Laboratory Equipment Qualification and System
Question and Answers
Frequent speaker and chair person at FDA, ISPE, PDA, USP. IVT,
and GAMP conferences and workshops
||Here Ludwig Huber
(right) in the Q&A Discussion Session on Laboratory
Compliance at an ISPE/FDA Beijing University Conference with
Nick Buhay, Acting Director in FDA/CDER's Division of
Manufacturing and Product Quality
For Dr. Huber's connection with the FDA,
Q: What is the difference between Performance Qualification (PQ) and
Performance Verification (PV)?
PQ is the last Q in the 4Q model for equipment qualification (see
figure below). The qualification terminology has first been
described by the pharmaceutical industry validation experts for
manufacturing equipment and has been used in official documents such
EU-GMP Annex 15 and ICH Q7. Later on the 4Q approach has also been
formally introduced to the analytical laboratory through the USP
Chapter <1058>: Analytical Instrument Qualification and Equipment
Qualification is frequently referenced in FDA warinng letters.. PQ means checking that the complete system works as intended
for a specific application, e.g.,
an HPLC system is not only comprising equipment hardware but also columns,
reference standards and the computer hardware and software controlling the system. PQ
should be checked with a standard that is similar to real samples
and the check should be frequently. According to USP <1058> in
HPC quality control analysis PQ can be substituted by system suitability
testing. The meaning of PQ is different from PV.
The 4Q model for equipment qualification
The term Performance Verification (PV) comes more from the ISO
and ASTM world, such as ISO 9000, ISO
17025 and ASTM E2500. The task of PV is in line with OQ from the 4Q
model and basically means that the equipment (hardware) performs as
specified by the user. The similarity is demonstrated in the figure
below that compare PV as defined in ISI/IEC Guide 2 with OQ as
defined in Pharma publications. The bottom line is that both terms
OQ and PV are used interchangeably for the same task..
The best statement I have seen comes from the GAMP® Good Practice
Guide for Laboratory Computerized Systems (Version 2): "Whatever
terminology is used for verification activity (could also state:
qualification activity), the overwriting requirement is that the
regulated organization can demonstrate that the system is compliant
and fit for intended use”
Performance Verification vs. Operational Qualification
Q: It is my understanding that USP General Chapters above <1000>
recommendations and are do not carry the same weight as those below
1000 they are not "official". The seminar on (AIQ) <1058> sounds as
though it is more than just a recommendation. Please comment
Thanks for your comment. You are right, general chapters from
1000 - 1999
are recommendations. as I have been told by the Chairman of the
chapter, they will become mandatory for any mononograph that
qualify instruments before running a sample. In light of this it is
recommendation to use always the same approach for instrument
but no question, the amount of testing and documentation will vary
complexity of the instrument and the criticality of the application,
leaves a lot of room for interpretation.
Also FDA typically asks for equipment or instrument qualification,
inspectors are quite familiar with the USP approach, so this also
sense. If the seminar description did give the impression, that USP
will always be mandatory, then I am sorry for this, nevertheless I
follow the approach, just to stop the endless discussions.
Q: Since firmware is typically part of the equipment and is OQ'd
by the supplier, does the supplier typically OQ the firmware as part
of a maintenance firmware upgrade?
Firmware is qualified as part of hardware qualification. I don't
believe that such qualification is part of a maintenance firmware
upgrade. But the vendor should give you recommendation if
hardware re-qualification is necessary after a firmware update. This
is only necessary the firmware update can have an impact on the
Q: We have a global LIMS system. At a site level, certain
instruments are linked to the LIMS. If we have to move the hosting
of the global LIMS system, what is the validation/qualification
requirements for the LIMS (global and site levels) and the linked
Most important is to test the communication when you move the
hosting of the global LIMS.
Q: How do you handle shut down of equipment for maintenance and
or removal? I worry about chemical and biological contamination of
personnel handling the shut down and removal
This should be covered by cleaning procedures associated with
safety precautions. This is not covered by USP <1058>.
Q: How does the USP defined IOPQ definition relate to the
current industry and FDA view of verification testing (as per
current ISPE initiative)
To be honest I don't think there is a conflict. All documents
recommend to do some kind of testing when equipment is installed
with to make sure that the equipment works in the intended
environment. Process equipment that is used for a For laboratory
equipment that is always used for specific processes, can be be
fully tested with PQ. Analytical instrument most frequently is used
for several applications and analytical methods. This makes it more
practical to verify first that the instrument meets specifications
which is QO. I am note aware that the FDA has a different view. And
ISPE has published a Good Practice Guide for validation of
laboratory systems. The guide is also based on IQ/OQ/PQ
Q: Why do we need to qualify the instrument if we do system
suitability testing anyway
There are a couple of reasons. I would recommend to look at the
presentation from FDA’s Dr. Nasr. You can download it from the reference
website. On slide 21 of that presentation you can find some reasons, for
example, he states that SST may not guarantee error free chromatography and
he also gives a scientific reference for this, for example SST may not be
comprehensive enough to detect critical HPLC system performance problems,
because you just don’t measure all parameters that are critical for
analysis. An other reason is that ensuring independent and objective
qualification of critical system components will aid in the development of
robust and transferable methods.
Q: Is the USP standard available on the Labcompliance website
No, all these chapters are copy-right protected by USP. You can get it
Q: The standard says, users of commercial systems should check the
vendors quality system. Is an ISO 9000 certification enough proof for this?
For equipment hardware yes, for complex software is it not enough. It
depends on complexity and the risk the software or computer system has on
Q: USP standards are for US FDA regulated environments. What about
The 4Q model is also well established and expected to be implemented also
in Europe. There is an Annex 15 to the European GMPs: Validation and
qualification. This document also requires all four phases for equipment
Q: If calibration is expired on a validated system is it considered
invalidated. If so, then what testing needs to be done to bring it back to
the validated state.
- Yes, it is considered to be invalidated
- Go back and repeat the initial calibration
Q It was implied that determination of system suitability and monitoring
of accuracy through a control sample may not be sufficient to fulfill PQ. It
would seem to me that a properly designed SS would show consistency of
operational parameters. For instance in a chromatographic system retention
within a specified window along with resolution and/or efficiency would
demonstrate that flow rate, temperature, detector parameters were all
functioning consistently. Can a well designed SS fulfill PQ requirements? If
not, what more may be needed?
I had hoped that I made this clear that a well designed SST test can be
sufficient for PQ, only the first USP draft said the opposite, but it has
been corrected. It really depends on how you design the SST runs. E.g., if
you need quantitative trace analysis it is not sufficient to check precision
of amounts at 'some' amounts, you either need to make the test at the
relevant amount in the LOQ range or you need to have a additional PQ test to
re-test the baseline noise.
Q: You discussed very much the implementation of the USP validation
approach in a GMP kind of environment, which tends to require much more in
depth validation. Do you have any insights into implementation in a GLP
environment. The GLP guidelines in general tend to be very vague about what
validation entails, and in my experience the level of validation expected
tends to be less.
I fully agree, but it is really nowhere written. This goes back to risk
based compliance. Some companies call this "GLP Light". Example for less
* Less frequent PQ
* Less frequent OQ requalification
* less detailed vendor assessment
But there is no official matrix
Q: Why is the GAMP terminology and way of categorizing instruments not
used? Why is there not a direct link between the GAMP and the UPS
categorization of instruments?
This seminar was limited to USP AIQ. Bringing in a discussion on the GAMP
Lab Systems Validation Guide and its 7 categories would not fit into the
scope and time frame of this seminar. Seminar 141 Validation of Automated
Laboratory Systems is just on the comparison USP vs. GAMP. Check
Q: This question is related to my previous question:
Calibration is expired on a validated system and after some time
there is a new platform (hardware)added to that system, what
document needs to be initiated.
Open a Deviation: Stating calibration is past due, then
calibrate the system
Initiate Change control to capture hardware change or is
there any suggestion to handle this type of situations
Your recommendation 2) is perfect
The risk assessment may or may nor trigger a new calibration.
Q: Two items are often mixed up: OQ and PQ.
Can you line out the main difference between these?
From my point of view the OQ is covered by the PQ if the PQ is successfully
performed, because one cannot bring a PQ to an end if the device would not
Typically OQ verified specifications of the instrument (e.g., gradient
accuracy of a pump). PQ is method specific, e.g./ checks the precision of
your compound, here in a gradient run. If you have a good OQ that verifies
the gradient precision from e.g., 1-99 %, you have good assurance that the
system will pass PQ tests, not only for one application, but for all
gradient analyses. For some instruments such as a balance OQ and PQ tests
are very much the same, so you can combine them.
Q: Regarding the validation of analytical software (e.g., ChemStation) -
is it preferable to purchase the computer with software installed and
validated by manufacturer vs. installing software on existing computer and
validating on site?
This is not a compliance but a business question. Most likely you get
sooner to your analysis results if you purchase preloaded software on
pre-tested hardware and the vendor provides IQ and OQ, but it may be more
Q: You said in a seminar that there will be a special seminar to go
through examples for various qualification deliverables, will that seminar
include all documents as hand-outs and if yes, when will that be?
In the meantime the seminar has been delivered and comes with a full sent
of qualification documents. For details,
click here .
Q: How frequently should we re-qualify analytical equipment
Frequency will depend on manufacturer’s recommendations, required
performance, degree of use, nature of use and instrument history. For a
typical HPLC instrument we recommend once a year.
Q: Is there any difference when we qualify new versus existing systems
Be aware, the question will never be: is this a new or an old system. The
question will be: how did you make sure that the instrument performs as
intended. This means all documentation should be available as foe new
systems. The process is different, for example, you don't formally define
requirement specifications but describe which functions the instrument is
Q: How long should validation documentation be retained
Validation documents should be retained as long as the regulated data or
records that have been generated by the system.
Q GAMP has published a good practices document on “Validation of
laboratory computer systems”. The new USP draft also says something about
software and computer systems. Is there a difference between both documents
and if so which guide should we follow?
It’s interesting that both have been published at around the same time. I
could not find any conflicting statements in both. The USP chapter <1058> is
very generic and includes all types of instruments. The GAMP guide is more
focused on automated systems. I would recommend to look at both and use the
GAMP guide for computer systems. It is just more specific.
Q: Do many companies performance checks with quality control samples
aside from system suitability testing (SST)?
Not in the pharmaceutical QC analysis. The reason is that SST is required
by USP and the FDA wants to see it. So other QC checks that have to be done
is additional work, not really required by USP or FDA. The concept of
quality control checks is a better one and widely used in other industries,
e.g., environmental testing or chemical testing.
Q: What do FDA inspectors expect from daily calibration of
Inspectors very much like to say a daily qualification of analytical
balances with three reference weights. Such test including
documentation takes about 5 minutes and inspectors get a better
For Usersclub Members: Further Questions or Comments?
Any further question? Members of the Labcompliance Usersclub can
ask any further question related to this site. For information and
registration to the usersclub,
Any further question?
(For Usersclub members only)