22 July, 2014

How many pieces of technology does it take to replace a pancreas?


In this case,
  • an iphone
  • a tablet
  • two continuous glucose monitors
  • one insulin pump
  • one wireless hotspot
  • a pump meter/remote

A couple weeks ago I got a call about participating in an Artificial Pancreas System trial at UVA, and being that the dates fit perfectly into my summer schedule and I've been wanting to participate in one since I first heard about the system, I of course signed right up.  After the checks to make sure I was alive and not pregnant, I was screened in to the trial.  This meant a week of data collection at home, and then two stays at UVA each lasting about 30 hours.

For 30 hours I got to watch as the devices I listed above worked collectively  as my pancreas.  I watched (somewhat in disbelief) as the system time after time corrected predicted highs and lows to keep me within range for the entire time I was there.  Through meals, sleep, and exercise, I maintained normal blood glucose levels.  While I get some good days here and there when that happens, it is usually because of very careful calculations on my part, or just plain luck.  At one point one of the nurses said "Who would have thought math could do something this great?"  In my head, I responded, "Me Lady!  Of course math can do this - math can do anything you ask it to do!" But I understood her point - after five years of treating my body like a science experiment, I was hooked up to a system that was making all the decisions for me - and making all the right ones.

While at UVA, I had zero high blood glucose values, and zero lows.  The following day at home, I experienced one high, and five lows (three of which were overnight).  Getting to see the system in action gave me hope - I can say with certainty that this will become a reality for home use in my lifetime and will make diabetes management so much more effective and less of a mental and emotional burden.  That is also what made it difficult to go back to "just" my cgm and pump.  It is hard knowing there is a system out there that is safer and more effective than the guessing game I do all day everyday, but also knowing I have to wait, and wait, and wait for it to become available.  If you know me, you know waiting isn't really my thing.

In addition to my rejuvenated hope and faith that this will get easier in my lifetime, I also walked away with a few pieces of information that I think will improve my current methods.


  1. My body seems to do well with an 80% bolus upfront for meals, with the other 20% given over a half hour, one hour, or two hours (depending on the fat content).  
  2. It had been way too long since I strictly carb counted any meals, which meant that my estimates were getting further from the truth.  
  3. While I cannot make a decision about my insulin needs every five minutes all day and all night like the Artificial Pancreas can, I can test more often in the two hours after meals to see what adjustments I need to make in response to my mealtime guess.  

Sorry for the blur - but this is what an entire day of in range glucose values looks like :) 

I head back on Thursday for another 30 hours of closed loop diabetes vacation!  It is nice not having to devote any brain power to diabetes management, but I honestly didn't really know what to do with free time where I couldn't run, bike, or kayak.