Bioequivalence and Bioavailability Forum

Dr. Harish L. Rao ☆ India, 2007-09-08 16:16 (6473 d 15:37 ago) (edited on 2007-09-09 12:48) Posting: # 1060 Views: 17,545	Internal Standard Variation [Bioanalytics] Post reply
	Dear Group Members, How much of Internal Standard (ISTD) Peak Area Variation is allowed for bioanalysis during Method Validation & Study Sample Analysis by HPLC and LC-MS/MS? Is there any guideline for this? Thanks & regards, Dr. Harish L. Rao

Helmut
★★★

Vienna, Austria,
2007-09-08 18:43
(6473 d 13:10 ago)

@ Dr. Harish L. Rao
Posting: # 1061
Views: 15,228

Internal Standard Variation

Post reply

Dear Harish!

❝ How much of Internal Standard (ISTD) Variation is allowed for bioanalysis by HPLC and LC-MS/MS?

During validation: anything that leads to a validated method. :-D

Once you have set your specifications of the method, there’s no point looking at ‘Internal Standard Variation’. Plainly speaking: either a particular batch is valid or not.
The more similar an internal standard is in its physicochemical properties to the analyte, the less you should be concerned about varying detector responses.

Example: Sometimes cartridges get blocked in solid phase extraction (SPE), and only a small fraction of the usual volume will be obtained. If you are using a stable isotope IS in MS it simple wouldn’t make any difference. Of course your signal would be much lower, but who cares. On the other hand, if you are using a compound which is only fairly similar to the analyte, following situation may occur:
Although the elution from SPE-cartridges is expected to happen in an on/off manner (Agilent Technologies, formerly Hewlett-Packard Analytics called it ‘Digital Chromatography’ in the last century), in a partly blocked system the situation may be different. In the worst case all of the analyte or IS is still in the cartridge, whereas the respective other compound is already eluted. This is a pretty nasty situation. Therefore:

choose an IS as similar as possible to the analyte.
if blocking occurs, document it in the lab notebook.

But remember, if you get a ‘strange result’ based on the IS response, there's no way to resolve the problem from the chromatogram itself. The only remedy would be to run the entire method with constant volume transfers. In such a case it would be possible to track the error down to its source. But if doing so, one could also omit the IS at all, and go with an external standard only (no guideline forces us to use an IS!). See also this thread.

❝ Is there any guideline for this?

No, this term is not mentioned at all (there’s not a guideline for everything).

There is not difference in validation or routine analysis between HPLC and LC/MS (besides a higher likeliness of matrix effects in LC/MS). Guidelines treat only bioassays differently.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Ohlbe
★★★

France,
2007-09-09 00:32
(6473 d 07:21 ago)

@ Helmut
Posting: # 1062
Views: 14,927

Internal Standard Variation

Post reply

Dear Harish,

I agree with HS, the main question will be: is the peak area ratio affected by the IS variation or not ?

Supposing the ratio is not affected, another issue is the case of samples with a very low concentration, close to the LLOQ. If your signal (both for your analyte and your IS) is only e.g. 1/4 of what it should be, you may end up with an analyte peak that cannot be measured correctly (too small, or signal to noise ratio too high). You would then incorrectly report it as < LLOQ.

You can define criteria for repeat analysis due to IS variation either in a general SOP or in the method SOP itself (based on what you observed during method validation).

Regards
Ohlbe

Helmut
★★★

Vienna, Austria,
2007-09-09 02:37
(6473 d 05:15 ago)

@ Ohlbe
Posting: # 1063
Views: 14,762

Internal Standard Variation

Post reply

Dear Harish & Ohlbe,

now it becomes interesting!

100% agree with Ohlbe about low responses, but first we have to sort out what exactly we are meaning by ‘IS variation’.

Maybe I’m wrong, but I got the impression from Harish’s post that he was talking about varying responses of single measurements within an analytical batch.

❝ I agree with HS, the main question will be: is the peak area ratio affected by the IS variation or not ?

❝

❝ Supposing the ratio is not affected,...

Ohlbe, I’m not sure whether you are talking about duplicates?
If measuring singlets, we don’t know whether or not the ratio is affected; we only may notice an ‘unusual’ IS response.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Charl
●

2007-09-09 14:25
(6472 d 17:27 ago)

@ Helmut
Posting: # 1064
Views: 14,705

Internal Standard Variation

Post reply

Dear Harish, Ohlbe, & HS..

We should all go back to the concept of WHY do we need the IS.....!

as far as I know it is used to minimize the calculation errors raised by doing the Absolute peak (height or Area)...!!!

so... variability of IS should be at the minimum, it should be studied throughly while developening your method.
(e.g... as Ohlbe mentioned about LLOQ, the case will be due to only the area of the analyte (drug) that would vary to the 1/4, so having a minimal IS variability in this case, we will know that some thing went wrong).

from my personal experience.... :-P

I always think fresh, from scratch, for studying the variability of IS while developing, thus for each HPLC detector there is different approach, e.g, the electrochemical detector would be more sensitive than the UV, the LC MS/MS would take a different approach taking into account the ion enhancing and suppresing, matrix effect, etc.....

So finally I recommend that the IS variability is crucial, so we should do the simplest procedure of extraction, cause there were the most error would occur without noticing

regards....
Charl

Jaime_R
★★

Barcelona,
2007-09-09 15:13
(6472 d 16:40 ago)

@ Charl
Posting: # 1065
Views: 14,788

Internal Standard Variation

Post reply

Dear Charl,

only responsing to part of you post.

❝ We should all go back to the concept of WHY do we need the IS.....!

❝

❝ as far as I know it is used to minimize the calculation errors raised by doing the Absolute peak (height or Area)...!!!

As far a I know - and stated by HS - the IS protects against variability in volume transfer. It has nothing to do with detector response.
Just to give an example (liquid/liquid extraction):
(1) External standard (ES) method
Exactly 500µl plasma + exactly 500µl organic → vortex → centrifuge → exactly 400µl organic to another vial → evaporate → dissolve in exactly 200µl mobile phase → autosampler: calculated amount depends on error of all volumes.
(2) ES with problems
e.g., a gel was formed during mixing, only 150µl organic can be transfered: the volume has to be measured - which is not that easy - and must be incorporated in calculation: calculated amount depends not only on the variability of volumes as in (1), but heavily on measurement error of the volume of organic.
(3) Internal standard (IS) method
As (1), but both transfers of organic don't have to be an exact volume, since the ratio of analyte/IS is employed in calculation - which remains constant with varying volumes.
(4) IS with problems
As (3), nothing has to be done in terms of calculation, although - as stated by Ohlbe - problems exist at low concentrations (but this is also true for all methods).

The IS method is more convenient, because we don't have to bother about exact volume transfers.
But: there are examples in the literature (please don't ask me where - buried somewhere in my files...), where the IS performed worse.

Routinely we do the following: During validation we perform all steps of an IS method as an ES method; in the validation report we include the entire data set in two 'flavours':
(1) the IS method as standard, and
(2) the ES method as an 'alternate'.
If (2) also fulfills generally accepted requirements, the method is validated for both.
In the analytical protocol we define a primary and an alternate method. If something goes wrong in routine analyses (i.e., a peak in a particular subject's samples interferring with the IS) we simply may switch to the ES method. Of course criteria have to be defined in an SOP or the protocol, and exact volume transfers have to be done.

—
Regards, Jaime

Helmut
★★★

Vienna, Austria,
2007-09-09 17:05
(6472 d 14:48 ago)

@ Jaime_R
Posting: # 1066
Views: 14,955

IS vs ES methods

Post reply

Dear Jaime!

❝ But: there are examples in the literature (please don't ask me where - buried somewhere in my files...), where the IS performed worse.

Agree. To quote Karnes et al. (1991):*

The assumption for use of an internal standard is that partition characteristics of the analyte and the internal standard are very similar. This can be a false assumption, […] the only appropriate uses of nonisotopic analogue internal standards are to serve as qualitative markers, to monitor detector stability, and to correct for errors in dilution and pipetting. […] The results of analysis can […] be adversely affected by the use of an internal standard because of the added variability of the internal standard measurement. This has been experimentally shown through a consideration of the law of propagation of errors. The internal standard technique will not inevitably improve, nor will it always adversly affect, the precision of an analytical method. Internal and external standardization techniques may both be evaluated when this issue is in question.

(my emphases)

See one example from our files (spironolactone and metabolites). IS was epoxyprogesterone; don’t worry about the chromatograms, different wavelengths (240 nm[nb] / [nb]280 nm) were recorded in two channels.

Day-to-day accuracy & precision (3 validation batches)

Spironolactone


conc.      IS method      ES method  

ng/ml   acc.%    CV%    acc.%    CV%

 41.2    99.6   2.77    102.8   5.62

183      99.7   3.35    101.2   1.23

733     100.3   1.12     99.7   1.05

7-alpha-thiomethylspirolactone


conc.      IS method      ES method  

ng/ml   acc.%    CV%    acc.%    CV%

 52.5    98.2   4.25     98.2   2.79

233      99.2   2.36    101.4   2.24

933     100.6   1.24     99.7   1.80

Canrenone


conc.      IS method      ES method  

ng/ml   acc.%    CV%    acc.%    CV%

 45.0   102.8   2.61     98.1   2.06

200     100.5   0.62    101.2   0.65

800     100.4   0.90     99.3   0.21

I don’t think the external standard method performed worse; the sponsor insisted in using the IS method anyhow.
The price we had to pay were runs-times prolonged by 56%… :crying:

Karnes TH, Shiu G, and VP Shah
Validation of Bioanalytical Methods
Pharm Res 8(4), 421–6 (1991)

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Charl
●

2007-09-10 12:10
(6471 d 19:42 ago)

@ Jaime_R
Posting: # 1068
Views: 14,527

Internal Standard Variation

Post reply

Dear Jaime_R

Responding to your post.

❝ As far a I know - and stated by HS - the IS protects against variability in volume transfer. It has nothing to do with detector response.

❝ Just to give an example (liquid/liquid extraction):

Variability does not occur in volume transfer. A good chemist can transfer the needed volume exactly.
It is the method of extraction that can or cannot guarantee to transfer the needed amount each time...!!!!! ;-)

detector response has "SOME THING" (1) External standard (ES) method to variability
1- What if I cant have a Standard similar to the analyte in chemical behavior?
-then I might apply or compromise my detection method for the favor of IS, ES
-I might need a lamda switching. ... variability is there.
2- What if my standard has low detection response?
-then I need to add a huge amount to the sample. ... variability is there.
-the matrix would be affected, and the lower concentrations will be affected in response. ... variability is there.

❝ Exactly 500µl plasma + exactly 500µl organic → vortex → centrifuge → exactly 400µl organic to another vial → evaporate → dissolve in exactly 200µl mobile phase → autosampler: calculated amount depends on error of all volumes.

Where exactly you add the ES...?

❝ (2) ES with problems

❝ e.g., a gel was formed during mixing, only 150ul organic can be transfered: the volume has to be measured - which is not that easy - and must be incorporated in calculation: calculated amount depends not only on the variability of volumes as in (1), but heavily on measurement error of the volume of organic.

If the method where developed well, then this e.g., would be an issue back then, nevertheless errors occur, and when they do, I just need to repeat this sample.

❝ (3) Internal standard (IS) method

❝ As (1), but both transfers of organic don't have to be an exact volume, since the ratio of analyte/IS is employed in calculation - which remains constant with varying volumes.

❝ (4) IS with problems

❝ As (3), nothing has to be done in terms of calculation, although - as stated by Ohlbe - problems exist at low concentrations (but this is also true for all methods).

But the detection response would not be consistent, thus reading the area or height for the lower points would be difficult and inaccurate...!!!

❝ The IS method is more convenient, because we don't have to bother about exact volume transfers.

❝ Routinely we do the following: During validation we perform all steps of an IS method as an ES method; in the validation report we include the entire data set in two 'flavours':

❝ (1) the IS method as standard, and

❝ (2) the ES method as an 'alternate'.

❝ If (2) also fulfills generally accepted requirements, the method is validated for both.

❝ In the analytical protocol we define a primary and an alternate method. If something goes wrong in routine analyses (i.e., a peak in a particular subject's samples interferring with the IS) we simply may switch to the ES method. Of course criteria have to be defined in an SOP or the protocol, and exact volume transfers have to be done.

Well,, do you do full and partial validation to the same method for the same study???
"interfering peak" is there weather you applied ES or IS, how would the ES eliminates that "peak"...!!!!! ;-)

Regards
Charl

Jaime_R
★★

Barcelona,
2007-09-10 20:58
(6471 d 10:54 ago)

@ Charl
Posting: # 1072
Views: 14,786

Internal Standard Variation

Post reply

Dear Charl!

❝ Variability does not occur in volume transfer. A good chemist can transfer the needed volume exactly.

If your chemist is a human being, he/she will not be protected against errors, being a good one or not. If there's a small air bubble unoticed in the pipet it will not effect the result after addition of the IS, whereas in an ES-method it may lead to errors in all consequent steps.

❝ It is the method of extraction that can or cannot guarantee to transfer the needed amount each time...!!!!! ;-)

Yes, this favors SPE over liquid-liquid-extraction.

❝ detector response has "SOME THING"

❝ (1) External standard (ES) method to variability

❝ 1- What if I cant have a Standard similar to the analyte in chemical behavior?

❝ -then I might apply or compromise my detection method for the favor of IS, ES

❝ -I might need a lamda switching. ... variability is there.

❝ 2- What if my standard has low detection response?

❝ -then I need to add a huge amount to the sample. ... variability is there.

❝ -the matrix would be affected, and the lower concentrations will be affected in response. ... variability is there.

Sure, but what are you trying to say? :-D

❝ ❝ Exactly 500ul plasma + exactly 500ul organic -> vortex -> centrifuge -> exactly 400ul organic to another vial -> evaporate -> dissolve in exactly 200ul mobile phase -> autosampler: calculated amount depends on error of all volumes.

❝

❝ Where exactly you add the ES...?

Me personally? Depends on the method, but there's no difference whether an IS- or ES-method is applied.
For daily fresh prepared calibrators just spiking plasma, or diluting it (mainly for SPE) 1:x with buffered stock solutions to yield the target concentrations.

❝ ❝ (2) ES with problems

❝ ❝ e.g., a gel was formed during mixing, only 150µl organic can be transfered: the volume has to be measured - which is not that easy - and must be incorporated in calculation: calculated amount depends not only on the variability of volumes as in (1), but heavily on measurement❝ error of the volume of organic.

❝

❝ If the method where developed well, then this e.g., would be an issue back then, nevertheless errors occur, and when they do, I just need to repeat this sample.

Very nice. If a gel has formed, the reason probably (likely?) is a plasma constituent (lipid?) in that particular sample. Maybe re-analysis will solve your problem, but I'm having some doubts.
If you don't want to waste your back-ups, why don't you simply correct in numerically?

❝ ❝ As (3), nothing has to be done in terms of calculation, although - as stated by Ohlbe - problems exist at low concentrations (but this is also true for all methods).

❝

❝ But the detection response would not be consistent, thus reading the area or height for the lower points would be difficult and inaccurate...!!!

Yes, but so what?

❝ ❝ Routinely we do the following: During validation we perform all steps of an IS method as an ES method; in the validation report we include the entire data set in two 'flavours':

❝ ❝ (1) the IS method as standard, and

❝ ❝ (2) the ES method as an 'alternate'.

❝ ❝ If (2) also fulfills generally accepted requirements, the method is validated for both.

❝ ❝ In the analytical protocol we define a primary and an alternate method. If something goes wrong in routine analyses (i.e., a peak in a particular subject's samples interferring with the IS) we simply may switch to the ES method. Of course criteria have to be defined in an SOP or the protocol, and exact volume transfers have to be done.

❝ Well,, do you do full and partial validation to the same method for the same study???

OK, this is definitely a misunderstanding (you used 3 question marks)...
We perform the full validation of all IS-methods as if they were ES-methods (i.e. taking care about exact volume transfers). By this we kill two birds with one stone:

calculating calibration / QCs, whatsover by ratios analyte/IS: the method is validated for IS use
calculating calibration /QCs etc by peak areas / heights (ignoring the IS): the method is validated for ES use

Nothing has to be done in terms of additional laboratory work; it's just a little bit sitting by the desk working on your PC.

❝ "interfering peak" is there weather you applied ES or IS, how would the ES eliminates that "peak"...!!!!! ;-)

Don't quote me wrong, I said:

❝ [...] (i.e., a peak in a particular subject's samples interferring with the IS)

I meant (and talked about) a peak interferring with the IS, not the analyte. If you are observing an interference at the retention time of the IS and your method is also validated as an ES-method and you have stated this in the protocol (better than in an SOP, because this is more transparent to inspectors) you may simply run this batch based on the ES-method. By this it's not even a protocol deviation; just report it.

If you are observing an interference with the analyte you will have to modify the method and do a partly re-validation (linearity, etc).

P.S. Please keep your quotes short (I'm not sure whether I answered your points correctly - so I kept them mainly in my post); single exclamation and question marks are suitable to get the emphases. ;-)

—
Regards, Jaime

Ohlbe
★★★

France,
2007-09-10 01:20
(6472 d 06:33 ago)

@ Helmut
Posting: # 1067
Views: 14,656

Internal Standard Variation

Post reply

Dear all,

❝ Maybe I’m wrong, but I got the impression from Harish’s post that you are talking about varying responses of single measurements within an analytical batch.

❝ Ohlbe, I’m not sure whether you are talking about duplicates?

I was talking about single measurements, where you sometimes have an IS response which is very different from what you have for other samples (for instance half the "usual" response).

❝ If measuring singlets, we don't know whether or not the ratio is affected; we only may notice an 'unusual' IS response.

True, and that's really a complex issue.

Actually I've come across different situations, which may have different causes.

One situation is when all samples from one or several given subject(s) have an IS response which is significantly lower, or higher, than the response of the calibration and QC samples. I've seen runs where subject samples all had an IS response 20 to 30 % lower than cal and QCs. There you may suspect matrix effect (this was LC/MS/MS) and investigate whether it affects your analyte and your IS in a similar way, and thus whether the ratio is modified or not.

Another situation is isolated IS "abnormal" response. HS, you have given a number of reasons which may result in a decrease in IS and may, or may not, affect the ratio. This makes it important to note any abnormal event during sample processing (such as a blocked SPE column). The only way to check whether the ratio was affected or not is to repeat the assay for that sample... The question is: how much IS variation can we accept and what would be the limit to trigger a reassay ? As already discussed, there are different ways to deal with this, for instance:

define criteria in a general SOP (for instance, reassay the sample if the IS peak area differs by more than (50 % ?) from (the mean IS peak area in your calibration samples ? in your calibration and QCs ? In the run ?) (many possibilities !)
define criteria in your method SOP, based on the variations during method validation, and the impact they had - or not- on the accuracy of the affected samples.

Whatever you do, should be done consistently: I've seen runs where samples were reassayed for "IS variation", but in which samples with a higher variation were not repeated. Or failing QCs not reported due to "IS variation", while subject samples which had the same variation (or worse) were accepted and reported...

Regards
Ohlbe

Helmut
★★★

Vienna, Austria,
2007-09-10 22:33
(6471 d 09:20 ago)

@ Ohlbe
Posting: # 1073
Views: 14,744

Internal Standard Variation

Post reply

Dear Ohlbe!

❝ ❝ Ohlbe, I’m not sure whether you are talking about duplicates?

❝ I was talking about single measurements, where you sometimes have an IS response which is very different from what you have for other samples (for instance half the "usual" response).

Thanks for the clarification; I wasn't sure.

❝ ❝ If measuring singlets, we don't know whether or not the ratio is affected; we only may notice an 'unusual' IS response.

❝ True, and that's really a complex issue.

❝ Actually I've come across different situations, which may have different causes.

❝ One situation is when all samples from one or several given subject(s) have an IS response which is significantly lower, or higher, than the response of the calibration and QC samples. I've seen runs where subject samples all had an IS response 20 to 30 % lower than cal and QCs. There you may suspect matrix effect (this was LC/MS/MS) and investigate whether it affects your analyte and your IS in a similar way, and thus whether the ratio is modified or not.

This is the ‘classical’ example of a matrix effect in LC/MS. Although LC/MS was just fashionable a decade ago, it became the standard in routine analyses for many compounds. Unfortunatelly many analysts still are not aware of the potential ion suppression/enhancement by co-eluting compounds and follow the ‘precipitate-and-inject’ approach propagated by instrument manufacturers/vendors.
With rare exceptions 'cleaner' extracts are a prerequisite for LC/MS compared to sample pretreatment needed for 'old-fashioned' HLPC/GC (OK, electrochemical detectors in HPLC are also quite unforgiving).
Everybody should remember that 'not seeing' a signal does not mean 'nothing' is there - analysts can physically touch it sticking in the ion trap after 500 biosamples… ;-)

❝ Another situation is isolated IS "abnormal" response. HS, you have given a number of reasons which may result in a decrease in IS and may, or may not, affect the ratio.

Not me, this was another guy. :cool:

❝ This makes it important to note any abnormal event during sample processing (such as a blocked SPE column). The only way to check whether the ratio was affected or not is to repeat the assay for that sample...

Agree.

❝ […] there are different ways to deal with [re-anyalsis], for instance:

❝ - define criteria in a general SOP (for instance, reassay the sample if the IS peak area differs by more than (50 % ?) from (the mean IS peak area in your calibration samples ? in your calibration and QCs ? In the run ?) (many possibilities !)

Oh yes, fully agree! But since the topic is IS variation, I’m having my doubts whether anybody right now looks at the actual peak area of the IS during method development and/or validation. Normally it’s recorded and stored in the depths of the data system and only the ratio analyte/IS is used.
But as Jaime just stated in his post in another context:

❝ Nothing has to be done in terms of additional laboratory work; it's just a little bit sitting by the desk working on your PC.

In every lab there’s freetime; it should be possible to re-evaluate existing methods in that respect and come up at least with some ideas.
Just my 2 cents (only gut feelings, haven't done it myself): 50% seems to be high (30%?); I would go with calibrators only (because QCs come from a frozen stock, and including the unknowns also will be 'masking' a matrix effect). But this is just a quick shot.

❝ Whatever you do, should be done consistently…

You should have formatted it consistently! Your examples are scarry ones; trouble is preassigned.
Being a certified cave diver we like this one (to be followed in case of problems):

Stop! - Think! - Act!

The order has to be kept, and no step has to be omitted. :-D

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

vamshi
●

2008-08-14 14:33
(6132 d 17:19 ago)

@ Dr. Harish L. Rao
Posting: # 2185
Views: 14,758

Internal Standard Variation

Post reply

Hari,

as in my practice for study sample analysis, the Average area of IS of all CC and QC will be calculated in MS EXCEL sheets.

Just calculate the 50% and 150% for the Average area of IS for CC and QC.

If your study sample IS area is falling out of +/- 50% area you need to identify as IS variation.
or near to LOQ samples if IS variation (is there but with in 50% +/-) perform repeat in Duplicate samples and conform.

For validation if 50% at each level QC and 67% for overall QC pass is sufficient.
if Any one or two runs have drastic variation is there and still the QC samples Pass as per the Area Ratio you can consider.

If frequently IS variation occurs in your method, Use the VALCO Valve.
Valco switching programme allows the Analyte and IS portion of chromatographic elution and So Curtain Plate will be neat.

Use Electonic multi pippette for Reagent or IS addition.

But if frequent IS variation where the Run sequence contain the BT stability or FT stability samples or recovery experiments where you compare the IS area with extracted IS area, your %CV will be not satisfactory.

ElMaestro
★★★

Denmark,
2008-08-14 16:09
(6132 d 15:44 ago)

@ vamshi
Posting: # 2187
Views: 14,053

Internal Standard Variation

Post reply

Imagine:
Throughout the validation you see a stable/constant IS response (e.g. by plotting the IS response as function of sample number within each run).
Then, in the bioanalysis production you see 6 runs with similar IS behaviour, but in the last 5 runs we have a steadily decreasing IS-response.

Now what?
The QC's take care of that and disqualifies runs that are nogos ?? Or stop-think-act a little more?? How?

EM.

Helmut ★★★ Vienna, Austria, 2008-08-14 22:28 (6132 d 09:24 ago) @ ElMaestro Posting: # 2189 Views: 14,108	Internal Standard Variation Post reply
	Dear ElMaestro! ❝ Or stop-think-act a little more?? How? `^^^^` Use the magic “Valco switching programme” — Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes

ElMaestro
★★★

Denmark,
2008-08-15 18:51
(6131 d 13:01 ago)

@ Helmut
Posting: # 2192
Views: 14,443

Internal Standard Variation

Post reply

It never hurts to have an SOP. Set limits for IS-variability. In addition, perhaps limits for a non-param. correlation coefficient would be accepable (Spearman's rank or Kendall's tau; "With X samples per run, IS-response as function of sample number should not have an absolute corr. coeff. larger than 0.whatever" or something like that)???

Oder was?

OK: This would work if the IS response follows a unimodal trend. But might not work if the IS-response has one or more extrema.

EM.

Helmut
★★★

Vienna, Austria,
2008-08-15 19:29
(6131 d 12:24 ago)

@ ElMaestro
Posting: # 2193
Views: 14,005

Internal Standard Variation

Post reply

Servus, ElMaestro!

❝ It never hurts to have a SOP.

Yeah!

❝ Oder was?

I don’ know either; nice suggestions of yours (BTW - for our non-German speaking friends: “Oder was?” means “Or what?”).

❝ OK: This would work if the IS response follows a unimodal trend. But might not work if the IS-response has one or more extrema.

Ha! Of course we even can start some kind of control-chart business, but what for? QCs should be spread within the batch; if the batch passes, so what? An IS protects against the need of exact volume control during sample preparation – if half of the volume is spilled in one step, the analyte/IS ratio should stay the same (OK, as Ohlbe already pointed out last year, problems may arise close to the LLOQ).
I must confess, that I don’t get the point of the IS response variation discussion. :ponder:

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Internal Standard Variation [Bioanalytics]

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

IS vs ES methods

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation

Internal Standard Variation