Contamination of insulin pump cartridges

For the discussion of the sciences. Physics problems, chemistry equations, biology weirdness, it all goes here.

Moderators: gmalivuk, Moderators General, Prelates

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Contamination of insulin pump cartridges

Postby Eebster the Great » Thu May 07, 2015 12:52 pm UTC

So I was filling up a cartridge for an insulin pump just now and the following consideration occurred to me.

Insulin vials (that you get from pharmacies or delivery services) contain 10 mL of dilute insulin solution (in my case, 100 IU of insulin per mL of solution), while the cartridges you fill have a capacity of 3 mL (in reality, many pumps can accommodate slightly more insulin than that). So inevitably, you end up with a vial with a substantial amount of insulin remaining that will still only partially fill your cartridge. You then need to fill the rest from the next vial. Combine this with the fact that you need to push the same volume of air into the vial that you are removing in insulin to equalize pressure, and you get an interesting problem of contamination.

Insulin can be contaminated for several reasons, mostly concerning denaturation due to light and heat and bacterial colonization. When you are filling a single cartridge from multiple vials, which in theory should happen exactly 2/3 of the time (but in practice happens even more often, since filling is imprecise and insulin is very expensive in the U.S.), you will inevitably push some insulin from the older vial into the new one. (Even if you try to only push the air out, some insulin will come with it, and since you absolutely need to remove all bubbles, the amount is often significant.) If the only concern were denaturation, you would only get first-order effects of significance, since serial dilution very rapidly decreases concentration. But since vials last a long time relative to bacterial generation, the problem of colonization is more pernicious. As you continue to cross-contaminate one vial to the next, does the risk of pathogenic colonization continue to increase as your lifetime? If I keep doing this, am I eventually inevitably driving to a bad vial? Worse still, if my insulin formulation contains some antibiotics, will I culture resistant bacteria this way?

To my knowledge, no experiment on this effect has been performed. This is the kind of thing I wish I had done in my middle-school science experiment. But even without actual data, maybe some people here can come up with educated guesses as to the risks here, and insight into whether such experiments should be performed.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Contamination of insulin pump cartridges

Postby Izawwlgood » Thu May 07, 2015 2:08 pm UTC

Can you post a photo of this system? Feasibly there's minimal exchange of contaminated surfaces, and thus the risk is somewhat low. Additionally, I wouldn't be surprised if insulin was shipped with some form of antibiotic or growth retardant. Furthermore, I wager aside from water, there's not a whole lot in this system that bacteria can consume, afaik. It's not like you're injecting a sucrose solution.

Oh, and is this stuff stored cold? For comparison, I routinely work with solutions that are basically salt, small peptides, and some stabilizing sugars, and they can be kept for months.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Fri May 08, 2015 12:37 am UTC

Image
This is an illustration of the setup. The dark plastic piece contains a hollow needle and unscrews after the cartridge is filled.

The insulin formulation used in pumps consists of water, a pH buffer (to a pH of 7.0-7.8), meta-cresol, glycerin, recombinant substituted human insulin (usually insulin aspart or lispro), and zinc chloride. Apparently the m-cresol is the preservative. The concentration of insulin is too low to serve as a food source, but the concentration of glycerin is not. However, the biggest risk comes from chemicals introduced by the needle.

Insulin is supposed to be stored cold after opening, but in practice I rarely refrigerate it for reasons of convenience.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Contamination of insulin pump cartridges

Postby Izawwlgood » Fri May 08, 2015 1:07 am UTC

So after you withdraw the insulin, how do you cap the needle? Do you use a different needle on reloading? Dirtying the needle strikes me as an issue.

I looked up m-Cresol, it's pretty nasty! Zinc chloride also has antimicrobial properties.

I've always wondered about glycerol, since it can be metabolized, but is used in so many applications.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Fri May 08, 2015 6:55 am UTC

Izawwlgood wrote:So after you withdraw the insulin, how do you cap the needle? Do you use a different needle on reloading? Dirtying the needle strikes me as an issue.

The entire system is disposable, including the needle. It's one-time use.

I looked up m-Cresol, it's pretty nasty! Zinc chloride also has antimicrobial properties.

Yeah, it's super harmful. Not sure why that's the preservative of choice. Novalog also uses phenol.

I'm actually not sure about the zinc. Normally it's used to allow hexamer formation, but fast-acting analogs don't form hexamers, so maybe it isn't used in pumps. Either way, the quantity would be far too small to matter.

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Contamination of insulin pump cartridges

Postby Neil_Boekend » Fri May 08, 2015 7:04 am UTC

I find it odd that the cartridge is refilled at all. I always assumed it would just be a slot into pump, recycle when empty thing. Exactly because of contamination issues.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Fri May 08, 2015 8:02 am UTC

Neil_Boekend wrote:I find it odd that the cartridge is refilled at all. I always assumed it would just be a slot into pump, recycle when empty thing. Exactly because of contamination issues.

It is exactly that. They are filled once, used until empty, then discarded.

Some companies have considered pre-filled cartridges, but there are a number of logistical and quality-control challenges, so the idea was basically abandoned.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Contamination of insulin pump cartridges

Postby Izawwlgood » Fri May 08, 2015 11:38 am UTC

So how often do you replace the needle, and between that, how often are you refilling the cartridge and injecting yourself?

Eebster the Great wrote:Novalog also uses phenol.
m-Cresol is a derivative of phenol. Take a look, they're very similar chemically.

Eebster the Great wrote:It is exactly that. They are filled once, used until empty, then discarded.
Wait, so... Ok, I'm really confused what you're saying is going on here now.

You have a syringe thing, with a needle. You fill it up from a vial, and then use it x times. When you're done with it, you throw it away? How many times do you draw from the vial?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Fri May 08, 2015 12:42 pm UTC

Izawwlgood wrote:So how often do you replace the needle, and between that, how often are you refilling the cartridge and injecting yourself?

This is a cartridge for an insulin pump. You do not inject yourself with it.

Eebster the Great wrote:Novalog also uses phenol.
m-Cresol is a derivative of phenol. Take a look, they're very similar chemically.

Yeah of course. It's just 3-methylphenol.

Eebster the Great wrote:It is exactly that. They are filled once, used until empty, then discarded.
Wait, so... Ok, I'm really confused what you're saying is going on here now.

You have a syringe thing, with a needle. You fill it up from a vial, and then use it x times. When you're done with it, you throw it away? How many times do you draw from the vial?

The cartridge comes empty, attached to a plunger on one end and a needle in plastic housing on the other. You fill it with insulin from a glass vial, first pushing 3 mL of air into the vial, then withdrawing 3 mL of insulin. Then you unscrew the needle and plunger and dispose of both. The cartridge is placed in the pump and connects to plastic tubing, which connects to the infusion set in your stomach or leg. A motorized plunger in the pump administers the insulin as directed until the cartridge is empty. Then you dispose of the cartridge, rewind the pump's plunger, and start over.

A cartridge is only every filled once. I really don't know why I need to keep repeating this. But the same vial is used to fill multiple cartridges. The vial contains 10 mL of insulin, so on the fourth cartridge, you will want to withdraw the last 1 mL from the old vial and the first 2 mL from the new vial. In order to do so, you need to first withdraw 1 mL of insulin and 2 mL of air from the first vial, then push the 2 mL of air into the next vial before withdrawing 2 mL of insulin. Every time you do this, you are potentially contaminating the new vial with any colonies present in the old one. Since you do this pretty much every time (as I said, the numbers do not work out as neatly as they would seem to, so rather than just two thirds of the time, this happens almost every time), if one vial becomes contaminated, you have the potential to immediately contaminate each new vial you use for the rest of your life.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Contamination of insulin pump cartridges

Postby Izawwlgood » Fri May 08, 2015 12:58 pm UTC

Because the way you outlined it at first was not entirely clear to me - I apologize for asking you to clarify.

It seems to me the only point of contamination (the one time needle on the cartridges) makes for a pretty low point of contamination likelihood. Is the needle sheathed or covered or such before it's used?

Also, the chance of passing on contamination seems to really only occur when you switch to a new vial - considering each vial has two antimicrobial agents, I imagine the only way contamination can occur is if it's particularly egregious (i.e., you don't inject 3mL of air into the vial, but 3mL of dirty water/spit), or if a vial has been compromised (broken, left at room temp for too long, etc). I also wager such a contamination would be quite evident, though I'm not sure if the cartridge or vial are made with clear glass, since you mentioned light can degrade insulin.

How often do you clean the whole system? Like, flush it out with hot water or scrub any gunk or such off the cartridge/pump connection?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
eta oin shrdlu
Posts: 451
Joined: Sat Jan 19, 2008 4:25 am UTC

Re: Contamination of insulin pump cartridges

Postby eta oin shrdlu » Fri May 08, 2015 7:37 pm UTC

It seems like you could eliminate the contamination problems if you allow a little waste (but less than the 10% you're avoiding by using the last 1mL), by filling from the new vial first. This would mean you'd have to estimate the amount left in the old one so that you know how far to fill it, though. (Weigh it on a postage scale?)

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Fri May 08, 2015 11:52 pm UTC

Izawwlgood wrote:It seems to me the only point of contamination (the one time needle on the cartridges) makes for a pretty low point of contamination likelihood. Is the needle sheathed or covered or such before it's used?

The risk of contamination is one vial to the next, not directly from the needle. Yes, it's shipped in plastic.

Also, the chance of passing on contamination seems to really only occur when you switch to a new vial - considering each vial has two antimicrobial agents, I imagine the only way contamination can occur is if it's particularly egregious (i.e., you don't inject 3mL of air into the vial, but 3mL of dirty water/spit), or if a vial has been compromised (broken, left at room temp for too long, etc). I also wager such a contamination would be quite evident, though I'm not sure if the cartridge or vial are made with clear glass, since you mentioned light can degrade insulin.

The vials are clear, but there is virtually no probability of spotting a bacterial infection the way you imagine. There is only one antibiotic, m-cresol, which doesn't kill 100% of bacteria.

How often do you clean the whole system? Like, flush it out with hot water or scrub any gunk or such off the cartridge/pump connection?

Never, but the connection to the pump isn't really a risk anyway, since it's just a plunger. If the rubber seal broke, it's possible bacteria could get in that way, but only one cartridge works be affected.

eta oin shrdlu wrote:It seems like you could eliminate the contamination problems if you allow a little waste (but less than the 10% you're avoiding by using the last 1mL), by filling from the new vial first. This would mean you'd have to estimate the amount left in the old one so that you know how far to fill it, though. (Weigh it on a postage scale?)

I've been thinking the same thing. By changing the order, there should be no risk of contaminating the new vial. It's not important that I estimate exactly, since wasting a few units is not a huge deal.
Last edited by Eebster the Great on Sat May 09, 2015 12:00 am UTC, edited 2 times in total.

User avatar
bigglesworth
I feel like Biggles should have a title
Posts: 7461
Joined: Sat Apr 07, 2007 9:29 pm UTC
Location: Airstrip One

Re: Contamination of insulin pump cartridges

Postby bigglesworth » Fri May 08, 2015 11:55 pm UTC

Eebster the Great wrote:Some companies have considered pre-filled cartridges, but there are a number of logistical and quality-control challenges, so the idea was basically abandoned.
What, like Novo Nordisk's Novorapid and Levemir 10ml vials that have a big market share? At least in the UK they're huge.
Generation Y. I don't remember the First Gulf War, but do remember floppy disks.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Sat May 09, 2015 12:01 am UTC

bigglesworth wrote:
Eebster the Great wrote:Some companies have considered pre-filled cartridges, but there are a number of logistical and quality-control challenges, so the idea was basically abandoned.
What, like Novo Nordisk's Novorapid and Levemir 10ml vials that have a big market share? At least in the UK they're huge.

I don't think you understand. Insulin vials are large, glass, and partially filled with air. You can't load them directly into pumps. You need to fill plastic cartridges from the vials. That's the whole point. Also, insulin detemir (Levemir) is slow-acting and not used in pumps. NovoRapid is the British brand name for insulin aspart (called Novolog in the U.S.).

E: OK, apparently 3 mL prefilled glass cartridges for insulin pens exist in the U.K. They don't seem to exist for pumps. There is a difference in management, since pens are turned down for infection, allowing any air to float away from the syringe, while most pumps are operated upright. They also use different materials, and you look at the pen for every injection, unlike the pump cartridge.

E2: I guess this exists now too: http://www.snappump.com/videos/Prefille ... are-a-Snap.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Contamination of insulin pump cartridges

Postby Izawwlgood » Sat May 09, 2015 2:28 am UTC

Eebster the Great wrote:Never, but the connection to the pump isn't really a risk anyway, since it's just a plunger. If the rubber seal broke, it's possible bacteria could get in that way, but only one cartridge works be affected.
Without seeing it, I'll have to take your word - I imagine the pump-cartridge connection to be just as much a contamination risk as the needle-vial point.

Eebster the Great wrote:There is only one antibiotic, m-cresol, which doesn't kill 100% of bacteria.
You said there was zinc chloride in the mix as well.

In any case, I think the suggestion of drawing from the new vial first, and estimating the remainder from the old vial is a pretty simple solution to your problem!
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Sat May 09, 2015 2:59 am UTC

Izawwlgood wrote:
Eebster the Great wrote:Never, but the connection to the pump isn't really a risk anyway, since it's just a plunger. If the rubber seal broke, it's possible bacteria could get in that way, but only one cartridge works be affected.
Without seeing it, I'll have to take your word - I imagine the pump-cartridge connection to be just as much a contamination risk as the needle-vial point.

There is no contact between the pump and the insulin. The motor pushes on the bottom of the cartridge, which slides up. There is a rubber seal to keep the insulin in and bacteria out. Even if there are cracks in the seal, only that one cartridge would be at risk.

Eebster the Great wrote:There is only one antibiotic, m-cresol, which doesn't kill 100% of bacteria.
You said there was zinc chloride in the mix as well.

Possibly in micromolar quantities, but I'm not sure. It's usually only for intermediate or long-acting insulin.

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Contamination of insulin pump cartridges

Postby qetzal » Sat May 09, 2015 3:11 pm UTC

The US label for Novolog says the following:

NovoLog is a sterile, aqueous, clear, and colorless solution, that contains insulin aspart 100 Units/mL, glycerin 16 mg/mL, phenol 1.50 mg/mL, metacresol 1.72 mg/mL, zinc 19.6 mcg/mL, disodium hydrogen phosphate dihydrate 1.25 mg/mL, sodium chloride 0.58 mg/mL and water for injection. NovoLog has a pH of 7.2-7.6. Hydrochloric acid 10% and/or sodium hydroxide 10% may be added to adjust pH.


Assuming that's 19.6 mcg/ml as zinc (not as ZnCl2), that corresponds to 0.3 mM. Not sure how much Zn is needed to be antimicrobial, but I'd guess that's too low.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Sat May 09, 2015 3:48 pm UTC

I think it's just to balance the charge. U100 insulin is about 0.6 mM (for monomers, though it actually forms dimers in formulation), and the monomer has a -1 net charge at neutral pH. For native insulin, one zinc ion coordinates the hexamer, while the other five float at semirandom positions around the outside. For aspart (NovoLog), there are no hexamers, but there is still a net charge.

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Contamination of insulin pump cartridges

Postby qetzal » Sat May 09, 2015 9:16 pm UTC

There's plenty on sodium already for charge neutralization. I'm pretty sure they wouldn't bother adding zinc just for that. Most likely the zinc helps stabilize the monomers or something.

I'd also guess that someone in pharma and/or FDA (and their equivalents in other countries) have already considered the issue you've raised, but I'm not sure where you would find it written up - except probably in regulatory submission dics that aren't readily available.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Sun May 10, 2015 2:19 am UTC

qetzal wrote:There's plenty on sodium already for charge neutralization. I'm pretty sure they wouldn't bother adding zinc just for that. Most likely the zinc helps stabilize the monomers or something.

Zinc has no effect on monomer stability. It's possible it has some effect on dimer stability, but I doubt it. Zinc is excluded from the dimer interface. Zinc is just the cation of choice because that is how insulin exists natively. The sodium chloride is just to add salinity and the sodium hydroxide alkalinity. Incidentally, phenol does have a very profound effect on insulin stability, but probably only at higher concentrations.

I'd also guess that someone in pharma and/or FDA (and their equivalents in other countries) have already considered the issue you've raised, but I'm not sure where you would find it written up - except probably in regulatory submission dics that aren't readily available.

They've certainly considered it, which is the reason the formulation requires preservatives. This web page describes basically the justification I gave in this thread, and also mentions that the oral insulin spray does not use preservatives (probably since meta-cresol would inflame your lungs the way it inflames infusion sites). But I haven't seen anywhere the idea of serial contamination mentioned.

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Contamination of insulin pump cartridges

Postby qetzal » Sun May 10, 2015 5:21 am UTC

Eebster the Great wrote:Zinc has no effect on monomer stability. It's possible it has some effect on dimer stability, but I doubt it. Zinc is excluded from the dimer interface. Zinc is just the cation of choice because that is how insulin exists natively.


I searched a bit but couldn't find a good explanation of why Zn is added to insulin aspart. However, I'm sure it's not for charge neutralization. There are lots of proteins that have a negative charge at physiological pH. Most of them are quite happy without any Zn or any other divalent cations. A modest amount of NaCl is usually all that's necessary.

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Sun May 10, 2015 6:01 am UTC

Not sure then. It's stoichiometric, so the zinc presumably bonds in some way.

User avatar
Minerva
Posts: 947
Joined: Sat Nov 11, 2006 2:58 pm UTC
Location: Australia
Contact:

Re: Contamination of insulin pump cartridges

Postby Minerva » Wed May 20, 2015 10:31 am UTC

You may find this interesting with regards to what the Zn does.

http://www.europeanreview.org/wp/wp-con ... ds/176.pdf
...suffer from the computer disease that anybody who works with computers now knows about. It's a very serious disease and it interferes completely with the work. The trouble with computers is you play with them. They are so wonderful. - Richard Feynman

User avatar
Eebster the Great
Posts: 3487
Joined: Mon Nov 10, 2008 12:58 am UTC
Location: Cleveland, Ohio

Re: Contamination of insulin pump cartridges

Postby Eebster the Great » Thu May 21, 2015 2:56 am UTC

It sounds like Lyspro at least (and presumably also Aspart) does form zinc-coordinated hexamers at the comparatively high pharmaceutical concentrations, then immediately dissociates after injection, which makes sense from the standpoint of thermal stability.


Return to “Science”

Who is online

Users browsing this forum: No registered users and 6 guests