Why don't we inject insulin intravenously?

I had a thought in the shower today (as one does). Two of the major obstacles to better glycemic control are the slow absorption of subcutaneously injected insulin and the delayed/less accurate blood glucose readings available subcutaneously. A healthy body can both infuse insulin directly into the circulatory system and detect blood glucose levels accurately. So why don't we inject insulin intravenously? And wouldn't a continuous glucose monitor with a probe in a vein instead of in subcutaneous tissue be significantly more accurate?

I suppose injecting insulin into veins might be too rapid, at least to allow single bolus injections, it might have to be trickled in, as the body does. But at least for a pump that would seem to be perfectly feasible. And while setting up an intravenous line is trickier than placing a subcutaneous infusion set, it's something with which plenty of folks with other conditions have to deal. I assume if a person with diabetes is checked into a hospital, insulin does get delivered intravenously (or maybe I'm wrong about that).

As for placing a CGM sensor in a vein, clearly there are various challenges and potential health risks associated with doing that, but I wonder whether it's something that's been explored much (and if not, why not).

Any doctors/RNs/others on here who might shed some light on this? For my part, I'm none of the above, the only formal medical training I have is as a former EMT, so these are just the musings of a diabetic who likes to better understand his condition and related issues.

I have seen posts here from folks inject in a vein when the BG is really high. I would love to know how and feel comfortable doing that, but only with a high BG. The four hour wait to bring a high BG down is torture.

Interesting, I was thinking of that as one occasion that seems to warrant intravenous injection. Though I wonder how one manages the dosing, because presumably the usual ratios, derived based on subcutaneous dosing, don't apply. I have no idea what typical blood concentrations of insulin might be, but I could imagine that injecting a big bolus to try to bring down a big BG high might actually be quite dangerous, because it would raise blood insulin levels so dramatically. Or maybe not. Where's an endocrinologist or perhaps any MD to chime in when you need them?

It seems to me that MDs must intentionally avoid commenting in public forums like this— or maybe they just don’t acknowledge that they are a MD… Probably a code of ethics / professionalism sort of thing

Yep, makes sense, huge potential liability issues.

I don't have enough knowledge to know if this would be feasible or not, but I can tell you that every time I've been hospitalized and on an IV, the insulin was still given by shot, not through the IV.


I understand that hospital intensive care units use IV insulin for post surgical diabetic patients. They also deliver dextrose when needed to bring BGs up. I also understand that this protocol is highly effective at tight BG control.

One of our members does dose IV shots once in a while, but I’ve seen him characterize it as risky and possibly dangerous. I suffered a 15-hour hypo induced hyperglycemia yesterday. I would have gladly tried an IV dose if I knew how to do it and that it would be effective.

My experience with hypo caused rebounds is that sub-q injections have very little to no effect on my BGs. It takes patience and forgoing meals to bounce back.

The only two times I've had intravenous insulin was when I was hospitalized with DKA. They gave me insulin through IV at a slow drip. I also had glucose administered me intravenously once in the ER because I was sick with vomiting and diarrhea and couldn't keep anything down to bring my BG up. So they shot glucose directly into my vein...boy was that stuff thick and painful!

I've had another thought about IV insulin. Sometime during the last year TuD featured a live video with a researcher that was testing an agent that when added to insulin would enhance the effect it had on the liver. If memory serves, in a gluco-normal person, the release of insulin is felt immediately by the liver and that subdues the liver's release of glycogen.

Perhaps when insulin is administered via IV that that suppressing effect on liver sugar happens as well. If that's true then I wonder if IV insulin could have rescued me yesterday when I was stuck in high-BG misery for 15 hours due to a hypo-induced rebound high.

The pancreas dumps insulin into the portal vein which is a direct route to the liver right next door, which is why human insulin, which is actually slower-acting than the rDNA human-analog fast-acting insulins we use, can still control BGs. An injection elsewhere in the circulatory system wouldn't be quite as effective, but since it doesn't take blood very long to circulate, it would be an awful lot faster than a subcutaneous injection that only slowly seeps into the bloodstream.

But I don't know how much insulin the pancreas can actually dump at once (and since healthy BGs stay within a fairly small window a firehose is probably not needed, even if someone eats a pretty high-carb meal), so dump is probably misleading, I guess it's more like trickle.

Ok, I found the video interview with Robert Geho, CEO of Diasome Pharmaceuticals. I found it interesting and educational. It's worth the investment of time.

The only time I have ever had IV insulin was when I was in labor with my son in 1990. The OB had instructed the nurses to do blood sugar checks every hour and start IV insulin to keep me under 100 throughout labor. I labored for 17 hours and they managed my blood sugar really well keeping it ~100 with intermittent use of dextrose and insulin via the IV. While it was an effective method to keep my blood sugars steady, it is not something I would want to do on my own nor would I want to wear a CGM sensor in my vein. As flawed as the system and devices I currently use, I am still able to maintain a non-diabetic A1C which is good enough for me.

Same here . . subcutaneous injection while hospitalized, even in ICU.

Even if IV is easily manageable with low risk for lows how about the logistics, the ease of subcutaneous vs intravenous injection. And in the case of pumping an IV connection is rather uncomfortable to wear on a 24 hour basis. Fine enough when your confined to a hospital but out and about living our lives. Not so much.

One thing that occurs to me is that IV use has limited "real estate". I would want to "save" my veins for emergencies (that I hope not to have). A friend who is an ER phlebotomist used to get frustrated by overused access by heroin addicts, finding a place for a blood draw and for the RN's to set up an IV gets harder and harder. Se'd have to use areas like between toes!

It works ok but is super dangerous. Only a dangerous lunatic would do anything like that!

Strong language, not sure I follow. If someone had a crazy high, and knew that a modest amount of intravenously delivered insulin could quickly correct it, in some respects that seems less dangerous to me than injecting inefficiently large amounts of insulin subcutaneously, being back at healthier BG levels some hours later, and then fighting a severe hypo some hours after that.

Of course, for me this is all hypothetical, both because I'm fortunate to not experience extreme highs for now, and because I have no experience with how I would dose intravenously (and no near-term intention to find out). So for now this is more in the category of thought-experiment, to better understand our shared condition, than something I'm actually considering doing.

That’s a disclaimer! I’ve done that for a while.

Ah, tone is hard to gauge online! Can you shed any light on the circumstances under which you do this, how you think about dosing, etc.? If you've already posted the answers in another thread, maybe provide a link? I'm intrigued by the mechanics of how something like this would work.

Most of my diabetes decisions tend to be situational and depend on how I would evaluate all of the variables cooking away at any given time, insulin on board, food on board, germs on board (the current situation...it seems that 200% basal is covering it for now though...), stress, exercise, planned exercise, etc. I would find that an IV correction would generally need less insulin than a pumped injection. I do all that stuff by SWAGGING it which is, of course, extremely dangerous.

Also, it's not really "mechanical", it's biochemical, which is why these things are so hair raising. Sometimes, you can bang away at a stubborn high (I look at anything above 130 as a "high" although, if I have enough IOB, I don't always act...) for a long time before it starts to move. The other potential advantage is that an IV shot might clear out faster but, if you don't catch it, you had better have a chute packed!!

Chute being glucagon, I guess. Though having both significant active insulin and glucagon flowing through your veins at the same time sounds a little scary. Do you inject glucagon IV also?

What does SWAGGING mean?

I'd love even more specifics on IV insulin if you're amenable. What insulin would you inject IV? How much under what circumstances, and how does that relate to what you'd do if you were injecting it subcutaneously? What is its action profile, i.e. how quickly does it start reducing BGs, when does it peak, when is it gone? Have you learned anything else by doing this that I'm not informed enough to ask about?

As you can tell, I'm pretty fascinated by your experience here.