Field trials. Taking the @MedtronicUK 640G and Smartguard into the Glastonbury 2016 Wilds…

01/07/2016 admin artificial pancreas, Diabetes, Enlites, Glastonbury Festival, Medtronic 640G, mud, OpenAPS, pump, Smartguard 0

As many will know by now, this year’s Glastonbury Festival was declared “the muddiest ever” by Michael Eavis. It had rain, sun, and mud. Lots of it. Rather more a case of “Mud trials” than “Field trials”, you could say.

But both of these too:

And it had me, with my 640G and Smartguard enabled. This was to be a whole new exercise in trialling the semi-artificial pancreas. Miles to be walked in thick, sticky mud with completely variable weather conditions. What would the results be and how well would it work?

Well the answer would turn out to be “Superbly” as it happened!

Before we start, what exactly is Smartguard? Well, Medtronic says this about the feature:

“SmartGuard®, our exclusive technology, mimics some functions of a healthy pancreas to provide advanced protection against hypoglycaemia.4*

  • Based on your sensor glucose values, SmartGuard® can predict when you are approaching low glucose levels 30 minutes in advance and automatically stop insulin delivery**
  • When your glucose levels recover, SmartGuard will automatically resume insulin delivery***.
  • You have the possibility to set multiple low limits throughout the day to give you increased protection when you need it most.”
It almost sounds too good to be true!
Following advice from a number of online sources, I had placed the Enlite sensor on my arm. This was mainly because I am less likely to sleep on it there and because the skin is less likely to articulate significantly, giving the sensor a chance of life extension. It also seems to have proven to be an accurate location to put the sensor.
Charged up and ready to go, we embarked on our mammoth journey around the festival, walking around 35 miles in 3 days. Through this:

These are the type of conditions that are a challenge for anyone. Wellies at the ready, we were walking through deep, sticky mud that really tested stamina, balance and strength. AFter three days of it my quads were sore. Very definitely a testing workout.

Now the other thing about festivals is that food is, at best unpredictable in terms of what you’ll eat and how the timing works, and at worst, downright horrific, so I give myself a little latitude as to what I’ll tolerate in glucose levels. CGM makes managing this a lot easier, and the Medtronic equipment was no different in this respect.
Where I found it made a huge difference was in how I was able to manage levels. With best guesses for carb counts all that were available on the food front, I was able to take largish doses of insulin, with the knowledge that should I start to drop, the Smartguard would kick in. It really suprised me how effective it was.
I know that there has been a lot of discussion about whether Smartguard considers IOB from bolusing as part of its inputs, but based on my experiences walking around Glastonbury, I’m not convinced this is not the case. On more than one occasion it cut out whilst glucose levels were falling relatively slowly and provided a very soft landing, missing a low. To test the idea of this out, I’ll be looking at rate of change of sensor glucose levels versus when the system decided to suspend. Aside from this, it proved to be a very effective tool.

Throughout the time spent marching round the fields, eating frankly rubbish food and drinking far too much cider, I was able to maintain my glucose levels in reasonably good levels. Over the three days of the festival, without altering my basal rate, the stats were what I can only consider to be really good:

  • Time below 3.9mmol/l (lower threshold): 2%
  • Average glucose level: 6.8mmol/l
  • Time above 7,8mmol/l (upper threshold): 29%
  • Standard Deviation: 1.9mmol/l (29%)
The picture, on the other hand, doesn’t look quite so impressive:
The great thing about the stats is that they are relatively normal and in line with everyday numbers that are generated when I’m working in the office, at home, etc. That average blood glucose level is equivalent to an Hba1C of 5.9%, or just about in the normal range.
You can see from the graphs that it all got a bit lumpy from time to time, but that was driven by miscalculation of eating boluses and cider, and while not ideal, I was aware of them. Overall, it wasn’t bad at all. The combination of CGM and Auto-Suspend really makes managing a festival so much easier, and I can only imagine what the 670G will do, with its reassessment every five minutes, in a mechanism similar to that which OpenAPS currently uses. If the ability to manage BG rises is anything like what I’ve seen from HAPP, it will be spectacular. When you have them in the UK Medtronic, I’d like to use it please!

Coming back to the point about dosing, due to effectively being able to give a superbolus whilst maintaining a basal rate, it was possible to ameliorate the peakiness that often occurs when eating at short notice with no lead time on the insulin. It was surprisingly effective strategy and the Smartguard made it feel a great deal safer than doing it without. There’s definitely something interesting going on in the algorithm.

Ignoring most of the other aspects of pumping at Glastonbury (changing sets, refilling cartridges, etc) I encountered one interesting aspect that was worth bearing in mind when looking at pumps. Most of those on the market currently use an AA battery as the power source. These do run out, and on the Sunday morning, mine ran out. Having forgotten to pack spares I had to drop in to one of the festival shops and buy a pack. While I’d probably have recharged a pump unit before I took the spare with me, it does show that when you have memory failure and forget to take something with you, having the ease of access to batteries that AAs provide is a boon.

So what did I take away from my “Glastonbury Field Trial”?

There were a few conclusions:

  • Smartguard works, for me, really effectively
  • Enlite II Sensors aren’t anywhere near as bad as I had heard them to be – notably I’ve seen fewer compression lows with them versus the Dexcom sensors
  • With Smartguard in place, I was able to be more aggressive in my bolusing which made managing eating significantly easier
  • I didn’t have to worry about changing basal rates in an environment where I was being more active than usual. Smartguard managed this for me
  • Carrying no additional gear around with me for CGM was a massive benefit. At Download with xDrip, I had a phone and the xDrip box, which I had to keep dry, and more importantly charge. Okay, so there’s no watch app with this, but not needing to charge the other bits is a godsend.
Much of what I’ve written above applies whether you are using Smartguard or the OpenAPS artificial pancreas software. What, for me, it shows, is that AP software, while we continue the (probably interminable) wait for a cure can deliver a good alternative.
If you are hypo unaware, this really is a system worth considering, as it’s the only one available right now. I await the Animas alternative with interest.
Meanwhile, we now need to integrate this into the OpenAPS stack, so that we can run with the full AP capability while we wait for the 670G. Based on what I’ve experienced here, OpenAPS plus Smartguard is a viable alternative approach when you can’t send remote temporary boluses.

Be the first to comment

Leave a Reply

Your email address will not be published.


*