/20/36/362201
- 2.Lalhmunsiama, Lee S, Choi S, Tiwari D. Simultaneous Removal of Hg(II) and Phenol Using Functionalized Activated Carbon Derived from Areca Nut Waste. Metals. July 2017:248. doi:10.3390/met7070248
- 3.Berlinger W, Spector R, Goldberg M, Johnson G, Quee C, Berg M. Enhancement of theophylline clearance by oral activated charcoal. Clin Pharmacol Ther. 1983;33(3):351-354. doi:10.1038/clpt.1983.44
- 4.Juurlink D. Activated charcoal for acute overdose: a reappraisal. Br J Clin Pharmacol. 2016;81(3):482-487. doi:10.1111/bcp.12793
- 5.Isbister G, Kumar V. Indications for single-dose activated charcoal administration in acute overdose. Curr Opin Crit Care. 2011;17(4):351-357. doi:10.1097/MCC.0b013e328348bf59
- 6.Cooper G, Le C, Richardson D, Buckley N. A randomized clinical trial of activated charcoal for the routine management of oral drug overdose. QJM. 2005;98(9):655-660. doi:10.1093/qjmed/hci102
- 7.Eddleston M, Juszczak E, Buckley N, et al. Multiple-dose activated charcoal in acute self-poisoning: a randomised controlled trial. Lancet. 2008;371(9612):579-587. doi:10.1016/S0140-6736(08)60270-6
- 8.Friberg L, Isbister G, Hackett L, Duffull S. The population pharmacokinetics of citalopram after deliberate self-poisoning: a Bayesian approach. J Pharmacokinet Pharmacodyn. 2005;32(3-4):571-605. doi:10.1007/s10928-005-0022-6
- 9.Isbister G, Friberg L, Hackett L, Duffull S. Pharmacokinetics of quetiapine in overdose and the effect of activated charcoal. Clin Pharmacol Ther. 2007;81(6):821-827. doi:10.1038/sj.clpt.6100193
- 10.Cooper J, Duffull S, Saiao A, Isbister G. The pharmacokinetics of sertraline in overdose and the effect of activated charcoal. Br J Clin Pharmacol. 2015;79(2):307-315. doi:10.1111/bcp.12500
- 11.van G, Duffull S, Hackett L, Isbister G. Population pharmacokinetics and pharmacodynamics of escitalopram in overdose and the effect of activated charcoal. Br J Clin Pharmacol. 2012;73(3):402-410. doi:10.1111/j.1365-2125.2011.04091.x
- 12.Kumar V, Oscarsson S, Friberg L, Isbister G, Hackett L, Duffull S. The effect of decontamination procedures on the pharmacokinetics of venlafaxine in overdose. Clin Pharmacol Ther. 2009;86(4):403-410. doi:10.1038/clpt.2009.114
- 13.Duffull S, Isbister G. Predicting the requirement for N-acetylcysteine in paracetamol poisoning from reported dose. Clin Toxicol (Phila). 2013;51(8):772-776. doi:10.3109/15563650.2013.830733
- 14.Friberg L, Isbister G, Duffull S. Pharmacokinetic-pharmacodynamic modelling of QT interval prolongation following citalopram overdoses. Br J Clin Pharmacol. 2006;61(2):177-190. doi:10.1111/j.1365-2125.2005.02546.x
- 15.De W, Snoeckx A, Germonpré P, Jorens P. Rapid-onset adult respiratory distress syndrome after activated charcoal aspiration. A pitch-black tale of a potential to kill. Am J Respir Crit Care Med. 2015;191(3):344-345. doi:10.1164/rccm.201409-1607IM
- 16.Juurlink D, Szalai J, McGuigan M. Discrepant advice from poison centres and their medical directors. Can J Clin Pharmacol. 2002;9(2):101-105. https://www.ncbi.nlm.nih.gov/pubmed/12172589.
- 17.LoVecchio F, Shriki J, Innes K, Bermudez J. The feasibility of administration of activated charcoal with respect to current practice guidelines in emergency department patients. J Med Toxicol. 2007;3(3):100-102. doi:10.1007/bf03160918
- 18.Kirshenbaum L, Mathews S, Sitar D, Tenenbein M. Whole-bowel irrigation versus activated charcoal in sorbitol for the ingestion of modified-release pharmaceuticals. Clin Pharmacol Ther. 1989;46(3):264-271. doi:10.1038/clpt.1989.137
Maarten Van Hemelen says
How do you feel about sdac for sedative overdoses? I have been weary of it since I had a young pt who was intubated to receive it after bzd overdose. She formed a bezoar and aspirated in spite of ett, resulting in several days of mechanical ventilation. I felt that she would have been fine if we had just waited, but maybe there is some nuance to this?
Dave Juurlink says
I’m not enthusiastic about AC for sedative-hypnotic overdoses because most of these do fine with supportive care. It’s hard (although not impossible) to envision a scenario in which AC would be expected to meaningfully influence such a patient’s outcome. I for sure wouldn’t intubate such a patient for the sole purpose of giving AC, and I wouldn’t give it if the patient were already intubated, since in this scenario i) the airway is already protected (in other words, the main toxicity of the OD has already been addressed) and ii) the expected utility of AC is lessened by the fact that a fair bit of absorption must already have occurred.