Report: Medtronic files for FDA approval of MiniMed 670G ‘artificial pancreas’

Diabetes Technology Insulin Pumps
21

From the article on the FDA monitoring the AP movement:

Dr. Lias said physicians should ask these questions if patients say they
are using a DIY artificial pancreas: Do you understand exactly what
algorithm is being used? Is it right for you? Have you checked the code
to ensure it implements the algorithm correctly?

This made me do a double-take. Open APS is open source. Not only can I read the code, I can modify it to fit my needs, fix bugs, or if they make a change I don’t like, go back and use the older code. I can’t say the same thing for any of the commercial devices. This feels very hypocritical to me.

22

I’ve been wondering if it might be possible to achieve good control using this system with a calibration offset (ie if my BG is 80, I could calibrate to 140 and hope to average 90).

There is a danger, though, that the signal may not be linear and this would distort the data… does anyone have a Medtronic CGM and know roughly what the linear range is for BG vs ISIG?

I’m frankly horrified that Medtronic chose such an unambitious target as an average of 150. Maybe this higher target is necessary for safety in people who eat large quantities of simple carbs? Has anyone tried OpenAPS with a very high carb diet?

23

150 isn’t an odd target. The target A1C for everyone with diabetes is 7.0 from the which is in that ballpark (http://www.ndei.org/ADA-diabetes-management-guidelines-glycemic-targets-A1C-PG.aspx.html). I think AACE may be a shade lower at 6.5 but that still works out to an average c. 145 mg/dl.

I’m an Enlite user but am not sure about the answer to your questions re ISIG. I am pretty engaged with my diabetes but am also horribly lazy and kind of roll with it.

I’ve never looked into OpenApps either. The Medtronic stuff worked ok to get things nailed down although it hasn’t worked since I got Windows 10. #technologicalterror

24

I don’t think a higher target is beneficial for eating carbs. I usually run 150g/ day but am all over the place, some days a lot more and some days a lot less. I did a week of 50g of carbs and it was fine but I like carbs.

1 Like
25

At an average A1c of 6.9 its not good enough. With t:slim and G4 achieving 6.2.

26

So you might be able to tweak the 670g by lying to the sensor when you calibrate. If you want an average of 90-100, calibrate your BG at 120 when your meter tells you it’s 100. I can’t openly say on a forum how I know this or even recommend it, just saying this possibility may exist. If you email me privately I can explain more.

1 Like
27

It seems like that might work however then you’d sort of be doing pumping. I want an AP that wants my average to be 90, or 85. I use 85-90 as the “target” for my Medtronic 630 with Enlite and but I the threshold suspend off. I also will do “moving” calibrations, if my BG is like 63, I’ll have a few G of fast carbs and caiibrate, or even bolus and eat as if I were at 70 as it seems to work out ok most of the time, result in reasonably accurate calibrations and helps me achieve decent control, without drying out dinner waiting 45 minutes for my BG to come around!

1 Like