Very interesting bit of info Scott. Having used 5 dexcom sensors so far my average life with 4 have been 22 days. Only one sensor hasn't made it this far, 9 days and it failed completely. This just happened to be the week I was struck down with the "Aussie flu ".
Biofouling???? Maybe or maybe not, but would be interesting if it was to happen again.
I've lost track of the link for it, but I read a paper a while back setting out a list of things which conspire against accurate readings. Lots of them, biochemical, bioelectrical, biomechanical. It's extraordinary these things work at all and understandable when they don't
I suppose the trick for those of us lucky enough to be using them is to figure out how to be savvy enough to know when a reading is "right" and when it might be "wrong". Gets easier to do that the more they're used.
There's a pay access paper so haven't read it but the abstract says it looks at sensor development ideas for reducing biofouling. Copied below, it's obvious they don't just buy the bits from Maplins, there's some serious engineering going on here at a microscopic level!
Nine sensor modifications are discussed herein: hydrogels, phospholipid-based biomimicry, flow-based systems, Nafion, surfactants, naturally derived materials, covalent attachments, diamond-like carbons, and topology.
PS: there's been a libre sensor availability blight recently, so I hedged my bets and bought a dexcom G5 sensor and transmitter just in case I ran short of libre. It's sitting on the sideboard just now, but I'll probably join you dexcommers at some point to give it a trial run (although from some youtube videos, looks like I'll have to learn some yoga and gymnastics to attach it to the back of my arm singlehandedly!).