Highs when flying and Lows when landing

Have any of you experienced unwarrented lows after a flight? It doesn't always happen on short flights (2 hours or less), but on long flights I have been experiencing highs (cgm alarms on a 240+). I bolus the high. Upon landing, I exit the plane and my bg's are in the 40's. I have experienced this several times so far over the holidays. I read that the Omnipod isn't as sensitive to 'pressure' as the corded pumps. I don't know what it is. Am I getting too much insulin with the bolus I give on the plane; should I be satisfied with a higher bg during the ride? I get off of the plane, and I need to stop for sugar before I continue. Anyone else have or cure this roller coaster problem?

On long flights to Europe or Hawaii, my two year old and his Omnipod will usually go high for two hours and then spend the next 8-10 hours spectacularly low. After the initial highs, I usually have to turn his pump off for the rest of the flight. I have no explanation.

Hmmm, interesting. With the lower pressure of a flight, small air bubbles (microscopic) may form inside the reservoir and tube, leading to you getting less insulin that normal.

On landing, and pressure returning to normal, the air bubbles collapse and you get more insulin than normal.

I would re-prime on attaining cruise altitude, and again on landing.

Do you bolus the high or do you just wait it out?


This might explain what is going on. I recently switched from OmniPod to the t:slim pump. I experienced the same issue on flights to and from California and the mid west.

I appreciate the comments! Would Omnipod react the same way as a tubed pump with regards to pressure and bubbles?

You should also suspect your CGM, but I'm assuming you verified the high before doing a bolus.

There were reports of excessive basal insulin delivery, possibly because of the reduced pressure causing some pumps to over-deliver. I've never experienced this effect and I've been on a number of 11 hour flights since I started using the Omnipod.

Generally long haul flights go to a higher altitude than short haul ones, and the pressure in the cabin will go lower because it is decreased gradually.

I do know that the worst aspect of air travel for a diabetic is the food; don't eat it, it invariably consists of a large amount of high glycaemic index carbs; bread, potatoes, sugar. I did once have a spectacular high because I requested diet sprite and got non-diet, I didn't check and you can guess what happened.

I also always eat on flights (pretty much a requirement on an 11 hour flight!) I actually eat quite a lot compared to other passengers - we stock up on salads in the departure lounge and start eating within a few minutes of take off. I generally consume wine with the food which might be relevant.

Lowering pressure happens fairly rapidly I believe because it tracks the aircraft's passage to cruise altitude. Since that can take 15-20 minutes I suspect that your experience is the result of one of the physiological changes that happen during pressure changes. I'm guessing here, but the obvious explanation is a temporary reduction in your rate of insulin absorption, so your basal is temporarily reduced but the insulin is still there; when your body adjusts you get a boosted basal.

Given the symptom the obvious answer is not to bolus for the high. However you have a CGM trace! So you know exactly what happened, though not why. You have the data up until the time of the bolus, so you can simply do a temp basal rate change to match the bolus then, at the time where the CGM trace shows your blood sugar dropping (I'm guessing maybe 30 minutes after take-off) counter the original increased basal by dropping (or even stopping) your basal for the another 30 minutes.

You could also ignore the CGM alarm and collect the data without any bolus, on a mid-length flight where you don't eat. Then you will know how long after take-off before your insulin response returns to normal - the point where your blood sugar stabilises.

I don't think it matters whether the problem is purely physiological or whether it is caused by the Omnipod; the answer is the same.

John Bowler

I bolus the high, but nothing explains the 6-8 hours of 40'ish lows that I experience on the rest of the flight. Now I know to wait to see if I really want to bolus that high and let him ride high for a while.

Thanks John! My high looked like a steady climb. Sometimes my pod only lasts 2 days, so I have seen that type of climb before (I was into the 3rd day on the pod)....not a spike...gradual without dips (looks like it isn't coming down again and on the 3rd day sometimes it doesn't until I change the pod).
Since their wasn't a good justification for the high...other than adrenalin, I think you are right John, I shouldn't bolus it.
My cgm wasn't catching up with the lows. They came on fast, and by the time I left the terminal and had the opportunity to test, they were plenty low...44 off of one flight...49 the next time.
The next time, I am going to pay particular attention to my cgm and try to notice the time...as Jack's mom does.

Curious. Adrenaline causes a reduction in the skin blood supply - that's why we go white. The 'pod injects insulin about 5mm below the surface of the skin and adsorption is, I think, dependent on the skin blood supply.

That might be all that is happening.

John Bowler

It must be noted by others that you are a Type 2 diabetic. Type 1 diabetics should in general NOT request a diabetic meal because airlines always assume they are Type 2, and give them low-carb and diet drinks.

A tubeless pump should react less than a tubed pump, but I don't know to what extent.

I'm Type 1, I've tried the diabetic options a couple of times over the 40+ years I've been a diabetic and neither time were they low carb, indeed on both occasions they were actually higher carb than the regular meal option.

I believe the diabetic meals might be *low fat*, not low carb.

What I do see consistently is that the diabetic meals have the triple carb overload thing: rice/potatoes *and* bread *and* fruit. Worse this is often not backed by fat, which at least seems to slow down adsorption a little for me.

John Bowler

Interesting, but I don't see why. I guess the tube can expand or contract, but the contents are liquid and therefore not significantly compressible. (I.e. a liquid filled tube will hardly expand at all if the exterior pressure is halved.)

John Bowler

I recently switched to the t:slim insulin pump. It is a tubed pump with a entirely different delivery mechanism. All other tubed pumps deliver insulin by pushing from behind the entire contents of the reservoir. Have you ever noticed the push back that you get when you draw up insulin from a new bottle of insulin. I usually use around 150-160u in three days. Most of the time I get what I call an auto-fill from a new bottle. The t:slim doesn’t have a motor pushing a plunger; it draws micro doses from the reservoir, then pushes that into the tubing. Therefore…no problems with change in pressure.

That being said…I believe that, although the OmniPod does Not have tubing, it does have a motor and piston pushing on the entire contents of the reservoir.

Simon's explanation would appear to me to be the biggest factor in this. I can say that over the years of flying, even when I was on MDI, I had highs during trips then too. But any change in atmospheric pressures could affect any numatic pump, bubbles or not. I don't know if enough research has been done on this to recommend any course of action.

Ability to contract depends on the liquid e.g. water is very hard to compress. But bubbles can form in any liquid under low pressure, which is what we're talking about.

The tube doesn't make any difference to that; that effect is determined by the total volume of water (insulin) in the reservoir. I don't know how much a typical tubed pump holds, but the Omnipod holds 2ml, and this will will cause a *maximum* of 1.2IU insulin to be released if the insulin in the pump is saturated with air.

There are all sorts of reasons why this is an upper limit. Firstly I assumed 9000ft cabin elevation (an upper limit) and secondly I assumed inconsistent temperatures (see below) to maximise the result.

Much more important, I assume the insulin is fully aerated. I suspect this is unlikely and it is easy to prevent with a vial - just suck 30% of the air out of the vial *after* each time you fill the pod (so now the insulin is being stored at 9000ft effective altitude.)

I use a pen, not a vial, and so I don't inject air in to balance the pressure. The insulin probably isn't aerated as a result (but I don't know the manufacturing process.)

Detailed calculations:


Then run the calculations in the page for sea level vs 9000ft. We know the temperature is less than 98.4F, I used 77F, an underestimate (so overestimate of the dissolved gas):

Moles of gas dissolved at 2ml of water at 1 atmosphere pressure, assuming 21% Oxygen and 89% Nitrogen (I ignored the CO2; it behaves differently because it reacts with the water, replacing it with Nitrogen overestimates the total effect.)

Oxygen: 0.002 * 0.21/756.7
Nitrogen: 0.002 * 0.89/1600

The difference in moles between the Oxygen+Nitrogen dissolved at sea level (1atm) and 9000ft (about 0.714atm), using Henry's law (reference above) which says that the dissolve gas is proportional to pressure, is therefore 28.6% of what was original dissolved:

(1-0.714) * 0.002 * (0.21/756.7 + 0.89/1600) ~= 5E-7M

Volume at 98.6F (37C - overestimate) of gas assuming the ideal gas law (and from the well know 22.4l volume of 1M at STP):


(273.15+37)/273.15*22.4l = 25.4l

So the volume of gas released is 0.012 ml, which is about 1.2IU of insulin.

Thanks for doing the maths :-)

The medtronic pump I used to use had a 3ml reservoir, so when full the effect could be even greater.

How the 1.2 units impacts you depends on your ISF. For me, 1.2 units less would cause a rise of 3mmol (54mg), but for my daughter, it would cause a rise of 14mmol (252mg) depending on the time of day.

It's 1.2IU more, not less; it corresponds to a lowering of blood sugar not a rise. The dissolved air comes out of solution and pushes extra insulin into your body.

The tiny air bubble in the reservoir that has been discussed before has the same effect; it expands with the decrease in pressure by about 40% (of the original volume). However the bubble is very small - if we assume about 1mm across that's about .1IU of insulin, so the expansion only accounts for an additional 0.04IU.

The fact that people experience highs after takeoff means this isn't the explanation. It also makes me very suspicious of the scare story that ran a while back about air travel causing some pumps to deliver extra insulin, resulting in possible lows.