-
N&PD Moderators: Skorpio
How is amphetamine neurotoxic? What does it do to you?
- Thread starter TheSwede
- Start date
ebola?
Bluelight Crew
Hah...I was about to reply on the basis of that being a correct reading. you would need opulent statistical power to tease out an influence with an effect-size that small.
ebola
Seppi
Bluelighter
- Joined
- Feb 22, 2014
- Messages
- 294
To be fair, I think it's kind of odd to write a computed P value as an inequality, even if some software package rounds it to 0 due to machine precision. It should be computed/stated as an equality because it's a statistic (esp. since precision is important when P is very close to the significance level). I can't think of any software I've used that doesn't just use P=0 when P is stupidly small.
Lightning-Nl
Bluelighter
- Joined
- Nov 11, 2012
- Messages
- 1,247
^^ Yup.
Excuse me for not knowing what a p-value is. Although I like to think so sometimes, I don't know everything. Which is unfortunate because I could be an even bigger asshole than I already am if I knew more.
endotropic
Bluelight Crew
- Joined
- Feb 14, 2012
- Messages
- 1,140
That's typically how P values are reported. Basically P is compared against certain predetermined thresholds (.05, .01, .001, etc.), and once P falls below one of those thresholds it's reported that P < threshold. It might not be completely logical, but that's the convention.
ebola?
Bluelight Crew
seppi said:
Reporting as an inequality provides information about the cut-off used for hypothesis testing. I always just reported the statistic, as this provides more information, but for whatever reason, it's not that common to do that.
Swamp Fox said:
This is pretty important for understanding research. The p-value gives the probability that the observed experimental results have occurred due to chance rather than an association between the variables being investigated, given that sampling is perfectly non-biased and that hypotheses are a priori.
ebola
sekio
Bluelight Crew
[borat voice] very nice! I would have figured you'd put that shit into google before calling people out on it.
ebola?
Bluelight Crew
In research practice, situations where the results are 'borderline significant' usually provoke additional data-collection and small adjustments to the experimental procedure, leading to violations of the assumptions that preserve the meaningfulness of what the p-value says about the probability of the experimental results.
ebola
Seppi
Bluelighter
- Joined
- Feb 22, 2014
- Messages
- 294
I wasn't talking about stating the testing rules/cases(i.e. the wiki page inequalities); I meant the computed statistic itself.
It's being written as P<X for very small X due to limits in machine arithmetic (this X has no relation to alpha/significance level - more related to computational/numerical analysis). The computed p statistic is just a single real number.
https://en.m.wikipedia.org/wiki/Numerical_analysisLast edited:
ebola?
Bluelight Crew
This happens, but it is for willful truncation, not to limits of machine arithmetic.
ebola
Seppi
Bluelighter
- Joined
- Feb 22, 2014
- Messages
- 294
What're you referring to when you say "this happens"?
The IEEE 754 standard for machine arithmetic has +0 and -0 defined for this reason (truncation for infinitesimals). P<0.0001 would be P=0 if it were being truncated.
Just to be clear, I used the phrase "limits of machine arithmetic" as a catch-all way of describing the failure of the code file (machine) to produce the correct output - a single value, not an interval (technically, a non-degenerate interval). I'm not ascribing a specific cause (like rounding error or hard-coded truncation) though, since I'd obviously have no clue without looking at data (code+output) to which I have no access.
If that's not the case, then the author doesn't understand P values (this is why I assume it's the script's fault).
__________________________________________________
You two are thinking of the relationship between P and significance level/alpha (the name for those predetermined thresholds). That's not really relevant to this, since no one in their right mind chooses .0001 as a significance level.Last edited:
Seppi
Bluelighter
- Joined
- Feb 22, 2014
- Messages
- 294
thx for your help so there is no supplements that can reduce the neurotoxicity of meth usage ? (average dose, 1 or twice every 3-4 months, no binge or sleep deprivation) ? all the talk about these supplements in the amphetamine neurotixicty/tolerence reduction thread are worthless ?
I wouldn't say worthless so much as misinformed. It's not worthless, because its often beneficial to consume the types of antioxidants that reduce meth neurotoxicity in rats. It's misguided/misinformed because there's no neurotoxicity being prevented when used concurrently with therapeutic amphetamine (as there's no neurotoxicity to begin with).
This unfortunately results in pointless fear-mongering about a non-existent problem with amphetamine when used at therapeutic doses though.
As for methamphetamine, I'm assuming someone else who has read more of this site will have a thread link on hand about supplement research (i.e., I'm too lazy to look for it myself). Off the top of my head from studies with meth on rats, there's: N-acetylcysteine, Vitamins C+E, acetylcarnitine, coenzyme Q10, and melatonin.Last edited:BallsStapledToLeg
Greenlighter
- Joined
- Jan 1, 2014
- Messages
- 41
What are you talking about? Neurotoxic effects have been reported in non-human primates using ADHD level dosing of an Adderall like mixture. That's pretty much the gold standard for neurotoxic studies, until someone sneaks sacrificing human users past an ethics board.
http://www.ncbi.nlm.nih.gov/pubmed/16014752
http://pharmacology.ucsd.edu/graduate/courseinfo/documents/ricaurteadhd.pdf (Free full text)
sekio
Bluelight Crew
Ricaurte's studies have a shadow of gloom upon them ever since he fucked up a MDMA study by giving monkeys meth instead. I would be wary of any single paper of his that is not corroborated from outside sources.I heard that he additionally attached a mask to their face, effectively asphyxiating them.
agreed. i wonder how he manages to still get published in reputable journals.
and he still uses that ridiculous scaling scheme resulting in huge doses for the animals, which he claims are equivalent to normal human doses, even when we have good evidence that the animals reach the same plasma levels with similar (mg/kg body weight) doses and due to their metabolism are more susceptible to damage than humans. just in 2013 there's a new paper on mdma neurotoxicity wherein the lowest dose they gave was nearly 6mg/kg...
anything that has the name "ricaurte" on it is a hot candidate for the trash can (at least if it isn't backed up by the work of other authors).
Seppi
Bluelighter
- Joined
- Feb 22, 2014
- Messages
- 294
Note: I'm not re-citing the papers I've already linked earlier in this thread again in this reply.
I read that paper as well as newer research on them (those noted in http://www.ncbi.nlm.nih.gov/pubmed/23070200) a while ago. Research on nonhuman primates is still animal research / animal models for neurotoxicity. It's not a gold standard for humans at all - it's just preclinical evidence; it's a gold standard for rhesus monkeys though. More seriously, it's not necessary to dissect a brain to see whats going on inside DA neurons. Human meth NT evidence is based upon autopsies of human meth abusers and fmri scans of living individuals - there's a couple good reviews documenting human neurotoxicty in the meth wiki article. This is the best IMO. I mention this because it's entirely possible (in fact, apparently not that hard) to detect this type of neurotoxicity in humans.
If you had read this thread completely, you'd have read that there's clinical evidence of increased DA transporter availability in humans who have used amph at therapeutic doses. I also cited a very new paper indicating that the reverse case (fewer transporters) is observed in heavy abusers (intake range being 10-40 times the max therapeutic dose) of dextroamphetamine. Taken together, that means what happens in humans and rhesus monkeys at therapeutic doses is exactly opposite. However, at much higher doses, there are marginally statistically significant and permanent (could be irreversible) adverse effects on DA neurotransmission via DA transporter loss.
New evidence reviews that are solely on humans w/ acute toxicity of amphetamine don't even mention neurotoxicity... because it's just never observed in humans. Methamphetamine, however, does produce acute neurotoxicity, as I noted before. >.>
Example reviews for acute amphetamine toxicity without neurotoxicity : review 1 + review 2
*The newest amphetamine neurotoxicity review (which is what I used to write/cite that section on wikipedia) indicates that there's more metabolic pathways in rhesus monkeys/rats than there are in humans - one among those has highly neurotoxic metabolites; so, there's is a reasonable explanation for why this difference is observed.
There's more that I've mentioned earlier in this thread but haven't summarized here.
Just FYI, Ricuerte has been doing this study - http://projectreporter.nih.gov/project_info_description.cfm?aid=8429516&icde=0 - and I'm willing to bet he won't find anything in his data based upon similar research attempted by others.Last edited:BallsStapledToLeg
Greenlighter
- Joined
- Jan 1, 2014
- Messages
- 41
Note: I'm not re-citing the papers I've already linked earlier in this thread again in this reply.
I read that paper as well as newer research on them (those noted in http://www.ncbi.nlm.nih.gov/pubmed/23070200) a while ago. Research on nonhuman primates is still animal research / animal models for neurotoxicity. It's not a gold standard for humans at all - it's just preclinical evidence; it's a gold standard for rhesus monkeys though. More seriously, it's not necessary to dissect a brain to see whats going on inside DA neurons. Human meth NT evidence is based upon autopsies of human meth abusers and fmri scans of living individuals - there's a couple good reviews documenting human neurotoxicty in the meth wiki article. This is the best IMO. I mention this because it's entirely possible (in fact, apparently not that hard) to detect this type of neurotoxicity in humans.
If you had read this thread completely, you'd have read that there's clinical evidence of increased DA transporter availability in humans who have used amph at therapeutic doses. I also cited a very new paper indicating that the reverse case (fewer transporters) is observed in heavy abusers (intake range being 10-40 times the max therapeutic dose) of dextroamphetamine. Taken together, that means what happens in humans and rhesus monkeys at therapeutic doses is exactly opposite. However, at much higher doses, there are marginally statistically significant and permanent (could be irreversible) adverse effects on DA neurotransmission via DA transporter loss.
New evidence reviews that are solely on humans w/ acute toxicity of amphetamine don't even mention neurotoxicity... because it's just never observed in humans. Methamphetamine, however, does produce acute neurotoxicity, as I noted before. >.>
Example reviews for acute amphetamine toxicity without neurotoxicity : review 1 + review 2
*The newest amphetamine neurotoxicity review (which is what I used to write/cite that section on wikipedia) indicates that there's more metabolic pathways in rhesus monkeys/rats than there are in humans - one among those has highly neurotoxic metabolites; so, there's is a reasonable explanation for why this difference is observed.
There's more that I've mentioned earlier in this thread but haven't summarized here.
Just FYI, Ricuerte has been doing this study - http://projectreporter.nih.gov/project_info_description.cfm?aid=8429516&icde=0 - and I'm willing to bet he won't find anything in his data based upon similar research attempted by others.
I looked through the reviews you posted and most of their sources, while they do discuss various metabolic pathways amphetamines (mainly MDMA) go through in mice I didn't see anything explicitly saying that humans had fewer metabolic pathways with regards to (meth)amphetamine. Mind posting a source for that? Its something I'd be very interested in.
Thanks
Seppi
Bluelighter
- Joined
- Feb 22, 2014
- Messages
- 294
No, you read it correctly, but you're not seeing through the authors' bullshit. Compare fig. 4. with what the metabolism section says about amphetamine. You'll then wonder why they chose to included alpha-methyldopamine (shown as 3,4-dihydroxyamphetamine) as a metabolite in that graphic. That is one of the highly neurotoxic metabolites in other species.
Human CYP2D6 is responsible for 4-hydroxylations in the human metabolic pathway. This DOES NOT 3-hydroxylate any amph metabolites in humans. Hence, humans DO NOT produce any 3,4- (catechol type) metabolites. Read the papers they cite on hydroxylation. You could also read the refs in the amphetamine pharmacokinetics section and you'll see that neither their references nor my references state that humans produce 3-hydroxy metabolites from amphetamine.
The whole cardiotoxicity section contradicts the massive FDA studies on the lack of cardiac effects in amphetamine at therapeutic doses (see amphetamine - side effects section). The FDA sample was practically a census (literally, massive, and afaik included 100% of the target population for statistical inference in the sample). They used case reports (imo: ROFL...anecdotal evidence) to argue their point that amphetamine is has severe cardiac adverse effects at therapeutic doses.
The sentence below is flat out wrong since FMO3 is just as dominant (conditionally, more dominant) as CYP2D6 metabolism in humans. The review doesn't even mention FMO3 or DBH. But, it mentions GSH conjugates which have never been observed in humans.
Since, in humans, amphetamines are mainly metabolized
by the enzymatic complex CYP450, their effects can be
influenced by PK/PD interactions with other compounds
(pharmaceuticals, drugs of abuse, nutrients, etc.) that share
or affect the same metabolic pathway (Oesterheld et al.
2004).
Toxnet/pubchem (basically, NCBI) indicates w/ multiple monographs on both pages (link to HSDB/Toxnet) that there is a LOT of interspecies variability in metabolism.
I have no clue how this paper passed peer review. I'd have forced them to fix their shit if I reviewed it, which is sad because I'm not even a researcher in their field; yet, I can spot glaring errors or misleading statements on every page with text.
Researchers that use the term "amphetamines" to refer to anything but amphetamine also annoy me since it's an abuse of language.
In any event, the part of the review on indirect neurotoxicity in amph is actually correct - it involves DA autoxidation and ROS production in humans. That's why I cited it. The rest is crap - I personally believe these authors, instead of simply being crappy researchers, just have an agenda to trash these drugs.
I can't imagine why else there isn't a single positive or differentiating factor mentioned for amphetamine in humans from subst. amphetamines in humans OR amphetamine in other animals. These aren't obscure facts; Pubchem or Toxnet/HSDB alone differentiates amphetamine in humans and other animals completely (see animal monologues). They even states there is likely no direct toxicity in humans, differentiating it from other subst. amphs (see human monologues).
***EDIT: FMO3 and CYP2D6 work together in the phenylacetone pathway (deaminating amph); only CYP2D6 is involved in 4-hydroxylations. What I originally wrote about FMO3 being involved in 4-hydroxylations as well was incorrect.***Last edited:
