Variation in HPLC results

[jvtailor]

Hello,
The lens of UV detector of my HPLC get fogged. We are facing some inconsistency in results in active analyte area. May the fogging of lens result in inconsistency of results? We checked the injection accuracy and found within the limits.

[tom jupille]

I'd say it's "possible, but not probable".

First off, what level of inconsistency are you seeing (RSD for, say, 5 replicate injections from the same vial)?

Second, how did you check the autosampler performance? with what sample? at what wavelength?

Third, when you got inconsistent area for your analyte, was this with a pure standard or with a "real" sample? and at what wavelength? and was this wavelength on an absorbance maximum for your compound or on a slope?

[jvtailor]

I'd say it's "possible, but not probable".

First off, what level of inconsistency are you seeing (RSD for, say, 5 replicate injections from the same vial)?

Second, how did you check the autosampler performance? with what sample? at what wavelength?

Third, when you got inconsistent area for your analyte, was this with a pure standard or with a "real" sample? and at what wavelength? and was this wavelength on an absorbance maximum for your compound or on a slope?

I'd say it's "possible, but not probable".

First off, what level of inconsistency are you seeing (RSD for, say, 5 replicate injections from the same vial)?

Second, how did you check the autosampler performance? with what sample? at what wavelength?

Third, when you got inconsistent area for your analyte, was this with a pure standard or with a "real" sample? and at what wavelength? and was this wavelength on an absorbance maximum for your compound or on a slope?

I'd say it's "possible, but not probable".

First off, what level of inconsistency are you seeing (RSD for, say, 5 replicate injections from the same vial)?

Second, how did you check the autosampler performance? with what sample? at what wavelength?

Third, when you got inconsistent area for your analyte, was this with a pure standard or with a "real" sample? and at what wavelength? and was this wavelength on an absorbance maximum for your compound or on a slope?

Hello Tom,
Thanks for your quick response.

Inconsistency was observed in area of different samples of same lot. But re-injection of same solution on second day obtained consistent.

We checked injection accuracy by means of water by weight difference after injecting. The wavelength of measurement is 278 nm. % RSD of 5 replicate injection of standard is less than 2%.

The variation observed only in samples which seems sample solution may be faulty but re-injection obtained very good consistent results of sample?

Then on first day what was the cause of inconsistent area of sample. We investigated but could not concluded. Please have your thoughts on this case. Is it fogging effect?

[DR]

(in my opinion)... An issue with the autosampler is a more likely culprit than fogged windows. Perhaps an air bubble worked its way out of the injector over the course of the initial injections, rendering the sampler more suitable for the reinjections thereof on the next day. This assumes that the unanswered question (about whether the wavelength used corresponds to a local absorbance maximum for the analyte) would not be best answered by saying that teh absorbance for teh analyte is very different 2nm in either direction from the wavelength used.

Also - gravimetric determination of injector precision and/or accuracy is, at best, tricky business. It does not work at all for many types of injectors and does not work well for injectors amenable to the technique unless you have adequate balance sensitivity and range.

[tom jupille]

DR said it better than I would have.

[unmgvar]

i ask myself.
why would the detector be consistent for the same vial, and give a different results to the next vial?
if i understand well, a vial gives stable results. so the precision between the samples is the problem
if the problem is between injections of same sample but from different vial, i would ask:
how do you prepare the samples out of the same batch?
is there something in the product itself that can be inconsistent?

[jvtailor]

i ask myself.
why would the detector be consistent for the same vial, and give a different results to the next vial?
if i understand well, a vial gives stable results. so the precision between the samples is the problem
if the problem is between injections of same sample but from different vial, i would ask:
how do you prepare the samples out of the same batch?
is there something in the product itself that can be inconsistent?

Hello,
Thanks to attend the issue. Product is consistent. It is very much sure because the content of analyte found to be consistent in its assay. If product was inconsistent then re injection of same sample solution never obtained consistent next day!

Inconsistency obtained in samples analyzed for dissolution testing which are collected on specified time interval and injected in HPLC.

[jvtailor]

(in my opinion)... An issue with the autosampler is a more likely culprit than fogged windows. Perhaps an air bubble worked its way out of the injector over the course of the initial injections, rendering the sampler more suitable for the reinjections thereof on the next day. This assumes that the unanswered question (about whether the wavelength used corresponds to a local absorbance maximum for the analyte) would not be best answered by saying that teh absorbance for teh analyte is very different 2nm in either direction from the wavelength used.

Also - gravimetric determination of injector precision and/or accuracy is, at best, tricky business. It does not work at all for many types of injectors and does not work well for injectors amenable to the technique unless you have adequate balance sensitivity and range.

Thanks DR,
So as you say there may be air bubble entrapment in injector that may be generated uncertainly. On this account right now we are flushing the system with priming the lines and injector purging followed by needle priming and washing. Or we may have to replace the injector seal.
Is there any alternative or any preventing actions to be suggested to arrest such issue?

Hello,
Let me have your opinion for corrective and preventive action.

[jvtailor]

DR said it better than I would have.

Thanks Tom. we will be in touch.
My ID: jaynish.med@sahmed.com

[jvtailor]

i ask myself.
why would the detector be consistent for the same vial, and give a different results to the next vial?
if i understand well, a vial gives stable results. so the precision between the samples is the problem
if the problem is between injections of same sample but from different vial, i would ask:
how do you prepare the samples out of the same batch?
is there something in the product itself that can be inconsistent?

Hello,
Thanks to attend the issue. Product is consistent. It is very much sure because the content of analyte found to be consistent in its assay. If product was inconsistent then re injection of same sample solution never obtained consistent next day!

Inconsistency obtained in samples analyzed for dissolution testing which are collected on specified time interval and injected in HPLC.

Hello,
Please have your thoughts in this case.

[Klaus I.]

Hello,
The lens of UV detector of my HPLC get fogged. We are facing some inconsistency in results in active analyte area. May the fogging of lens result in inconsistency of results? We checked the injection accuracy and found within the limits.

Servus jvtailor,
Problems with UV-detection are based on bad lamps in most of the cases. Your have replaced the UV-Lamp and also the VIS-Lamp with new ones?

[unmgvar]

jvtailor
you say the test is dissolution. a profile, and you get a difference between the different samples of the product and at the different intervals between the 6-8 glasses.
and from the same vial results are consistent even after 24 hours
the assay and UOC is perfect, all is great all is accurate.
but the dissolution is something else
well maybe:
it is in the formulation maybe, maybe the type of dissolution buffer you are using,
how you do the tablets, the technology, is it paddles, is it baskets what are the time points of sample collection? is it a slow release?
what model of instrument? maybe vibrations differences between your baths?
how do you collect? manually? automation? do you filter? and with what? only when you collect or also prior to HPLC vials? do you make sure the depth of the cannula is the same in all glasses?
what is your injection volume for the HPLC?
and I can go on and on of all the things that could go wrong in a dissolution test with HPLC

[jvtailor]

jvtailor
you say the test is dissolution. a profile, and you get a difference between the different samples of the product and at the different intervals between the 6-8 glasses.
and from the same vial results are consistent even after 24 hours
the assay and UOC is perfect, all is great all is accurate.
but the dissolution is something else
well maybe:
it is in the formulation maybe, maybe the type of dissolution buffer you are using,
how you do the tablets, the technology, is it paddles, is it baskets what are the time points of sample collection? is it a slow release?
what model of instrument? maybe vibrations differences between your baths?
how do you collect? manually? automation? do you filter? and with what? only when you collect or also prior to HPLC vials? do you make sure the depth of the cannula is the same in all glasses?
what is your injection volume for the HPLC?
and I can go on and on of all the things that could go wrong in a dissolution test with HPLC

Hello, Thanks for comments.
Our formulation is controlled drug release. We keep our samples in orbital shaker incubator (OSI) for dissolution/release testing. There is no vibration during process no ultrasonic bath is used. Samples are collected manually to fill HPLC after keeping out from OSI. Sample solution is not filtered as solution remains clear and injected directly. Injection volume is 20µl. Inconsistency obtained only in first time interval but samples of other time intervals obtained consistent. The vials of first time interval which were stored obtained consistent on second day on reinjection.

[HPLCaddict]

Ah, OK. First time I seem to understand your problem. Doesn't sound like a problem of the HPLC itself to me, though.
Question:
- What level of inconsistency do you see at the first time point (%RSD, better yet, what are the absolute values). How much % dissolution at this stage?
- What do these values look like upon reinjection on the next day?
- You're re-injecting from exactly the same vials, right? No new sample pretreatment or something else?
- Are you sure that you're samples are absolutely clear? There might be small particles which are hard to see. When standing until the next day, these might get completely into solution altering the values...

[unmgvar]

why are you using this technique for the dissolution test and not using a standard dissolution tester apparatus?
is the product you have a powder form, a tablet, a capsule, knowing this could help.
it looks like you put the sample in a flask or something equivalent, fill with medium and start shaking?
do you do a prep stage, in which you make sure that the flasks that will be used are pre-heated in the incubator with the medium? what temp?
what is the medium?
do you degas the medium?
what is the speed of the orbital?
have you checked if the results are in correlation with sample position in the instrument?
how much time elapses until you do the first sample collection?
in any case it is something to do with that dissolution procedure that is inconsistent for the first time stage.
it is not an HPLC issue, it never was