Share on Pinterest
Mike Hoskins/DiabetesMine

For many years at DiabetesMine, we featured a regular weekly column known as Ask D’Mine. It was a place where the community could ask questions, and get answers from Wil Dubois, a veteran type 1 and diabetes author who worked for over a decade as a diabetes educator at a rural clinic in New Mexico.

Questions we often received centered around insulin temperatures, particularly how cold insulin can get and if it’s still OK to use if a vial or pen has been frozen? Whether it be the result of frigid winter temperatures outside, power outages that lead to no heat, emergency situations, or storage that might lead to a person with diabetes (PWD) putting their insulin on ice for too long… it’s a common question from the Diabetes Community.

This particular question came in to Ask D’Mine, from Mary in North Dakota who lives with type 1 diabetes: “Lots has been written about insulin and heat, but what about insulin and cold? How cold can insulin get and still be “OK?” I know we store it in the fridge, but can it freeze? Well, of course it can, but is it more like water, or more like anti-freeze? How cold does it need to get before it freezes solid? If it did get frozen, can you thaw it back out and still use it?

With that topic in mind, Wil experimented with his home freezer one time to see firsthand how cold insulin could get before it proves unusable. He also researched the issue, consulting insulin manufacturers and product labeling, to offer the best guidance possible. Read on: you might just get chills hearing what he discovered!

For the sake of science, I put the last dredges of a vial of Novolog into my kitchen freezer last night. Now, there wasn’t that much left, maybe only 20 units or so, but this morning I was rewarded with Novo-ice in my Novolog vial.

How cold is my freezer? I have no idea. It’s a garden-variety Kenmore. It will make ice cubes and turn Häagen-Dazs into a solid rock, while leaving my pecans soft enough to eat right out of the freezer. So it’s pretty much just like every other freezer in the country.

My son Rio chilled a glass of wine in the freezer for me this summer, but I got home late and the vino was about half frozen. On the other hand, we chilled some whiskey shots overnight with no whiskey-ice at all. So from all this comparative science, we cansafely infer that the freezing point of insulin is much closer to that of water than it is to wine (typically 13.5% alcohol) or to whiskey (typically 40% alcohol).

Therefore the answer to your question is: Insulin is more like water than anti-freeze.

Share on Pinterest
Wil Dubois/DiabetesMine

So, if you left your insulin in your car overnight in many areas of the country right now, you’d have a block of insu-ice by morning. Of course, a full vial might not have fared as well as my science experiment did. The glass vial could have cracked as the freezing insulin expanded, or the pressure might have blown the top membrane off and sent quasi-frozen insulin everywhere, making your Häagen-Dazs taste like Band-Aids (a Ben and Jerry’s flavor that never worked out).

One thing I did notice was that the insulin thawed much more quickly than I would have expected when I was holding the chilled vial in my hand to try to get a picture of it. That might just be my warm hands and cold heart. Or the small size of the insu-ice cube.

But is the insulin still usable after it thaws? Is it like Shackleton’s 100-year-old whiskey from the South Pole expedition – frozen, but preserved for all time? Ready to use when returned to balmier climates?

Insulin getting frozen happens a lot more often than you’d think. Check out the discussions about it here. Or here. Or here.

Much of the conversation online revolves around how to get replacement insulin from your pharmacy or health insurance plan; or verbatim quotes from the prescribing information sheets about the dangers of frozen insulin. And what do they say?

The prescribing info for Novolog specifically states, “Do not freeze NovoLog® and do not use NovoLog® if it has been frozen.” And it says it all bold-y like that, so we know they mean business. The same kind of wording can be found on Apidra, Humalog, Lantus, and Levemir insulins. So the folks who make the insulin think freezing it is a bad idea. But does that really tell us that freezing is a problem, or just one they choose not to study? If freezing really did preserve insulin, then expiration dates might be out the window. We could stockpile. Profits might plummet.

So what are the real facts? Has it been studied? Is there any word on frozen insulin from outside the manufacturers? The BD Diabetes site says that “once insulin is frozen, it loses its potency.” And the Wisconsin Department of Health agrees.

But wait a sec. Endogenous human insulin can be frozen. In fact, blood samples for insulin level tests are supposed to be frozen. And look here: a study at the Collaborative Studies Clinical Lab at Fairview-University med center showed that human insulin in blood “is stable up to five freeze-thaw cycles.”

And you might find this case report interesting. It’s about a 28-year-old guy who thawed his frozen insulin in a 600-watt microwave oven (on defrost mode). It ended… umm… badly… for him. But that said, his actions aren’t really as stupid as they look at first glance. Frozen plasma can be stored frozen in hospitals and thawed in microwaves in emergencies.

Was it the freezing or the microwaving that trashed his insulin? This is an older case, but a follow-up study at the time looked at freezing and microwaving vs. freezing and room-temp thawing. Of interest, his “R” type insulin survived either process just fine, while his “N” type fared much more poorly. Freezing apparently changed the “N,” in a way that caused it to be clumpy, so that when drawn up, the suspension wasn’t consistent.

Would that mean a modern basal would be more sensitive to freezing than a modern rapid-acting? I think that’s too far a stretch, but based on this study, I’d sure has hell throw away any mix-type insulin that got frozen. The juice in the modern mix is different, but the suspension fluid used is still zinc protamine, the same clumpy culprit behind the 28-year-old’s misadventures in ice-land.

What about the modern basal and fast insulins? Well, modern basals work a lot differently than “N,” and for that matter, Lantus and Levemir couldn’t be more different from each other in how they work. Lantus is probably somewhat closer to “N,” but much higher-tech; while Levemir is closer to the “R.” But either way, I think I personally would only use a modern basal that was frozen in an emergency.

But, hey, if my fast-acting got frozen I think I’d try it out. Cautiously. With lots of fingersticks.

And a cold, stiff drink.


“This is not a medical advice column. We are PWDs freely and openly sharing the wisdom of our collected experiences — our been-there-done-that knowledge from the trenches. But we are not MDs, RNs, NPs, PAs, CDEs, or partridges in pear trees. Bottom line: we are only a small part of your total prescription. You still need the professional advice, treatment, and care of a licensed medical professional.”

Wil Dubois lives with type 1 diabetes and is the author of five books on the illness, including “Taming The Tiger” and “Beyond Fingersticks.” He spent many years helping treat patients at a rural medical center in New Mexico. An aviation enthusiast, Wil also works as a private flight instructor. He lives in Las Vegas, New Mexico, with his wife and son.