My 8-year-old daughter started using MMInsulin pump last September and we keep having lots of trouble with airbubbles in the reservoir. The events happen at night usually, at 10:30pm when we discover high bg for no particular reason. We assume that it is caused by airbubbles in the reservoir which probably get into the tube and block the way of insulin. Sometimes we change everything(reservoir and set)and still find it difficult to normalise bg. Sometimes we have to wake the girl up every other night (around midnight) a few times to change the set, which makes her exhausted the next day. We have asked the world and his wife but no solution has been found so far. we are still investigating. Any ideas?Ildiko

most of the tiny air bubbles have no effect on blood sugar, and they wouldn’t normally stop or block insulin, they just push right through.

when I fill my reservoir I tap the sides with a pencil. “whack” might be a better description. this pushes the big bubble toward the transfer guard. then I use the plastic plunger to push the big bubble back into the insulin vial.

once the infusion set is attached, I give that reservoir another “whack” and manually push the bubbles through the set. THEN i put the reservoir back into the pump. then I “prime” (using the pump they way MM tells you to do it) to get the first drop out of the set. works for me. no bubbles worth mentioning.

using cold insulin often makes the bubbles worse - more gas can dissolve into the insulin when it’s cold. then it warms up and forms bubbles. (think soda bottle when you unscrew the top). my working bottle is room temperature, only my inventory is in the fridge.

I never, ever, change sets at night. for me, the irritation caused by the infusion set insertion can interfere with insulin absorption (I can run very high after a set change), I change them in the mornings, so I can keep a close eye on BS until things are under control.

good luck - sorry to hear your daughter is having troubles.

Hi Joe,
Thanks for your advice. We also push the big bubble out after filling the reservoir and I know of the temperature effect. We try to pay attention to these details. Despite this fact, after a day of using the new reservoir the bubbles are back - sometimes tiny, sometimes big.
Do you have to 'deair’the reservoir within the 2 or 3 days of use, or do you not get any bubbles whatsoever?
Best wishes, Ildiko

hi Ildiko

once I have knocked the big bubble out and most of the “champagne bubbles”, I do not really see any more bubbles in the line set. given the diameter of the tubing, I would have to see a bubble about half an inch in length or more, before I would even care to clear the line. in the last 5 years I haven’t seen any bubbles worth mentioning.

at set change time (day 3 or 4), I often observe a minor bubble, or some tiny bubbles clinging to the side of the reservoir and that is normal. insulin de-gasses when it warms up, when there is a change in pressure (such as in an airplane) and when it’s agitated such as it is on the belt of an active person.

good luck. let me know how big of a bubble you are talking about. reply here or inbox me.

Doubt the highs are related to the air bubbles. I’ve used a Minimed for 12 years. As long as you remove all the air from the reservoir when doing a site change, it is fine.

During the 3-4 days of wear you’ll have a few little bubbles appear. Love the description of “champagne bubbles” because that’s what they look like. But they are so tiny it won’t have any measurable impact on insulin absorbed. The truth is, not 100% of the insulin that’s pumped or injected is absorbed. It’s not an exact science, which is one of the variables that makes managing blood sugars such a challenge.

My guess is that your bedtime highs are either caused by a change in your daughter’s basal rate. You’ll know it’s this if she’s high every night. If it’s not that, then it’s probably inaccurate dinner carb counting. It’s easy to underestimate food at dinner and because our bodies slow down in the evening and we often eat bigger or high fat and carb food at night, it doesn’t show on the glucose meter until bedtime. Try to track certain meals and see what her glucose readings are at bedtime. If you notice consistent readings for spaghetti and meatballs or a hamburger or whatever you had for dinner, then doing the match of the amount of insulin needed for the correction dose you can use the dual wave bolus setting on the pump the give an immediate dose for the meal and an extended dose to cover the part of the meal that digests later.

Hope that helps. If I’m off track in my guess about what’s causing it, let me know your thoughts.

I’ve used a minimed pump for 12 years. As long as you remove the air bubbles when you fill a new reservoir, there shouldn’t be any problem.

The little champagne bubbles are normal and have no effect on blood sugars. The truth is, not 100% of insulin absorbs whether is given by pump or shot. That’s one of the variables that makes blood sugar management so exciting (and sometimes frustrating).

My guess is that your nighttime highs are caused by something else. If it’s happening night after night, the pump’s basal rate setting in the evening hours many need to be increased.

If it happens sporadically, it probably means you aren’t giving an accurate bolus dose for dinner. This is common at night because most people eat a bigger meal and often a higher carb or higer fat meal at dinner. There’s a phenomenon called “Chinese Restaurant Effect” where a big meal takes more insulin than the actual carb count would indicate. Big meals or high fat meals can digest slower too, so you don’t notice the increased glucose level until hours later. For a couple weeks, try to track specific meals, meal bolus given, blood sugar 2 hours after eating, and blood sugar at bedtime. If you notice patterns of highs, you’ll see how much insulin is actually needed to cover hotdish or pizza or whatever you had for dinner and can dose accordingly using the dual wave bolus on your pump. If you haven’t used this yet, it’s pretty cool for fatty food like pizza. Or you can just give a bigger meal dose and give a correction at bedtime.

These are just my guesses. If I’m off base, let me know what you think might be causing the highs. Take care.

I live at a mile high elevation and bubbles are the worst problem I have to put up with. I am a mechanical engineer, and I think the bubbles come in through the O-rings, because I often see small ones collect there. Then they coalesce into a larger bubble. It is possible air also enters thru the vent in the top of the reservoirs.

Two years ago I tried to get Medtronic to take this seriously. They still do not. They ran a simple test on the reservoirs to make sure they were made according to design, but they have never challenged the design itself, which is a much more complex endeavor. So, fed up with them, I reported the problem to the FDA and followed through two or three more times, sending them samples and describing what I see and think is going on.

I believe the O-rings are a bad design, because the O-rings are round in cross-section, while the slots they fit in are rectangular cross-section. This allows the O-ring to twist as the pump rotates the plunger upward (you may have noticed that the plunger moves upward by rotating within the cylinder of the reservoir. I think the twisting allows the air to get in.

Funny thing, I don’t remember this being an issue before I started using the newer 530g model. It might have been there, but I rarely noticed it. What I do now is look into the reservoir frequently, several times a day, and if I see a big bubble I remove it.

All of us who have observed this problem need to complain to Medtronic and to the FDA. search for MAUDE, then air bubbles in insulin pump. Here is an example where someone ended up in the hospital. Be sure to send every one to Medtronic and tell them you want the samples tested for cause. They will come back saying the product functioned as designed, but like I said, you have to challenge the design when you contact Medtronic or use the MAUDE system at

I encourage every T1 with Medtronic pumps to check their reservoirs each day and if there is a large bubble, take it out, call Medtronic and complain about it. There is strength in numbers.

Here is the case from MAUDE, which resulted in this person being put in the hospital:


Model Number MMT-XXX
Event Date 10/18/2012
Event Type Injury
Manufacturer Narrative
Currently it is unknown whether or not the device may have caused or contributed to the event as no product has been returned. No conclusion can be drawn at this time. We therefore consider this report complete to the best of our knowledge.

Event Description
It was reported that customer was hospitalized for diabetic ketoacidosis (dka). Diabetic ketoacidosis (dka) hospitalization was from a malfunction regarding air bubbles around the o-rings in the reservoir. Customer stated the reservoir contained air rather than insulin. Customer reported having had blood glucose levels of 500 mg/dl with diabetic ketoacidosis and ketones. Nothing further was reported.

Search Alerts/Recalls22

New Search | Submit an Adverse Event Report23

Manufacturer (Section D) MEDTRONIC MINIMED
18000 devonshire street
northridge CA 91325

Manufacturer (Section G) MEDTRONIC MINIMED
18000 devonshire street

northridge CA 91325 121

Manufacturer Contact christopher tupper
18000 devonshire street
northridge , CA 91325-1219
(818) 818 -8185
81857643 8185764313

MDR Report Key 4264716
Report Number 2032227-2014-55124
Device Sequence Number 1
Product Code OYC24
Report Source Manufacturer
Source Type Other
Reporter Occupation Other
Type of Report Initial
Report Date 10/22/2014
1 Device Was Involved in the Event
0 PatientS WERE Involved in the Event:
Date FDA Received 11/20/2014
Is This An Adverse Event Report? Yes
Is This A Product Problem Report? Yes
Device Operator Lay User/Patient
Device MODEL Number MMT-XXX
Device Catalogue Number MMT-XXX
Was Device Available For Evaluation? No
Is The Reporter A Health Professional? No
Date Manufacturer Received 10/22/2014
Was Device Evaluated By Manufacturer? Device Not Returned To Manufacturer
Is The Device Single Use? No
Is this a Reprocessed and Reused Single-Use Device? No
Type of Device Usage Unkown

Links on this page:
Page Last Updated: 11/30/2015

Note: If you need help accessing information in different file formats, see Instructions for Downloading Viewers and Players.

Accessibility Contact FDA Careers FDA Basics FOIA No Fear Act Site Map Transparency Website Policies FDA
U.S. Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993
Ph. 1-888-INFO-FDA (1-888-463-6332)
Contact FDA E-mail RSS Twitter Facebook YouTube Flickr
For Government For Press Combination Products
Advisory Committees
Science & Research
Regulatory Information
Emergency Preparedness
International Programs
News & Events
Training and Continuing Education
State & Local Officials
Health Professionals
FDA Archive
U.S. Department of Health & Human Services
Links on this page:

hi @davyboy,

hey I am a mechanical engineer as well, 25 years experience designing machines and control systems.

Help me understand what you are saying? MM pumps use the ram to push the plunger at the bottom of the reservoir. the top of the reservoir is engaged to the pump casing by a screw fitting mating the top of the infusion set and trapping the entire reservoir.

given this arrangement, the entire assembly is either neutral or in compression at all times. if the o-rings leaked, why would air be introduced? Wouldn’t leaking o-rings cause insulin to bypass the o-ring and the amount that leaked would then not be delivered through the infusion set? I can’t imagine air is introduced unless the ram is pulling on the plunger.

I have had infusion sets that didn’t engage the top of the reservoir, and then during manual priming the top of the reservoir wasn’t tapped or vented. the o-rings didn’t leak even though I pushed quite hard (harder than the stepper motor could anyway).

the source of bubbles that appear after priming are most typically gasses that evolve from the liquid. Insulin manufacturers charge inert atmosphere during processing and often times it is at 2-3 PSI, given time and refrigeration, a considerable amount of gas can dissolve into the liquid. once the atmosphere changes (is reduced) at the top of the free surface of liquid insulin, the gasses evolve and bubbles form. This can happen even in compression. the insulin in your pump warms to near body temperature and any dissolved gasses will tend to come out of solution in the form of bubbles. you don’t have to believe me, it’s explained fully in in chemistry texts.

you live at high altitude. I suggest you keep your working insulin bottle at at least 72F. I recommend you leave the working insulin vial slightly negative when you draw up a reservoir, this can be done by simply charging the vial with less units of air than the amount of insulin you are drawing out, then removing the transfer needle from the vial. These things together (warm and slight vacuum) will cause any dissolved nitrogen in your vial to evolve in the vial and not in your reservoir.

MM, like all medical device manufacturers, are required to take all complaints about their products seriously, record, and track them. The DKA event you reported on is unclear. Nor is there a response. This could have happened for a variety of reasons including but not limited to if the person didn’t prime the reservoir and put the pump back into operation with both air in the reservoir and the infusion set. Don’t forget that many people do not report all of the facts: it’s typically just what they experienced.

anyway cheers, hope you figure it out. good luck.

Joe, thanks for the discussion. I did not think that much gas could come out of solution in the reservoir. I always see the bubbles start around the O-ring. I realize the reservoir is supposed to be positive or neutral relative to its surroundings. One thing I didn’t mention was that the problem was not nearly as bad when I lived at sea level. I actually moved from a mile high town to sea level, and after a year came to another mile-high city, where the problem again became severe. I like your idea of warming the insulin up to near body temp before filling the reservoir.

Okay. I have warmed up the insulin in warm water, which is maybe 90 degrees, instead of room temp. I am now getting far fewer large air bubbles. I am also getting no more above 400mg spikes in glucose. Many thanks to Joe for the idea. If I have more problems, I will post here.