A ramble as clumsy as the pump’s name itself

I told myself that I would give it some time before posting a recap of my trial on a new pump (you know, the awkwardly titled Medtronic Minimed 530G with Enlite).  Now, nearly two months in, seems as good a time as ever.  But just when I think I’ve formed an opinion on the Minimed 530G, I find a reason to change it.

Case-in-point.  It started out wonderfully. The sensor tracked my blood sugar really, really well. I even put this nifty mash-up of Carelink Pro data (provided to me by my trainer, as I apparently lack the credentials or qualifications to do it myself).


Enlite-CareLinkProThe line is the sensor readings. The dots are the fingersticks.  They are pretty damn close, if I do say so myself.

The red box at the bottom-left is the Threshold Suspend.  It was an annoyance.  I knew I was about to go low, and had already treated and suspended my basal (set a temporary basal of zero for an hour). Then it was just a waiting game… waiting for the CGM to tick down to 65 mg/dl so the pump could alarm and I could silence it.  This happened a few times, actually, and in every case I had already taken some precautionary action (the “Low Predicted” and “Low” alarms that precede it tend to help) and in every case the TS turned out to be a pain in the butt.

I suppose it will be an annoyance every time, except for the time it saves my life. If that’s the case, so be it.  In the meantime, I see it as a way to persuade me to treat an impending low immediately, rather than to push through it and treat it when convenient.  (Or to treat it with fast-acting carbs like glucose tabs, not delicious-but-slower-acting carbs like cake or ice-cream).

Anyway, the first to sensors worked well.  Really well.  I was impressed.

The next few sensors, however, were disappointing. One got pulled early, and I waited out the next despite its inaccuracies, because I had only a limited supply.

It seems that the system suffers from the same problem as its predecessor, where it doesn’t always see the highest highs or the lowest lows.  Or, more accurately, the first high tends to de-sensitize it to future highs. My tendency to troubleshoot leads me to two conclusions, which may or may not be correct.  Either (1) the high BG consumes the sensing gel faster than an in-range BG, thus leaving less of it to measure future highs.  Or (2) all my calibrations are done in a “good” range — and while the algorithm does a fine job at matching an ISIG (the electrical signal received from the sensor)  from a BG of 100 mg/dl with an earlier calibration done at 100, it has a tough time extrapolating an ISIG from a 300 BG from a calibration at 100 to give an accurate number.

Something I’ve learned over time is the way calibrations seem to work.  The relevance of a calibration seems to decrease over time and not with subsequent calibrations.  A new calibration (or series of new calibrations), can’t negate an old one.  In fact, a bunch of rapid calibrations tends to muck things up … really bad.  It’s better, in my experience, to wait 4-6 hours after a calibration before performing a new one.

Yes, like a shot of NPH, you need to wait it out.  A mega-bolus won’t immediately fix your blood sugar and a “correction” calibration won’t immediately fix your CGM. The only thing to get that bad algorithm out of the system is time *.

*Or a sensor restart.  Which is  against the rules of the trial, but in the first 12 hours while the sensor is still trying to settle-in and find its comfortable equilibrium point, I think it’s justified.

But this blog post isn’t about lessons learned, it’s about impressions of the system. And about a week and a half ago, I was ready to call it quits with the system. Or at least try to find a way to weasel myself into a Dexcom trial (I’ve never used a Dexcom, but I’ve heard great things).

But the sensor I’m wearing right now, which I put in on Sunday, has been spot-on.  Even as I type this, my CGM was showing an upward trend with an upward arrow (thanks to overtreating a low) next to the number 184.  My meter reading at the time: 184.


So I really don’t know what to think.  This last sensor was put in a little higher on my abdomen than the previous couple. Though my last few were a few inches above the belt, I’ve heard that the “squeeze” from clothing could affect things.  I don’t know if that’s the case here, but my “good-range” ISIGs are in the 30s again. With the bad sensors, they were in the teens. Maybe location is the key.

That’s one drawback with the six-day wear. There’s less ability to experiment with sites and insertion methods and such, because I’m committed to each one for twice as long.  And since this last one has been working so well, I’m questioning my techniques on the previous few, not the merits of the sensor itself.

Still, I believe that the perfect CGM shouldn’t require such a precise “technique”.  In fact, the mere use of the term “taping technique” implies that it isn’t a simple task like everything else we’ve ever taped  in our entire lives.  The process, admittedly a dramatic improvement over Sofsensor, is still unnecessarily complex. You’ve got to lay the tape so it’s half-on, half-off the back of the sensor, with the front of the sensor sticking through the hole, and firmly secured at the back but more loosely around the transmitter area.

Part of the complexity, in my opinion, is due to the attempt to be “backwards compatible” with the previous transmitter. If they had started from scratch and redesigned the mechanics as well as the chemistry, it would have been better. The other part is that the inserter is large and clunky, and doesn’t fit comfortable in my hand (which isn’t all that small!). I feel the designers focused too much on the sensor wire itself and not enough into the part that rest above the skin. You could drop a Porsche engine into grandpa’s old Buick — it’ll still work, but not to the best of its abilities, and it’ll still look old and clunky.

Also, someone, somewhere, got the false impression that they need to completely hide the needle, which means designing various plastic casings to surround it. This, in my mind, was a mistake — or at a minimum, an over-reaction. It’s not needles in general that we fear (Type 1’s grow to accept them. There’s really no choice). Most needles are small and relatively painless.

The problem was the Sofsensor needle, specifically, that made us cringe, tense up and loathe CGM insertions. Fixing the needle alone would have solved the problem alone, and would have resulted in a simpler, lightweight solution. But the attempts to hide the needle, with the awkward “double-click” of the inserter tool, made it complex and error-prone. I’ve made a few of those errors myself. In time, hopefully those errors will cease.

But, as I said, the sensor I’ve got right now is darn near perfect, so I’m reserving judgment on the Enlite.

* * *

Oddly (for me), this experience is the first time I have been a little bit emotionally fragile over diabetes. And it came from a machine.

I resent the implication (perhaps inflicted on myself, by myself) that the Enlite doesn’t work well because I chose the wrong spot on my body, or because I inserted it wrong, or because I didn’t do a good job taping it, or because I didn’t calibrate it at the proper times.  Over 33 years of diabetes, I’ve been fortunate to have escaped the feelings of self-blame that many others struggle with on a regular basis. When it comes to using the Enlite sensor, however, I have been resigned to, at times, it all being my fault.

And what I resent about that is doesn’t have to be that way. The biggest competitor in the CGM market doesn’t have all these restrictions or complexities. My CDE has seen patients walk in with a Dexcom properly inserted right after they opened the box, without any training. That can’t happen with Enlite.

So while I believe Medtronic does a great … no, a phenomenal job at customer support and training and public-relations, these folks are handcuffed by a product that has some deficiencies. I can only imagine what it’s like for these folks who need to defend and promote it.  Because there’s no doubt that they know.

Hopefully, my feedback from this trial will lead to improvements the next time around.  There are huge opportunities for Medtronic to take a triumphant lead against every other competitor with their next version of the product. I just hope they invest in the system. Not only have they made great strides in their outreach, but even the details of their packaging have improved. Marketing (other than the awkward product names) is their strong-suit. But stellar marketing can only go so far – you need a stellar product.  Their product is very good, but not stellar.

* * *

For some reason, it's missing that radiant glow shown in the ads.

And as a final thought to it all, I just received my second 530G pump today. That Motor Error annoyance that plagued the Revel? Still there.

(Yes, there’s some thoughts of self-blame with that, too.  What is it that I do to keep causing these things?? It seems to be just me.).

But the HelpLine rep asked me if I was looking at my sensor graph when the motor errors occurred, and he recommended that I not do so. Apparently, doing both at the same time can cause motor errors.

Sounds like a load of bull to me, and it tells me that nobody really has a clue as to what’s causing them. As I mentioned, though, the reps can’t fix the products, they can only talk about them. And that, sometimes, puts them in a really tough situation.

* * *

Bottom line: if you’re getting the impression that I dislike this pump, that is incorrect. I do like it…. this post is not a review about PROs versus CONs, and recollections and impressions tend to focus on the negative. Based on the options that are out there, it is still leaning as my pump of choice, but the race is neck-in-neck. They need to really WOW me with the next version, because the others are really close to pulling ahead.

* * *

Disclosure: I was given the opportunity to participate in a trial of the Minimed 530G with Enlite for 90 days. The cost of the pump, meter, strips, and supplies were covered by Medtronic. I am encouraged to freely express my own personal thoughts on the system, without any external editorial control or influence. Read more here.

Posted on May 23, 2014, in Diabetes, Insulin pump. Bookmark the permalink. 5 Comments.

  1. Great post and review man. Awesome. I’m currently on the trial and have many of the same feelings as you have stated above. The biggest area I disagree on is the inserted for the sensor. For me, that is the best part of the whole thing so much easier and I personally don’t even feel it go in. But YDMV.

    It’s funny how one day I also love the sensor data then the next I hate it also. My trainer specifically told me to not over calibrate. That if it’s off a bit, just give it time to catch up and it will, which has been the case most of the time. It’s just annoying that the catch up time has to exist. But at the same time there is no “perfect device” out there yet.

    I loved your point about the TS alarm being annoying until the time you actually need it. I’ve found myself turning the TS alarm off when i see that the low has been predicted and then I treat but the Bg is still low on the cgm.

    Great review.


    • Actually, Chris, I do think the new sensor is a dramatic improvement over the old one, but it’s cumbersome. I think the Enlite resembles the Mio quite closely in form, and wish the insertion was similar as well. I will give lots of credit to packaging of the Enlite though; the Sofsensor plastic bags were so flimsy that I feared the sensor would be damaged before I opened it. (I do wish Enlite’s rigid plastic packaging were recyclable, though). Thanks for the feedback.


  2. Great review Scott. Lots for me to think over here. Also: I think calibrating too often is an issue with all CGMs. I’ve heard it’s hard for things to settle in a range and spot trends if it’s getting various calibrations every hour or so that bump the number around.

    Thanks for sharing your experience… very helpful.


  3. Great recap, Scott. reading this post answered the question I asked on my last comment about the trial…that’s cool you got to participate in this!


  4. Also, the pro data is pretty cool! I wish I had access to that!



Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s