I went to visit one of the Diabetic nurses at Salisbury Hospital today (yesterday). It was kind of a follow-on from my visit to see the nurse at my local surgery (see here) but this guy was obviously far more specialised. I must write about the meeting more fully in the next few days, since it was quite productive, but for now I wanted to concentrate on the discussion we had about pumps.
“You know how they all work, right?” I thought I did, but allowed him to explain. You just basically set a rate manually on the pump, and it pumps the insulin into your body at that rate. Presumably for a certain length of time. And that’s it!
What a crock! So all my assumptions that it would measure your sugar, then pump in insulin as required are unfounded. My ideal pump would take a reading, then deliver a small amount of insulin, then keep repeating that loop until your sugar lies below a certain value (which would need to be user-configurable). Of course, it would have to be fast-acting insulin – there’s no point delivering a small dose then having to wait a few hours for it to take effect. So, maybe a cycle every half hour or so would be appropriate (fast acting insulin takes around 15 minutes to act. It varies by insulin type, so that’s a general number)?
So basically the pump does one simple action – to replace the manual process of pushing a plunger on a pen with something more automatic.
So why don’t they work the way I perceived? Well, at the outset I’d have to say that I’m still unsure that they don’t. I’m sure I’ve read in marketing blurb that some pumps will offer a kind-of “advanced warning” that you’re about to have a hypo, so how can they do that if they’re not measuring? So I need to clarify this.
But in terms of technical issues, I can think of two. First, if all you’re doing is delivering stuff, then you just need a one-way pump. However if you’re receiving blood as well, then you need a two-way pump. So things become more complicated. But, insurmountable? The second issue that needs to be overcome is the pressure difference. Obviously the blood within our bodies will be flowing at a higher pressure than the insulin in an insulin reservoir, which will just be sitting at atmospheric pressure. I say “obviously” – that’s why we bleed when we cut ourselves. So, given that the task is pumping from the low-pressure reservoir to the high pressure bloodstream, there is an obstacle there. But wait a moment! We already have insulin pumps which deliver from a low-pressure reservoir into our body, so this problem has already been solved!
Of course, I have to temper these thoughts with the high level knowledge that pharmaceutical companies are full of very bright people, and have a very keen eye on making a profit, so why are such pumps not available already? It’s a bit like the Fermi Paradox, but for insulin pumps!