In this video by CNN's Sanjay Gupta entitled "How dangerous is Fentanyl?", he can be heard saying at the end

"[Fentanyl] is so powerful that just a quarter of a milligram can be fatal"

Is that true? What do we know about the lethality of Fentanyl? I couldn't find the LD50 for it in humans. Is there medical speculation not quite so propagandistic?

  • I find this site which says fentanyl is more lethal than many of these other drugs because even an extremely small dose (2 milligrams) can be fatal. – Daniel R Hicks Feb 2 '18 at 0:55
  • 1
    I guess I wouldn't really expect there to be a firmly established LD50 for humans, because how would you determine it? Unlike with lab animals, you can't legally give humans overdoses of various sizes and see how many of them die. You could try to gather data from accidental or self-administered overdoses, but in such cases I would guess that the actual dose is not known very accurately. – Nate Eldredge Feb 2 '18 at 21:58
  • 1
    I think the quote misses the point: the danger comes from how close the effective and lethal doses are, rather than just the lethal dose measured in milligrams. A drug with an effective dose of 10 micrograms and a lethal dose of 100 micrograms is much safer than a drug with an effective dose of 1 milligram and a lethal dose of 2 milligrams. – Paul Johnson Feb 3 '18 at 19:03
  • "Mockumentaties" like this are made for entertainment and profit to the producers, not information. PS :the Dr. prescribed Fentanyl 50 for me when I had cancer and it did not kill me. – blacksmith37 Apr 19 '18 at 17:35

Fentanyl is about as lethal as nicotine, about 1 mg/kg, but we don't see people dropping dead from smoking cigarettes which each contain about 8-20 mg of nicotine, although long term the tar in cigarettes kill around ten times as many people as opioids, in the US.

What makes current opioids so lethal is the dosing methods. Injecting an amount that isn't precisely known or taking pills when it isn't known what is one's tolerance is dangerous. When opium was smoked a person would pass out before overdosing and smoking nicotine makes its dosing safe.

When substances are made illegal those producing the substances try to make it as concentrated as possible so it is easier to hide. In the Wild West there were more opium dens than saloons and opium users were preferred to alcohol drinkers because they were more predictable and less unruly.

This says 2 mg/kg is lethal: http://www.emcdda.europa.eu/publications/drug-profiles/fentanyl

This is the substitution guide for duragesic(fentanyl) patches vs other opioids: https://www.rxlist.com/duragesic-drug.htm#indications_dosage

According to the above guide 8-17 mg of oral hydrocodone(Vicodin) is roughly equivalent to 25 mcg(micrograms)/hr of duragesic. Using an average of 8 and 17 which is 12.5 mg that is 12,500 mcg, and dividing 12,500 mcg by 24 hours gives about 521 mcg/hr compared to 25 mcg/hr for duragesic/fentanyl, or about 21:1.

Interestingly only about 10% of patients prescribed opioids become addicted and most people have been prescribed opioids after dental procedures without becoming addicted. Humans and other animals produce neurotranmitters called endorphins that bind to the same receptors as opioids to block pain. At least in rats the production of endorphins increases during puberty, which makes sense. Adults can withstand much more pain that children. I would bet that those who get addicted don't produce enough endorphins for the number of receptors they make. Most addicts say they take opioids to feel normal. Maybe they are just trying to make up for a lack of endorphin production by their body, plus withdrawal is quite unpleasant.


As an ER physician who routinely uses fentanyl to treat acute pain in patients, I can confirm that 250 mcg (0.25 mg) would be a potentially lethal dose in some circumstances.

When I use fentanyl to treat pain a usual starting dose would be roughly 1 mcg/kg given intravenously. Because fentanyl acts rapidly and wears off quickly, we can give multiple doses over time that may exceed 250 mcg without risk.

If I were to give a patient a bolus (all at once) dose of 250 mcg of fentanyl (about 3 mcg/kg for an 80 kg adult male; much higher for smaller humans), there would be a definite risk that it would cause them to stop breathing. The smaller the person, and the less habituated they are to opioids, the higher the risk. Fentanyl is pretty short-acting, but it doesn't take long for your brain to be permanently damaged if you stop breathing. In fact, the risk that someone will stop breathing with any dose of fentanyl is small (but real), so we watch them closely in the ER.

AND... that's in a controlled environment. Multiply that risk by using a source of fentanyl where the concentration may not be consistent or what you expect (because it was made in a clandestine lab), and subtract the ability to monitor the recipient, and where the others in the room are medically untrained people, and it's easy to imagine that a 250 mcg dose could be fatal.

Ref: many many sources to look up fentanyl dosing. One example: https://reference.medscape.com/drug/sublimaze-fentanyl-343311


A DEA publication states:

Although the total quantity of fentanyls seized may appear small relative to other illicit drugs, fentanyl is more lethal to potential users than other illicit drugs due to its extremely small lethal dose (approximately 2 milligrams).

And I can't find a clear statement of this, but various references suggest that people (especially children) with no prior opioid exposure are extraordinarily sensitive to overdose.

  • 4
    I'm not sure that this source should meet our standard for evidence. It's a publication from a law enforcement agency which gives no indication of having been written by scientists. It doesn't contain any references to the original research literature, so we have no way of finding out how this number was arrived at - what kind of studies were done, what methodology, what statistical analysis, was it peer-reviewed? – Nate Eldredge Feb 2 '18 at 22:01

Per unit fentanyl is far more lethal than other typical opiates

Fentanyl is in the same class of medical compounds as heroin (medically known as diamorphine) and is also used in medially supervised pain relief. Both are opioids and both are dangerous because of the consequences of taking too much. The medical guides (this is from the British National Formulary or BNF) give this general warning about opioids:

Opioids (narcotic analgesics) cause coma, respiratory depression, and pinpoint pupils. For details on the management of poisoning, see Opioids, under Emergency treatment of poisoning and consider the specific antidote, naloxone hydrochloride.

Stoping breathing while in a coma is not usually regarded as an acceptable side effect of a pain-relieving drug so recommended doses are set to minimise that risk. The problem being that the effective window where the patient gets relief from severe pain but doesn't stop breathing is narrow (one reason why using illegal sources where the concentration is uncertain is very risky).

But we can gauge something about the relative "strength" of fentanyl from the medically recommended starting dosages of it and heroin when given for acute pain (these figs are from the BNF):

  • heroin: 5mg every 4hr
  • fentanyl: 12 𝝻g every hour

So the dose rate for fentanyl is about 1% that of heroin by weight (crudely suggesting that fentanyl is at least 100 time more potent than heroin.

The warning sections in formularies spend a lot of time on getting the dose titration correct because of risk of overdose and fentanyl seems to be more problematic than heroin (which is not exactly easy to start with).

So even under medical supervision fentanyl is dangerous because of the risk of overdose. The amount that will actually be fatal, though, depends on the patient and will be higher in a patient used to the drug than in a fresh user. On any measure though, fentanyl is far more potent than heroin which makes it especially dangerous in unsupervised hands or when sourced illegally as the dosage will vary unpredictably and dangerously.

Note. For those who have trouble accessing the BNF link (there are some restrictions on access), you can get the content as an iPhone app and in print. For other ways to access the information see the wikipedia page.

  • 1
    I'm getting a permission error (403) from the link. – Oddthinking Feb 3 '18 at 2:57
  • 1
    @Oddthinking Some online versions are geographically and specialist restricted (didn't think that one was, though). I'll add links to the print and app versions. – matt_black Feb 3 '18 at 12:56

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .