What's considered "routine" GC maintence/repair under cGMP

[KM-USA]

This topic asks what's considered "routine" maintence/repair for cGMP-qualified GCs under cGMP, specifically Agilent 6890/7673.

Our SOP states that routine maintenance can be performed, lists examples of changing inlet liner, septa, gold seal as routine. Last week I had to dismantle an FID because a column had broken off in the capillary jet; I ran an Agilent FID cleaning wire through the jet, flushed with methanol and dried, then put the FID back together, installed the column. Was this routine? It was to me, have done similar for over 30 years. But do I have a piece of paper in the file detailing official training on that? No. Would the FID need to be re-qualified by someone with a certificate, or does system suitability demonstrate to FDA that the system is suitable?

I also last week installed a new plunger motor on the 7673 tower, took me all of 3 minutes, and no tools. Was this routine? It was to me, have several other plunger motors over 30 years. But do I have a piece of paper in the file detailing official training on that? No. But the FID thing was way more involved. Is requalification of the autosampler required? The tech engineer said they'd do RSD of retention time and peak area for 5 injections; our system suitability does that anyway each time the unit is used. Or would this need to have that new plunger motor removed, then reinstalled by an engineer with a certificate and then be "requalified" by HIM running those 5 injections, or does system suitability demonstrate to FDA that the system is suitable?




Obviously, what is routine or not is way different for different people. I used to work as an auto mechanic, but I don't have any training documents for that either; all I have is a 1974 tax return which states my occupation as auto mechanic. Yet this very weekend I repaired a slipping air-conditioning clutch on one vehicle and replaced a bad power steering idler pulley on another vehicle. Each took under 5 minutes to diagnose, and another 5 minutes to fix. Have I ever experienced either repair before? No. Could a professional mechanic have done better? No, but opportunity to "tie" me; some would be as logical in troubleshooting such a noise to the power steering pulley bearing (typically tougher than the repair, especially for intermittents), but many would not.

My supervisor doesn't agree that there are many gray areas in cGMP regulations and our SOPs; I do. I understand that an SOP can't detail everything, but many SOPs leave stuff open for interpretation.

I still haven't figured out if solutions detailed to be made "every day" magically expire at midnight like Cinderella, or are good for 24 hours? If the first interpretation, what happens if the autosampler injects that solution at 12:01 am?

Thanks for any input or comments.

[CE Instruments]

Congratulations I think this is well on the way for "grumpy old man status"
As any of us who where there in the 1970s will have found jobs that were once routine are now things that require qualifications Under cGMP you are certainly not covered to carry out the maintenance that you have done this should have been carried out at great cost by a less well "qualified" (read experienced) engineer.
Here in the UK to carry out any profession you require to be qualified, really stifles moving between professions. Fortunately experience does count , otherwise I am not "qualified" to carry out mine
Add an element of danger (domestic electrics and gas) and not only do you have to be qualified but to work on them you must be licensed (at a cost) as the work is regulated. If I change a wall socket, I must pay a licensed electrician to come in and check it.

[tom jupille]

I'll paraphrase the old joke about European drivers' attitudes toward traffic regulations (leaving the nationalities out -- you know who you are!):

A. Everything is permitted unless it's forbidden
B. Everything is forbidden unless it's permitted
C. Everything is forbidden, even if it's permitted
D. Everything is permitted, even if it's forbidden

What you need is a "meta-SOP" which defines your company's approach to SOPs (personally, I lean toward "A").

[KM-USA]

Today, my supervisor said that disassembling the FID and taking out the jet to clean out the broken capillary was allowable under our SOP, but the 3-minute replacement of the plunger motor was not.

He says our GC SOP staes "repair by qualified personnel" that it means someone with a training certification, not that they possess the skills to do the job correctly. So it's like Yogi: everything's opposite to what you think. So I'm qualified technically but not qualified.

Anyway, it seems that the diagnosis of the intermittent plunger action WAS due to the old plunger motor. We'll see on Wednesday if the service guy has to take out that new plunger motor and then re-install it. Or if I'll have to take it out for him. I think the other tower on that same GC also has same intermittent, same symptoms; I've observed that error on other units in the past, why I had installed plunger motors before (and had a spare in-house).

What used to be a big cost-savings and added benefit of having me as an employee - is now being wasted.

Now, I have an in-house validated test method (I did the validation), and I wrote that the split ratio should be about 10:1. I wonder how much split I can run and still be "about 10:1" ???

[Peter Apps]

If SOPs are written in such loose language that people can waste time worrying about what they do and don't mean, than they are poorly written.

Peter

[JGK]

Now, I have an in-house validated test method (I did the validation), and I wrote that the split ratio should be about 10:1. I wonder how much split I can run and still be "about 10:1" ???

Anywhere between 9.5 and 10.4: 1 (with rounding)

[KM-USA]

Now, I have an in-house validated test method (I did the validation), and I wrote that the split ratio should be about 10:1. I wonder how much split I can run and still be "about 10:1" ???

Anywhere between 9.5 and 10.4: 1 (with rounding)

I'm not sure I agree: that would apply if it stated 10:1 split ratio, with no decimals. "about 10:1" would appear to be more wide.

[Peter Apps]

Now, I have an in-house validated test method (I did the validation), and I wrote that the split ratio should be about 10:1. I wonder how much split I can run and still be "about 10:1" ???

Surely if you validated the method you checked how robust it is to changes in split ratio ?, and so you could just as easily state that the split has to be between X:1 and Y:1, which is completely clear to anyone who has to follow the method.

Peter

[8jw8]

My interpretation would be that changing liners, gold seals, clipping the column and cleaning the jet would all be considered routine maintenace as you're just replacing consumable parts (ie..similar to changing changing the column) but something like a repair to hardware where you have to replace the non-consumable part (ie..broken plunger motor, faulty heater and/or sensor) would be non-routine maintenance.

[DR]

Ours says that for "non-routine" replacement of parts, qualification of the component or instrument must be repeated. We have teh option of doing "simple" repairs ourselves.