Ladi
☆    

Thailand,
2018-07-14 08:09
(2084 d 08:56 ago)

Posting: # 19061
Views: 3,374
 

 Repeat analysis and re-injection of samples [Bioanalytics]

Dear Forum,

Kindly give your suggestions on what steps should be taken when some subject samples in an analytical run have poor chromatograms such as peak split, peak shoulder, etc. The curve and QC all pass acceptance criteria with good chromatograms. Most samples also have good chromatograms. After studying the LC pressure, it was found that samples with poor chromatograms have abnormal LC pressure (swing, fluctuate). Currently, we do not have the strategy written in our SOP to handle this kind of situation. Since we think this is instrument issue, we have decided to reject the run and reanalysis the whole batch (please comment).

Please also give your comment on what should be done (investigation steps, reinjection, reanalysis, etc) if this kind of event happens again.

Thank you,
Ladi


Edit: Category changed; see also this post #1[Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-07-14 22:04
(2083 d 19:01 ago)

@ Ladi
Posting: # 19062
Views: 2,952
 

 Sticking pump valves?

Hi Ladi!

❝ Currently, we do not have the strategy written in our SOP to handle this kind of situation.


IMHO, you can (and should) have an SOP for reanalysis or reinjection (if stability of extracts is validated).

❝ Since we think this is instrument issue, we have decided to reject the run and reanalysis the whole batch (please comment).


A remote diagnosis is difficult but I also think that it is a hardware-problem. IMHO, it will be almost impossible to write an SOP describing all potential steps to solve the problem. Have the system manual handy, if possible get a service/repair manual, make yourself familiar with the system, think about getting training by the vendor. Only as a last resort, call the service technician.

Reanalyse the batch only once you solved the problem – otherwise you are wasting your precious samples. If you are in a hurry and have another system with the same spec’s, run the batch on the other machine.

My guess:
  • A faulty pump check valve could explain increased pressure and shoulders / peak splits because the flow temporarily stops.
  • Since in a batch there are much more unknown samples than CCs and QCs, it is possible that by pure chance all CCs and QCs are OK.
What I would exclude:
  • If you perform gradient elution a blocked solenoid valve before mixing.
  • Blocked tubing between pump and column.
  • Blocked inlet frit on the column.
  • Contaminated column.
All of those will increase the pressure but not come back to normal.

I suggest to run a batch with analyte / IS in mobile phase (like in system suitability; not extracts). If there is a hardware problem I expect that you see “poor chromatography” in roughly the same percent of injections like in the sample batch. If yes, hopefully you don’t have one of these new fancy systems where you cannot do anything on your own.
General rule in troubleshooting: “Divide and conquer”, i.e., change one thing at a time (otherwise you will not discover the true cause).
Remove one pump value, disassemble, and inspect it with a magnifying glass. Deposit on the sapphire seats and/or the ruby ball? If yes, that’s the smoking gun. It seems that deposits occur more often with acetonitrile in the mobile phase than with methanol. Even if you don’t see anything, ultrasonic both (separately!) in a PP vial with a mild detergent. Rinse with deionized water and ultrasonic with methanol, let dry, assemble, and check whether the problem is resolved. If not, repeat with the other valve(s). I know, this is a time-consuming procedure.

BTW, a fantastic resource for any analyst (and nice reading matter while waiting for the result of the batch): John Dolan’s article series LC Troubleshooting in the LCGC magazine.

Good luck!

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

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

Thailand,
2018-07-16 08:14
(2082 d 08:51 ago)

@ Helmut
Posting: # 19063
Views: 2,834
 

 Sticking pump valves?

Dear Helmut,

Thank you for the valuable troubleshooting insight and tactics. We actually called the service technician and he replaced the check valve. But will surely do more before calling technician next time.

❝ IMHO, you can (and should) have an SOP for reanalysis or reinjection (if stability of extracts is validated).


May I ask you and other members about reinjecting or reanalysis of samples when these things happens?
From The AAPS Journal, Vol. 16, No. 6, November 2014 (page 1170-1171) and The AAPS Journal, Vol. 16, No. 5 September 2014 (page 892) which says that partial batch reinjection is possible, providing it has been tested during validation.

From the past bad experience, we have had issues some study samples (sometime 1-2 QC samples in between too) in an analytical run do not have reportable data due to sensitivity/connection loss in the middle of the run or poor chromatograms. However, CC/QC/IS deviation pass-overall acceptance criteria. For example, Time point 1-12 has data/ good chromatograms, Time point 13-15 has no data or poor chromatograms, Time point 16-20 has data/ good chromatograms again. In these kind of situation, can we have procedure in SOP to partial reinject those problematic samples? Please advice what action should be taken?

Option A- partial reinjection
1) Have some steps in fixing instrument and testing it.
2) Re-run the system suitability test.
3) Reinject a passing QC sample against the previous injected Calibration curve.
4) Re-inject those problematic samples.
5) Re-inject more QC that bracket those problematic samples.
6) If overall run do not pass, do full reinjection from the calibration curve (if stability still valid)

Option B- partial reanalysis
1) Accept the run.
2) Have some steps in fixing instrument and testing it.
3) Repeat analysis those problematic samples only.

Option C- Full Reinjection
1) Have some steps in fixing instrument and testing it.
2) Reinject whole run starting from system suitability test and also Calibration curve.

Option D- Full Reanalysis

1) Have some steps in fixing instrument and testing it.
2) Reanlysis all samples since instrument goes bad in the middle of the run is not acceptable.

Any suggestions or comment is appreciated.

Regards,
Ladi


Edit: White papers of the Global Bioanalysis Consortium Harmonization Team linked. [Helmut]
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,638 registered users;
72 visitors (0 registered, 72 guests [including 7 identified bots]).
Forum time: 16:05 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5