Talk:Median lethal dose

From Wikipedia, the free encyclopedia

Contents

[edit] POV?

This article seems to be slanted towards the animal rights movement, e.g. putting "controversy" at the top, and the only links pointing to "Animal Testing" and "Vivisection" (which in its own right is a fairly pejorative term). LD50 is a value arrived at through animal testing, true, but if this article is to be considered the norm then all drug pages should have links to "the horrors of vivisection"...62.56.126.163 20:48, 15 October 2006 (UTC)

Yeah, the animal rights section should probably go after the basic concepts, and I hate 'controversy' sections - there's almost always a better title. Hopefully it's a little better now, though it could still do with external links on other angles of this topic. --Calair 16:53, 16 October 2006 (UTC)

[edit] Should LD50's of substances be given in Wikipedia?

There is at least one drug (fentanyl) for which the wikipedia entry has LD50 info. Agreed, recreational fentanyl users are usually hardcore types (I was one, in the early 1980s!). But there are so many OTC medications or easy-to-get-a-prescription-for medications for which the LD50 is much lower than what laypersons think, otherwise there would be lots more successful suicide attempts. Which don't occur because most people with suicidal feelings think that "if you can kill yorself relatively painlessly with something, it must be hard to get".)

Putting LD50 information on Wikipedia pages would be good, as wikipedia is an encyclopedia, although i think it would be necessary to add a small warning somewhere on page where the LD50 is included that they should not solely depend on this information and especially not if they intend to use it as a 'rubric' for consumption. --Neur0X .talk 00:46, 15 July 2006 (UTC)
I think listing LD50 data for substances would be appropriate, but only if we're careful about verifiability and qualification. LD50 values always depend on several qualifiers - mode of administration, subject species, and (very often) an extrapolation from animal results to humans, for instance. It often has political implications, and even without those, it's human nature to pick the most most impressive (i.e. most lethal) possible interpretation of the data. IMHO, LD50 data should only be presented if it states the relevant measurement circumstances and is directly supported by an appropriate source. --Calair 07:00, 17 July 2006 (UTC)

I have seen that almost all blokes on the usenet [incl rec.drugs.hard] are responsible enuff not to give out this info. Ditto for the web. I hope that wikipedians also will refrain from providing LD info for substances not-too-hard-to-get. I dont think that info belongs to a general-purpose encyclopedia (or, indeed, to any easy-to-access source).

23:53, 4 Jan 2005 (UTC)mp from Calcutta India

looking up LD50's on a substance requires 2 things, knowing what a MSDS is and knowing the chemical name or the chemical composition of a drug. Neither of these are hard to come by. I don't think LD50's should be listed on wikipedia, not because people would use it to commit suicide, but because Wikipedia is not a Chemical info warehouse. There are ALOT of compounds, naming them all I believe are outside the scope of this site, let alone provide detailed information. Maybe wikifoundation could setup a WikiMSDS site or something... (MSDS stands for Material Safety Data Sheet, which usually contains the LD50 dosage, but this is changing as LD50 is being phased out because of animal cruelty) --ORBIT 04:49, 16 Feb 2005 (UTC)

if we have an article on a chemical known for it's toxicity then we should include the LD50 (cyanide arsnic and the like). If not then it probably isn't relvant.Geni 18:17, 25 May 2005 (UTC)

IMO there should be a list of most toxic (eg. botulin) and least toxic materials (carbon, water...) and their LD50 values (if applicable)... As for the dangers of listing them, well, quick Googling can probably find them anyways. - G3, 16:08, 14 February 2006 (UTC)

[edit] Removed text

I removed this passage from the introduction:

Two species are normally required to be involved in the testing. A feeding syringe is forced down the animals’ throats, injecting into the stomach a chemical ingredient of the product, or the product itself, often household cleaners or gardening products like weedkillers. The forced feeding increases in amount until half the animals are dead, after experiencing vomiting and convulsions. Technicians note quantity required to kill 50% of the animals, and then the survivors are killed. No harmful effects are measured, just the lethal dose.

I did so because as it stood it was misleading; its position in the introductory section gives readers to understand that this description applies to all LD50 testing, which it certainly does not. For one thing, the methodology described would only provide a LD50 for ingestion - this might be useful for evaluating toxicity of a consumer product, but not for investigating (e.g.) a snake venom or an injected/inhaled anaesthetic. For another, while many toxins will cause vomiting and convulsions, this isn't universal. And whether harmful effects are measured depends very much on the researchers in question.

This material might be elsewhere in the article, if accompanied by information providing some context on just whose experimental process it's describing, but it doesn't belong in the intro. --Calair 03:14, 18 October 2005 (UTC)


I'm also quite concerned about the POV of the removed statement. I'm not exactly an encyclopaediac source, but it seems a bit fallacious to claim that household cleaners and weedkillers are 'often' used, without some sort of citation. I also doubt the claim presented as fact that they experience vomiting and convulsions, as my gut feeling tells me this cannot be the case all the time. 80.7.199.165 16:16, 10 March 2007 (UTC)

[edit] What when LD-50 information is wrong or misleading...

The LD-50 given for fentanyl is 3.1 mg/kg. The dosage administered by one of those duragesic fentanyl patches, total (since this is what the patch contains) is between 2.5 mg and 10 mg. This is administered over 48-72 hours.

Only people who are opioid tolerant are supposed to get the 75 ug/hour and the 100 ug/hour strengths of the patches.

Now, one might presume that an adult would weigh at least 40 kilograms, so a 10 mg dose would be an eighth of a mg/kg/day. And yet that dosage can still be fatal.

I know of one case where an abuser had to be taken to the hospital because of respratory depression after consuming 10 mg in a single dose. He was normal sized and not completely narcotics naive.

I know the point is not to promote recreational drug use, and I know that there are other issues, but someone trying to size a dose might die in short order at 10th the listed LD-50, and probably would at 1/2 the listed LD-50.

Simicich 00:16, 1 November 2005 (UTC)

This isn't really the place to ask such a question, try the reference desk. But you understanding of LD50 appears flawed. A dose significantly below the LD50 isn't necessarily a safe dose. The LD50 is simply the dose at which 50% of the test population dies. Clearly some will die at a lower dose, they don't all suddenly die when given the LD50. Depending on the substance in question, people's reactions could vary significantly.
Also your calculations are incorrect. If a 40kg individual received 10 mg, that's 1/4 or 0.25mg/kg. If the LD50 is 3mg/kg, this will be about 1/10 of the LD50 and I don't think anyone would suggest this would be a safe dose or a good idea without further information. I definitely don't think anyone in their right mind would suggest 1/2 of the LD50 could be considered a safe dose...
Finally, remember the LD50 is the LD50 in whatever animal was used for testing. Just because the LD50 is 3mg/kg in rats or whatever, doesn't mean it'll be the same in humans. Also, remember it's important to consider the method if administration. If you're talking about the oral LD50, then using this for patches which administer subdermally I assume is a very bad idea. The LD50 is arguably a useful measure to have but they aren't an end all and it's important to understand what they are. In your case, I would suggest the Lowest published lethal dose or LDLo is a more useful figure to consider.
Nil Einne 10:31, 21 December 2006 (UTC)

[edit] Plutonium

Removed this text: Similarly, in the case of certain metalic elements, plutonium in particular, the LD50 is properly expressed as µg/kg, or even as nanograms/kg. About 1 µg of plutonium, if inhaled, is presumed to be probably a lethal dose, in the long term, because it is known that it will promote and almost guarantee lung cancer.

As per discussion on plutonium, there is a lot of dubious information about plutonium toxicity doing the rounds without citing any good sources; AFAICT, a lot of it ultimately seems to have come from Heinlein's fiction story The Long Watch.

This article estimates the cancer risk from inhaling 1 µg of reactor-grade plutonium dust at around 1/200 (for Pu-239, around 1/1300) and cites several other sources that yield similar numbers. Markedly lower numbers need to be supported with citations. --Calair 04:11, 3 November 2005 (UTC)

OK; but I doubt you'd be willing to inhale some plutonium dust. Or would you? Show us all how brave you are.  :)

Depends. How much is 'some'? --Calair 00:55, 4 November 2005 (UTC)

I recall a Dr Cohen (probably Samuel, probably not the same Cohen) who proposed building an extremely radioactive wall, or trench, around Israel. It would kill anyone who went near it, to defend that country against its enemies. I hadn't read too many ideas that were crazier than that one. He was a scientist. I believe he worked at Los Alamos. It made me think of Dr Edward Teller and his insane idea, that the world could survive even a full-scale thermonuclear war; no problem.

I'm inclined to believe that *some* pro-nuclear scientists (like scientists who work for the food, drug or tobacco industry, or really any field where there is money to be made) have agendas that can make their scientific judgement suspect.

This is one of the reasons why we don't take science on trust. Scientists are expected to show how they got their results, so other scientists can examine their reasoning and attempt to reproduce the data that supports it. It doesn't entirely eliminate the problem of vested interest, but it goes a long way to reduce it; this is why controversial claims need citations. --Calair 00:55, 4 November 2005 (UTC)

For example. In the 1950's, one of the Atomic Energy Commission's top scientists began promoting the term "sunshine unit", to describe exposure to nucleotides and radiation. Like it was no more worrisome than getting a suntan. We all know that by the 1960's, fallout had become such a serious problem that it led to the end of above-ground nuclear testing.

However: point taken. I accept your edit. It was really a point about *units*, anyway. I gave plutonium as an example. Thanks for your explanation, but I don't think you need to reach as far as SCIENCE FICTION to express your objection. lol -- Stellar-TO

Come on, that image is horrible.Improfane 01:16, 1 January 2006 (UTC)


[edit] LCt50

It is usually expressed in terms of mg-min/m³ or ppm

I removed 'ppm' because the units don't match - ppm is a concentration, whereas LCt50 is a concentration * time. This may have been a typo for ppm*min, but I don't know for sure. Will also change 'usually' to 'often' since I don't know whether mg-min/m^3 qualifies as 'usually' without this second unit. --Calair 03:14, 1 January 2006 (UTC)

[edit] Database for LD50 values

Is there a general database for all (almost all?) known LD50 values for all tested chemicals? -Samulili 10:26, 16 April 2006 (UTC)

Some MSDSes will contain that information, and I think the Merck Index might also contain LD50/LC50 values for a wide range of chemicals. I don't know of a specific LD50 database, though. --Calair 01:18, 18 April 2006 (UTC)

[edit] Semilethal dose

Removed semilethal dose from the intro, I work with chemicals and haven't heard the term "semilethal dose" being used for LD50. It is also very misleading as it is a fully lethal dose for 50% of the population, while semilethal could be read as sublethal. Dictionary.com for example uses the term "median lethal dose" which is more accurate. Sad mouse 00:24, 20 June 2006 (UTC)

It's not the most common term, but Googling finds quite a few hits that aren't just mirrors of Wikipedia; searching PubMed for 'semilethal dose' found five hits, e.g. "The effect of intravenous endotoxin (Sal. abortus equi) in a semilethal dose in rats was very individual"[1] & "Total number of red blood cells, their acid resistance, deformability and fragmentation were studied in rats after intramuscular injection of a semilethal dose of viper poison" [2].
Searching Pubmed gives 18625 references for LD50, of the 5 references in Pubmed for semilethal dose three (including the one you cite) are non-English papers, and the other two referred to the genetic term semilethal and not to LD50s. Sad mouse 01:09, 20 June 2006 (UTC)
However, note that the genetic usage has a very similar meaning: a mutation that causes at least 50% mortality. --Calair 03:37, 20 June 2006 (UTC)
The genetic usage is not really similar at all. Regardless, the point remains that Pubmed references use "median lethal dose" not "semilethal dose"
"Sufficient to cause 50% mortality" is pretty similar, IMHO. --Calair 05:33, 20 June 2006 (UTC)
Actually semilethal can be used in genetics for a gene which is lethal sometimes, but not always, and does not have to be 50% of the time [3]. Anyway, not a big deal, the current wording is fine, personally I wouldn't even bother to qualify with the "non-English speaking" part. Sad mouse 17:24, 20 June 2006 (UTC)
Furthermore, a google search gives 76,200 results for "median lethal dose", 2,470,000 for LD50 and only 773 for "semilethal dose". Of those 773 hits, most (488) appear to just contain the phrase "LD50 or colloquially semilethal dose", others are actually referring to a sublethal dose (ie, a dose at which none die, such as 164 hits referring to the radiation references). Sad mouse 01:32, 20 June 2006 (UTC)
Can you point me at some of the articles that use it to mean a sublethal dose? I've come across several that use it in connection to radiation, but all that I've checked seem to be using it to mean LD50. For instance, US Patent #5008199 refers first to "X rays at a dose of 630 rad almost corresponding to the LD50" and then to "a semilethal dose of 630 rad". (The same patent also refers to a "semilethal dose" of an injected substance.)
I don't have time to look this up, it is a minor correction that I made. This reference for example uses semilethal where it means sublethal [4]. And again, google has very few references for semilethal and lots of median lethal dose.Sad mouse 04:23, 20 June 2006 (UTC)
I am not contesting the fact that 'median lethal dose' is far more commonly used than 'semilethal', only the contention that 'semilethal' is not used. --Calair 05:33, 20 June 2006 (UTC)
Other examples:
Website on Hiroshima Day: "It is said that 50% of persons who receive a whole body dose of 400 rad, which is known as a semi-lethal dose die, and that those exposed to whole body radiation of 700 rad or more have little chance of escaping death... The location that was exposed to the lethal dose of 700 rad was a point approximately 925 meters away from the hypocentre (in Hiroshima); and in the case of the semilethal dose of 400 rad, ap­proximately 1,025 meters."
'The trigger to cell death determines the efficiency with which dying cells are cleared by neighbours', Cell Death and Differentiation, July 2001: "...we examined engulfment following death induction with DNA damage-dependent and -independent stimuli. Etoposide induces apoptosis as a consequence of DNA damage, whereas the kinase inhibitor staurosporine induces programmed cell death even in anucleate cells. 293 cells exposed to a semilethal dose of etoposide for 48 h exhibited a more than 10-fold increase in phagocytes (cells containing one or more phagosomes) compared to controls..." This one doesn't specifically identify it with LD50, but since it's certainly not talking about a sublethal dose - the whole point is to kill cells - that's presumably what they mean.
I'm certainly not claiming that it is a common usage, but it is out there. It does seem to be the case that it mostly (not always) appears as a result of translation from non-English originals but I don't see that that makes a difference; if it's used in publications intended for an English-speaking audience, then that's a usage that should be acknowledge. --Calair 03:37, 20 June 2006 (UTC)
So we leave in the sentence saying it is another term, and try to clarify, like my edit did, yes?Sad mouse 04:23, 20 June 2006 (UTC)
I agree that 'semilethal' is ambiguous, but then that's why Semilethal dose redirects here, to a page which removes the ambiguity. Given that it is used elsewhere, better that we explain it than not. --Calair 00:36, 20 June 2006 (UTC)
Then maybe the intro could say that it is sometimes referred to as a semilethal dose, but this term is misleading. The reason I clicked on LD50 was because the feature article today inferred the amount of thujone in absinthe was safe because it was below the LD50, whereas that is only evidence that less than half the population would be killed by it. Misleading, I think you'll agree. Sad mouse 01:09, 20 June 2006 (UTC)
No, it didn't. It inferred that it was safe because it was much below the LD50; while that's not a completely cast-iron conclusion, it's far more reasonable than it would be without that 'much'. --Calair 03:37, 20 June 2006 (UTC)
Actually it is a completely usage conclusion, because an LD50 says nothing about the dose at which 10% of the population dies, for example. Anyway, are we happy with how it currently is written? Sad mouse 04:23, 20 June 2006 (UTC)
While knowing LD50 doesn't give us an exact figure for LD10, it gives us a great deal more than nothing - both a hard upper bound (LD10 can't be more than LD50) and a fair bit of probabilistic information beyond that. While variability of effects is different from one substance to another, it's not enough so to stop us from concluding that the LD10 is unlikely to be greater than 0.95x or less than 0.001x the LD50.
I think there are some problems with the discussion there (will tackle that over on absinthe if I get the time) but they look to be more to do with chronic vs. acute exposure issues than with the issue of variability of effect.
I'm quite happy for the article to deprecate 'semilethal dose'; I have some quibbles over the wording (it's not clear to me that 'semilethal' is a particularly misleading term for something that's lethal to half a population) but IMHO, the main thing that needs fixing is that the definition should go first. Will edit accordingly; see what you think. --Calair 05:33, 20 June 2006 (UTC)
  • Are you using median as in statistics? Does it apply to LD50?
  • The PubMed usage might be that they took it from abstracts written by non-native English speakers.
Thus, I would stick with the usage and definition by MeSH while you consult with the staff of the NLM about the correct term. For the sake of full disclosure I'm not a user of LD50 but I have deleted another mistaken definition that equated it to half-life. Jclerman 01:27, 20 June 2006 (UTC)
Yes, as in statistics, which is how LD50 is determined. As above, the term semilethal dose is not used in a single English-language articles in Pubmed to refer to LD50, while >500 articles use "median lethal dose" to refer to LD50. In light of the google search, semilethal dose should probably be taken out completely, it doesn't seem to be used.Sad mouse 01:38, 20 June 2006 (UTC)
Note that PubMed searches only pick up words contained in the abstract; the Cell Death & Differentiation paper I mentioned above, for instance, is indexed in PubMed but doesn't show up on a search for 'semilethal'. That paper is in English, written by five UK-based scientists and published by a UK-based journal, which IMHO is reasonable evidence that 'semilethal' is indeed used, albeit infrequently. --Calair 03:37, 20 June 2006 (UTC)

[edit] Added LD50 template

since it seems rather tedious to write out LD50 in it's html/wikified form, I made a template in which when you wish to say LD50 you simply type {{LD50}}--Neur0X .talk 00:40, 15 July 2006 (UTC)

[edit] ED50

The official standardized method to measure the LD50 is described well under the OECD-methods 401, 402 and 403. The LD50 is a statistical value obtained on biological material. Consequently it has mostly a broad standard deviation.

It might be an idea to mention the ED50 (Effictive dose, 50%) in here as well. This is the amount of a substance that gives the desired therapeutic effect. If these are in same page the LD50/ED50 ratio should also be mentioned, it is a good indicator of how easy it is to take an accidental overdose.

[edit] "LD50"-values in human.

One faces often an confusing (confused) values of LD50 in humans. For instance:
"LD50 of batrachotoxin: estimated at 1 to 2 µg/kg in humans."
Now, it is certainly possible to extrapolate some biochemical/toxicological research values to get an "estimated LD50 in human", but is it of any worth? Moreover, often the "LD50" dose is beeing unintentionally interchanged with the lethal/fatal dose in human (so often exaggerated use of specific expert terms by general public or media, sometimes even inappropriate use of these terms by an expert). That is, f.e., from the information, that a 300 mg dose of potassium cyanide ingested killed an 75 kg individual, is concluded, that the "oral LD50 in human" is 4 mg/kg. I think, that it is not right to mention an LD50 dose in human, unless it is really found out by an LD50-experiment (which is, beside some Nazi crimes during WW2, almost certainly not), because: a) it is almost never plausible enough precise, because not based on realistic, controlled experiments; b) it is of no practical value; even if an LD50 value for a specific chemical/poison was known, it would have no effect on medical and/or forensic aspect of the toxicity of substance/poisoning by substance; in most poisonings, intensive/specific medical care is needed even if LD10, LD1 or LDLo are ingested/resorbed. On the other hand, even a LD99 dose must not be inevitably lethal in an individual. I would suggest the more appropriate and rational use of terms "estimated LD in human" (which could be anything from LD10 to LD99), or "LDLo".--84.163.109.162 12:12, 27 March 2007 (UTC)

I think the answer here is to enforce Wikipedia:Verifiability. If we have a reliable source that lists a human LD50 figure, we can cite it (although it would be good to indicate how that figure is derived - I agree that extrapolated values are very unreliable). If we don't have a citeable source, the claim shouldn't be in the article. --Calair 12:43, 27 March 2007 (UTC)


I agree; however, I still think that the use of "estimated human LD" is more appropriate, reality-close and relevant.--84.163.109.162 12:55, 27 March 2007 (UTC)

[edit] WikiProject class rating

This article was automatically assessed because at least one WikiProject had rated the article as start, and the rating on other projects was brought up to start class. BetacommandBot 16:29, 10 November 2007 (UTC)