The Toxpod

5 in 30 (oral fluid, 2D LC, and automated calibration)

August 14, 2019 Tim Scott & Peter Stockham Season 2 Episode 2
The Toxpod
5 in 30 (oral fluid, 2D LC, and automated calibration)
Show Notes Transcript Chapter Markers

In this episode, we look at 5 recent publications in the field of toxicology.

  1. Oiestad, E. et al. Interpreting oral fluid drug results in prisoners: monitoring current drug intake and detection times for drugs self-administered prior to detention. (2019) Forensic Toxicology. 37 (1):59-74
  2. Erne, R. et al. Hair Analysis: Contamination versus incorporation from the circulatory system - Investigations on single hair samples using time-of-flight secondary ion mass spectrometry and matrix-assisted laser desorption/ionization mass spectrometry. (2019) Analytical Chemistry. 91 (6): 4132-4139
  3. Eckberg, M et al. Separation and identification of isomeric and structurally related synthetic cannabinoids using 2D liquid chromatography and high resolution mass spectrometry. (2019) Journal of Analytical Toxicology. 43 (3): 170-178
  4. Ivanov, I.D. et al. A case of 5F-ADB/FUB-AMB abuse: drug-induced or drug-related death? (2019) Forensic Science International. 297: 372-377
  5. Caspar A.T. et al. Blood plasma level determination using an automated LC-MSn screening system and electronically stored calibrations exemplified for 22 drugs and two active metabolites often requested in emergency toxicology.(2019) Drug Testing and Analysis. 11 (1): 102-111


Contact us at toxpod@tiaft.org

Find out more about TIAFT at www.tiaft.org

The Toxpod is a production of The International Association of Forensic Toxicologists. The opinions expressed by the hosts are their own and do not necessarily reflect the views of TIAFT.


Tim:

Hello and welcome to The Toxpod. I'm Tim Scott.

Peter:

I'm Peter Stockham.

Tim:

And today we're going to do a 5 in 30 episode looking at five recent publications and drawing out some of the interesting aspects of them.

Peter:

Should be good!

Tim:

So our first paper is from Oiestad et al and it's in Forensic Toxicology. The title is Interpreting oral fluid drug results in prisoners: monitoring current drug intake and detection times for drugs self administered prior to detention. So both oral fluid and urine testing are used in all sorts of circumstances. Prison testing obviously, workplace testing, forensic testing, and there's been a lot of studies in recent years about the advantages of one versus the other. Oral fluid obviously is pretty easy to take compared to urine.

Peter:

And you'd expect, oral fluid's often said to be more indicative of recent use than urine perhaps?

Tim:

They do say that although this paper maybe suggests that that might not always be the case. A lot of studies have been done where people are given drugs in a clinical setting and then tested. But in those kinds of studies, you're usually giving people fairly low doses.

Peter:

Yeah, you don't want to kill them.

Tim:

No, that's right. It's hard to get ethical approval to give people the kind of doses that people actually do take in real life. Regular users take high doses of drugs over long periods of time. So studies like this are really important to try and find out what the a detection window and the kinds of concentrations that you might find in oral fluid are in a real life situation where people have been taking drugs.

Peter:

So that also means that there's no indication of the actual amount of drug they took. So that's a, so real life but a little bit uncontrolled.

Tim:

And there's, a lot of the studies that have been done on oral fluid have really focused on just a few key drugs, cannabis, cocaine, some amphetamines. But there's lots and lots of drugs which haven't really had a lot of study so we don't really know what the typical detection windows are and what the typical concentrations are that you might find in oral fluid and then how that compares to something like urine for example. So here they sampled some prisoners on the day of admission into prison and then every day after that for nine days, both oral fluid and urine. So in a way they're assuming that they're not taking any drugs after their admission to prison, although there are a couple of cases which they mentioned here where maybe there was some drug use in prison,

Peter:

Surely not.

Tim:

But they're assuming that they had been regular drug users beforehand and then they stop at that point and so they want to see how long afterwards they can detect the drugs in the oral fluid and the urine.

Peter:

So there's 18 subjects, that's a decent number.

Tim:

Yeah.

Peter:

But of course all a mixture of drugs, you never know who's going to have what.

Tim:

To collect the oral fluid, they used the intercept device, which is a commercially available device. And with these kinds of papers, it's important to look at what the collection device is that they're using, because you can get pretty different concentrations depending on the, the recovery of the particular device that they're using.

Peter:

And some of the devices have buffers, etc, detergents, which can also upset your extraction method. So it's always a compromise between the sort of collection device you've got, the extraction method, how much you collect.

Tim:

So they screen the urine and oral fluid then after collection, using different methods, the urine was screened by a variety of different methods, which I assume is the current way that they do the urine testing in this prison.

Peter:

Right.

Tim:

The oral fluid was screened by a sort of comprehensive LC/MS method and as maybe you'd expect as we've seen from other papers in the past, there is some variation between subjects as to which specimen had the longer detection window. As they see here, some, some drugs like methamphetamine seem to be detected for longer in oral fluid than in urine.

Peter:

So more than nine days. So the study went for nine days and it was still being detected in this particular case in oral fluid for, at the nine day mark, but it wasn't detected in urine. So were the cutoffs the same for both the urine and the oral fluid?

Tim:

No, they're not. So that's another complication with these type of studies is that you, you have different cutoffs for oral fluid and urine.

Peter:

So the oral fluid's been done by LC/MS. Maybe it's more sensitive, maybe they've got more capability to go down further.

Tim:

Yeah. Urine cutoffs are normally set based on adherence to some standard or guideline. But because there's not as much data about concentrations in oral fluid, standards don't always have specific cutoffs for a whole range of drugs. They might have it for some where there's quite a bit of data, but apart from that, people are probably basing their methods off the limits of detection in their method.

Peter:

Yeah. Which is the only way you can do it. So when you're analyzing oral fluids, usually you go into much lower concentrations than you do with urine because often you're using LC/MS and you've got that capability. And although it's been around for 20 years or so, we're still not really sure about how low we need to go for some of these drugs. So we're always looking at lower levels. So you might say this is a, um, a comparison of two different methodologies, not just a comparison of two different body fluids because they're screening both by different techniques.

Tim:

So I guess the, the main practical aim of what they're trying to establish in this study is whether oral fluid is a suitable sample to use in this kind of situation, testing in prisons. And it seems to be, and their conclusion is that it seems to work pretty well. There is a lot more variability in the elimination profile for oral fluid than for urine. For urines, it seems to come down pretty neatly and sort of follow that usual curve profile. But with oral fluids it's a bit more erratic, especially once it gets down to quite low levels around the cutoff. It, it can vary quite a bit.

Peter:

So really we need some more of these sort of studies don't we, to get a better idea of whether it's a viable alternative, but even though they're looking at much lower levels, it does look possible.

Tim:

Yeah. I think we need more of these kinds of studies as well to see what kind of concentrations are appropriate to use for cutoffs and so on. And they've looked at quite a few drugs here, but there are a whole range of more drugs that could be looked at and should be looked at. It's going to take some time to build up the kind of data that we need, that we already have about urine for example, we've got a lot of data about urine levels, but uh, we do already have quite a bit of data about oral fluid levels for some drugs, but we need more and we need more types of drugs and we need more studies like this where it's not a controlled dose that they're being given, but it's based on real life, regular chronic users. So you can see the profiles that happen in real life. All right, so the next paper is from Analytical Chemistry. It's by Robert Erne et al and it's titled Hair analysis: contamination versus incorporation from the circulatory system- Investigations on single hair samples using time-of-flight secondary ion mass spectrometry and Matrix assisted laser desorption ionization mass spectrometry.

Peter:

So I remember seeing a talk at a TIAFT meeting in Argentina a few years ago where Eva Cuypers presented a paper on this and that sort of did send shivers up people's spine I think, who were doing hair analysis because it showed that, proved that there can be migration of material from the outside of a hair to the inside of a hair and to see it demonstrated so graphically as those um I think she was using a similar technique, secondary ion MALDI TOF. Yeah so this is an interesting paper, is it really a response to that do you think?

Tim:

Well, I think it's a response to to that and other similar papers, but yeah, I agree that that paper by Cuypers, I think that image is kind of burned into everybody's brains. That did send shockwaves through the whole hair testing community. In a sense, it wasn't, it wasn't new information. It was something that people had suspected that drugs contaminated externally could migrate into the center of the hair and be bound on quite strongly there. But that was the first time it had really been shown like that.

Peter:

And there's something about seeing a graphical image of it I think.

Tim:

So yeah, this paper I think is in response to that, and maybe in response to our podcast from the last season, Pete, where we were critiquing some of the...

Peter:

Perhaps Tim.

Tim:

No, I think this work was probably done well before that podcast was recorded.

Peter:

We won't claim responsibility for that.

Tim:

But yeah, it has been shown as we said, that drugs don't just stay on the surface of the hair. If they contaminate it externally, they can make their way inside. And so there's an interesting quote in the introduction to this paper. The authors say that interpretation of data from hair analytics seems to be impossible, given all these issues that we've talked about. This is why more knowledge on this subject is urgently needed. So I think the authors here, which are some heavy hitters in terms of hair analysis internationally, they are really trying to get to the bottom of what can we really know from hair analysis. Um, and so in this study they chose one drug. This is kind of a pilot study, I suppose, kind of a proof of concept type study. So they chose one drug Zolpidem, um, which is not one that's usually thought of as a high risk of contamination because it's usually taken in tablets and it's not smoked or anything like that. There's not, it's not powder form and, and it's a common drug of abuse as well.

Peter:

So they doped hair samples with Zolpidem solution...

Tim:

Yeah soaked them in some zolpidem solution.

Peter:

And I'd like to see this done when they cut longitudinally using some equipment along the shaft of the hair.

Tim:

Yeah. That's not just your uh garden scissors they're using to do that.

Peter:

Maybe it's common place in, uh, with the use of microtomes and things like that. But it's nothing that I've been exposed to, but I'd like to see that done in person.

Tim:

Yeah, you need very specialized equipment to be able to do that. And then the, uh, analytical equipment is very specialized as well, using a time of flight mass spectrometer, which is a pretty standard Mass Spectrometer. But the secondary ion part of it is that they're bombarding the surface with a stream of ions and then that's releasing secondary ions from the, the hair in this case, which are then incorporated into the TOF. And so they can measure the masses that way.

Peter:

And so because they can direct and control the beam of ions that are shooting into their hair, they can really accurately map which part of the hair these drug molecules are coming from.

Tim:

Yeah, that's right.

Peter:

So even across a cross section of the hair.

Tim:

So they soaked these hairs in Zolpidem solution and then after that they analyzed them and they did find that there was Zolpidem in the inner compartments of the hair. So obviously by soaking it in a solution, you're getting it penetrating into the inner parts of the hair. And so then they wanted to compare a number of different wash protocols. Cause really the practical side of this is we want to find a wash protocol that is going to remove all of the external contamination but not remove the drug that's got into the hair through use, through the bloodstream.

Peter:

Yeah, the stuff that's on the inside. So it's not from external contamination, it's stuff that's been ingested.

Tim:

Yeah, but really because they found that the zolpidem penetrates to the inner part of the hair when you soak it, the question then becomes, can you tell from that zolpidem which is in the inside of the hair, which has come from external contamination and which hasn't? That's really the question then. And so they did a range of different wash protocols looking at some which are, have been used by other publications or which are commonly used, and then they develop their own in- house one, which is a very extensive wash protocol. Actually, let me, let me find that. Yeah, their in- house wash protocol, normally you might have a couple of different stages of wash, you might have an aqueous, you might have an organic, maybe there's a few different steps. But in their in-house wash protocol, they washed 10 times for two minutes in five mils of methanol with ultrasonication. Then transferred it into a new tube and washed in five mils of methanol for 18 hours. And then after that, rinsed it again three times with methanol. So that's...

Peter:

That's an extraction!

Tim:

Yeah. That's actually more than what some labs do for their extraction.

Peter:

Yeah.

Tim:

So that's an extremely extensive wash. But what they found is that they could, by doing that, they could remove almost all of the Zolpidem that had come from soaking in that zolpidem solution. But then when they tested the hair of chronic zolpidem users, they found that there was some zolpidem that still remained in there, even after that really extensive wash.

Peter:

Oh so that's promising. So even after that extensive wash there's still some endogenously derived zolpdiem present.

Tim:

Yeah. Now what's the reason for that? There must be some kind of strong binding that's happening when it's incorporated through the bloodstream as opposed to when it's incorporated externally.

Peter:

So that was an extreme wash protocol that they used, particularly to demonstrate this, I don't think it's their normal laboratory wash protocol.

Tim:

Well maybe it will become their normal laboratory wash protocol. Who knows? Maybe it'll become part of the guidelines. One interesting thing they did, which I think was very clever to show what they wanted to show was that they soaked this, the hair of these zolpidem users in deuterated Zolpidem and then put it through this same wash process and then analyzed it and they were able to almost completely remove all of the deuterated Zolpidem but there was still zolpidem remaining which was from use, presumably.

Peter:

Yep, that's an excellent way to do it, so you're removing the variable of another sort of hair, different person's hair type.

Tim:

Yep. And obviously by doing that they did reduce the signal quite a bit for the Zolpidem but that's the problem with such a strong decontamination procedure you are going to remove some of the drug, maybe quite a lot of the drug that's inside the hair. And here they are just looking at zolpidem and so this whole process seems to work reasonably well for Zolpidem. It may compromise your ability to detect a single dose of Zolpidem if it's just very low concentrations because they are looking at chronic users here. And there is also the question of how does this then apply to other drugs? Do other drugs behave the same way? Maybe they do or maybe they don't.

Peter:

Looking at the figures here though, when you were talking about the non-deuterated zolpidem, it hasn't decreased by that much after all these washes. I mean it's only, it's very hard to tell from the type of plot that you get from these instruments, but you know, it looks like probably a third to a half of as much as it was there originally, so that's...

Tim:

Yeah, that's pretty good.

Peter:

That's very good considering the amount of washing they did.

Tim:

Yeah. You know what I'd really like to see and you'd, you'd need some kind of ethical approval to do this experiment, is to contaminate someone's hair while it's still on their head. So to soak some of their hair in a drug solution while it's still on their head. Then just let them go about their business for a month. You know, they can wash their hair as normal. They can get it wet. They can...

Peter:

Hopefully they don't have to apply for any jobs or anything.

Tim:

That's right. They can do whatever and then test the hair and see whether that has migrated over time into the inner compartments of the hair and bound more strongly. Because obviously in these kinds of situations where they're just doing it all in vitro, in the lab, you're soaking it, but then you're presumably washing it and extracting it reasonably quickly after that and it's not in a live situation. It's not binding on as strongly as the blood derived zolpidem, but would that be the case if it's just sitting on someone's head for a long period of time?

Peter:

Getting washed with the soapy detergent water every day?

Tim:

Yeah. Would it wash in more? Would it bind more strongly? I don't know the answer to that, but that'd be an interesting study to do.

Peter:

I don't think anyone knows the answer to that.

Tim:

No, and obviously as the authors say here, this is just a pilot study. We need more of these kinds of experiments looking at different drugs, but there aren't that many labs which have this specialized equipment to do this kind of work. So really the onus does fall on on the labs which can do it to do it so that we can move hair analysis forward in terms of the interpretation and get some really clear understanding of the kind of decontamination procedures that are necessary.

Peter:

So the next article we're looking at is an article by Eckberg et al and it's entitled Separation and identification of isomeric and structurally related synthetic cannabinoids using 2D chromatography and high resolution mass spec. So 2D chromatography is a relatively new thing in forensic tox. I don't think it's been widely applied. I know it's, 2D chromatography in terms of GC-GC has been applied in petrochemical instances where they've got, separating very complex mixtures of different hydrocarbons or things like that. But there's not much around in terms of forensic tox.

Tim:

So they're applying it here to NPS, trying to separate out various NPS. And this is one, because there's so many new types of NPS and some of them are just structural isomers and some of them are extremely similar. And so you can't, they have the same mass, you can't tell the difference based on that. And even the mass fragments can be the same.

Peter:

So the chromatography part's just trying to separate those things so we can discriminate between two. So the isomers they're looking at here, are looking at JWH-019, where the fluorine is on a different part of the pentyl chain. So obviously very difficult to tell apart using normal LC/MS techniques.

Tim:

Yeah. And there's five different isomers which all have that fluorine just, you know, moving up the carbon chain there. So very difficult to tell them apart.

Peter:

So I think there's mainly two types of 2D chromatography, there's, I think they call it heart cutting, where they have, they might have a very targeted analysis where they, the first column will give you a peak and then they will divert that peak to another column cause that peak might contain a number of different compounds, and that second column will separate them out. Or there's the other version which they call modulated, where they divert it to a switching valve. So periodically it switches on, off, on, off, and diverts periodically the eluant to another column. And at the same time, you can analyze the effluent of the first column using one technique and the effluent of the other column, with another technique. So at the end of it, you end up with the retention time from column one and retention time from column two and you end up with little pretty 2D topographical map.

Tim:

Yeah. Which is what they did here, they had, well, during the method development, they had two UV detectors, diode array detectors, one after the first column, and then one after the second column. But then when they actually had developed the method and they'd got the gradient the way they wanted, they replaced that second detector, the final detector, with a mass spectrometer,

Peter:

You might have the first column as a C18 and the second column, I think they use a biphenyl. They have to be separable on one of those columns, don't they, otherwise, there's really no point doing 2D, is that right?

Tim:

That's right. And there, there might be some compounds which don't separate on either of those columns and they did try a bunch of different columns to see what would give the best separation and that you are trying to use two different types of columns obviously.

Peter:

Yeah, I was just trying to think of an example where you might use it in a common... So for example, THC on many methods comes out very late in the program using reverse phase chromatography. So unfortunately that means there's lots of other rubbish that happens to come out at the same time because you've got all the material that accumulates in your mobile phase, accumulates in the column then it elutes at around about the same time as THC. So that can cause a lot of ion suppression. So you might be looking for methamphet or something in the first part of your chromatogram. And then when THC comes along you switch it to another column which then separates THC away from all that other rubbish that you've got in the last part of the chromatogram. Perhaps that would be an application.

Tim:

And they do say that it's not necessarily better for routine identification of compounds. It's, but it's where you suspect that there might be an interference or there might be an isomer in your normal method, this is a good way to separate that out. And it's not the only way to do this kind of thing. One of the other things that they mention in this paper as well is ion mobility spectrometry, which is, so following your chromatography, then you have uh, before it goes into your final mass spec detector, you have this ion mobility spectrometry where it's based on the movement of the ions across a very small path and so that you can separate them out in time based on their ionic properties.

Peter:

The simplified thing that I've been told is that it's based more, molecules can be separated in terms of their shape rather than their polarity in the gas phase.

Tim:

Oh right, that's interesting.

Peter:

I don't really understand it to be honest. We'll get one and work at how it works.

Tim:

Yeah, well I mean all of these are quite new techniques, very specialized equipment and still really in the infancy of trying to work out how can they best be used? What are the applications that they're useful for? So it's a promising technique but needs quite a bit more development. Is it going to end up being used more useful than other techniques?

Peter:

Is it very specialized as well? It might, I think it's very difficult to set these methods up and get them up and running. It wouldn't be for a general tox lab to be using this sort of technology I wouldn't think.

Tim:

Yeah.

Peter:

But who knows? Who knows where it's going to go. LC/MS was pretty useless when it first started too!

Tim:

Yeah, everything's pretty useless when it first started. Everyone thinks that's, that's completely impractical. We're never gonna be able to do that.

Peter:

Yeah like electric cars and stuff, mobile phones.

Tim:

We're still waiting for the electric cars, aren't we? Where are they?

Peter:

Australia is, I think the rest of the world's got them.

Tim:

Oh, we're a bit behind.

Peter:

So Tim, the next paper you picked out is relating to synthetic cannabinoids and the title of the paper is A case of 5F-ADB and FUB-AMB abuse, synthetic cannabinoids, is it drug-induced or drug related death? And it's by Ivanov et al and it's in forensic science international.

Tim:

Yeah, so this is a really vexing question. How much do synthetic cannabinoids contribute to death? There's been quite a few case reports in recent years, in the last 10 years, about synthetic cannabinoids being found in postmortem cases where people have died. Are people acutely overdosing on synthetic cannabinoids? Is there some kind of cumulative effect from chronic use of synthetic cannabinoids?

Peter:

Maybe it's affecting an existing health condition, just exacerbating it.

Tim:

Or are they just incidental in particular cases and they, they're there but they're not necessarily causing the death.

Peter:

And because there's so many different synthetic cannabinoids it's very difficult to get a decent study on any one type. When I say decent study, I mean a significant number of deaths for any one particular synthetic cannabinoid.

Tim:

There are some publications of fatal overdoses and then others which suggest that maybe some synthetic cannabinoids aren't as toxic as what they first thought. There was a recent paper about Cumyl Pegaclone suggesting maybe it's not as toxic as what's first thought, but we need more data. We don't have enough data.

Peter:

There were still deaths involving Cumyl Pegaclone though?

Tim:

There have been deaths involving Cumyl Pegaclone. We just don't have enough data to, to make firm conclusions about these synthetic cannabinoids. But this is a case where someone did die. They did have synthetic cannabinoids on board and it appears that, certainly it was the conclusion of the authors here, that the synthetic cannabinoids directly contributed to the death.

Peter:

So in this case there was significant organ damage, which they also attributed to synthetic cannabinoid use. So significant damage to the lung. And that combined with, I think they're saying it's a combined case of induced and drug related.

Tim:

Yeah. Prolonged exposure and acute intoxication. It's hard to really say which one's which. Yeah, they found lung edema, brain edema, various other things. But we don't have a lot of data as well about the longterm effects of synthetic cannabinoids. So it's hard to say.

Peter:

So in this case they found 5F-ADB, which is also known as 5F-MDMB PINACA and a very similar compound. So basically two synthetic cannabinoids. One's got a long chain on it with a fluroine at the end, the other one's got a benzene ring with a para substituted fluorine.

Tim:

And they were tipped off here about that these drugs might've been there because they had some of the material that was found because they do note that if they had just used their normal screening techniques they wouldn't have necessarily found these, or maybe they would have in this case because they weren't quite high doses, but if they were just present at the normal kind of doses you might see, they definitely wouldn't have seen them.

Peter:

The concentration of 5F-ADB was 3.7 ng/mL. And what do you think about the quantitative approach that they took?

Tim:

They use standard addition, which I think is a, it's a pretty time consuming approach but it's probably the, probably the best approach you can take in a case where you don't have a validated method

Peter:

It does make sense to do it in that manner. They're using LC/UV to quantify as well. So the concentration, they got for 5F-ADB was quite high relative to the papers they say.

Tim:

Yeah. Well this brings up an interesting issue because for the measurement of, the quantification of these types of compounds, usually they're not in people's standard validated methods and so you might have to do a bit of validation on the fly. Sometimes validation's very limited. They do mention here that they did some validation in terms of LOD's and recoveries, linearity, but not full validation obviously. And so I think it's, it's a little bit risky to rely on compilations that you might see in publications of... A publication might come out about a particular synthetic cannabinoid and it says these are the ranges that have been found in the literature, but I would really advise going back to those original papers and having a look to see what, what kind of validation was done because some of these results are probably more reliable than others to be honest.

Peter:

But you also have to be careful. You don't want to discourage people from publishing stuff, so they might not have the same laboratory capability as another laboratory. They might not be able to get to the same validation standards because of lack of staff or of things like that. But it's still more important I think to publish the results that they've actually detected this substance. And another thing is that I'm sure there's many, many, many more deaths which haven't been reported in the literature. So there might be databases where you can get this information, like coronial information systems, stuff like that. Not all of those deaths will be reported in the literature. So solely relying on publications to work at how many um, FUBINACA deaths there have been is a little bit tenuous. And not only are they not reported a lot of these because there may be a lot of other deaths that aren't even detected because they're so, in this case they only detected it because it was such a large peak in their routine screening, but at normal concentrations they'd never detect it. So there could be other deaths that they've not seen before.

Tim:

Yeah, that's right. And even these chronic effects over time, if people are just using it regularly but not at very high doses, you may not find it. And so it's, the symptoms might be unexplained in another death. Alright, let's move on to our last paper, which is by Caspar et al. It's in Drug Testing and Analysis and the title is Blood plasma level determination using an automated LC/MS screening system and electronically stored calibrations for 22 drugs and two metabolites in emergency toxicology.

Peter:

Yeah, so they've got a large, this is a commercially available screening platform, um LC/MS/MS screening platform. So they screen for like 900 drugs, but this paper is just talking about quantification of 22 drugs using a limited calibration curve.

Tim:

Yeah. Using a stored calibration curve that you can then use, because toxicologists are always looking for ways to make things more efficient.

Peter:

Lazy.

Tim:

Yeah, that's it, we're all lazy. And one of the things that takes time and resources every time you're doing an extraction, a quantification, is you've got to do your calibration curve. You've gotta extract your calibrators. You might have between four and 10 calibrators for any particular method.

Peter:

This particular instance here of emergency toxicology they might have one sample I guess at a time or a very limited number of samples.

Tim:

Yeah, because they're looking for sort of rapid turnaround.

Peter:

So you don't want to do more calibrators than samples surely that would, so let's, let's make it easier.

Tim:

So one common shortcut is often just to do a single point curve, don't do a full calibration. You might validate your method using a full calibration, but then just use a single point from that time on. That's a pretty common way of approaching it.

Peter:

Yep.

Tim:

But they're looking here at the possibility of storing your calibrators, which you use on one day and then keeping them for a period of time after that. I think here they're looking for up to two months and just using that same calibration curve. And so they're trying to test that here and see whether that's actually, whether that works, whether the validation parameters still hold up over that period of time.

Peter:

It's not just the validation parameters, you've got to make sure that your method is very, very rugged so that there's little variability between different people doing the extraction, different days, the materials you use are exactly the same. So you need a very rugged method to do this sort of thing.

Tim:

So here they're using a Toxtyper, which is an ion trap LCMS so they're, they're actually doing LCMS/MS and MS/MS/MS, MS to the third, which you can do with an ion trap. And as you said, they've got a library of almost a thousand compounds so that they had a screening method, which they'd already already developed, but now they want to add some automated quantitation. Pete where do you stand on the quantitation versus quantification debate?

Peter:

I prefer quantification, but I'm an olden day and I'm, I don't know, why don't we make it shorter, what about quantation?

Tim:

I think I prefer quantification as well. Although I do normally say quantitation, I don't have a strong preference, but some people get very upset if you use the wrong...

Peter:

Yeah, I think it's, you see it all the time now on papers. I've got to stop. I've got to get over it.

Tim:

Quantitation, it's become a common, commonplace word.

Peter:

I'm going to start saying aluminum soon too.

Tim:

No, we're holding out on that one. Aluminium all the way. So in terms of the validation over this two month period that they're storing the calibration curve, they did have a few individual failures outside of their acceptance limits, but on the whole it looked pretty good.

Peter:

I do like the report they get out of it. It's nice and concise. All it's got is a chromatogram with just a list of names, wish I could get a report like that.

Tim:

Yeah. Well it's hard to know if, from an efficiency point of view it's obviously best to just get that. You just want the list of names basically, the list of, give me the list of positives and don't tell me anything else. And the concentrations obviously. But in terms of actually reviewing the data, you do want to have a closer look I think. Or you have to make sure that your parameters are set up really well so that you're very confident that you're not going to get false positives, false negatives.

Peter:

They do mention there's a couple of drugs bromazepam and another one where there's a quite a high detection limit and so they're only able to detect toxic concentrations, so...

Tim:

Which suits their purpose I suppose in an emergency department setting where they are really looking for overdoses, that's probably fine. In a different kind of setting that's, that wouldn't be okay. But methods should just be fit for purpose. They don't need to go lower if they don't need to.

Peter:

So they validated this method according to the German speaking guidelines, the GTFCH guidelines, and they've got a specific guideline for clinical chemistry where there's a 24/7 analysis.

Tim:

Emergency toxicology, that kind of thing.

Peter:

Yeah. So they have slightly larger tolerances for the required accuracy and, accuracy, precision. This is, um, the toxicology testing they're doing here will direct the treatment of the patient. This is something that's not often done in a lot of hospitals around the world. So is that more important than say a forensic test where you're trying to convict someone on a, so should it be more stringent in clinical than it should be in forensic?

Tim:

Well, I guess the, sometimes in forensic settings the, the crime is dependent on the concentration of the drug that's there. There might be cutoffs, or whatnot. Whereas in these types of settings, you really just want to know is there a huge amount of this drug present? If there is, we're going to treat it accordingly. I mean in hospitals a lot of times they're just treating them symptomatically anyway, regardless of what the tox results show.

Peter:

But I mean if you got a false negative in this testing. Is that worse than a false negative from forensic testing?

Tim:

Well, who knows Pete? Who knows? That's a philosophical discussion isn't it for another time.

Peter:

But in terms of efficiency, you've got a... the extraction's quite um laborious, but I guess there's only a few samples at a time. I'm imagining that in a hospital situation you wouldn't be analyzing 40 or 50 of these samples at once. You may be just doing an emergency intoxication. So that doesn't really matter that it's a little bit laborious.

Tim:

Yeah, the important, but the important thing is to get this whole thing done quickly. And so from the processing side of it, it's been set up to do it in the quickest possible way. One thing they're doing is for concentrations which are above the curve because you can only go up a certain amount, obviously before the curve starts topping out going quadratic. So they're diluting those samples and they're, and they've tested the dilution integrity by diluting some concentrations above the curve, one in 10.

Peter:

What did they dilute them with?

Tim:

So they're diluting it with blank plasma and they found that this worked pretty well. They did have a couple of anomalous results. I think amitriptyline gave poor accuracy in the dilution, which again is very difficult to explain why that is, because the other drugs all seem to work very well. But when you're talking about such high concentrations, it's probably not that important anyway. They're obviously above the curve. They're very high. It might not matter so much what the exact concentration is.

Peter:

And so this is a fit for purpose method.

Tim:

Yeah. And the authors conclude that it is fit for purpose. It may not be the most rugged method, um, but in this particular context, it does the job that they're trying to do. And, and really that's what you're trying to aim for with any method.

Peter:

And in any case, the fail safe is they're always analyzing a QC with each batch, I imagine. So they'll know if there's a serious issue with the problem, with the method. And it's obviously so robust that they can use it with, without a worry that it's going to change from day to day. So yeah, it's a good way to do it.

Tim:

Yeah. All right. That's it for this 5 in 30 episode.

Peter:

Thanks for listening and tell your friends.

Tim:

And if you want to contact us, you can email us at thetoxpod@sa. g ov.au. See you next time.

Oiestad, E. et al
Erne, R. et al
Eckberg, M et al
Ivanov, I.D. et al
Caspar A.T. et al