I have not used Dexcom but have used Libre since they were first introduced into Europe (before USA!). The basic technology is the same. You can read it in the hundreds of pages of the patents. The major difference is in so called calibration and smoothing. Both systems rely upon a foreign body being inserted below the skin level to measure glucose within the interstitial fluid that surrounds all our cells, including the brain cells. IMHO That is the bit that really matters.
With both systems the body reacts to the foreign body by producing large cells, macrophage, that coat the foreign body and attempt to protect us from the invader. Those large cells use up glucose! But everyone is different. We all react differently to a simple cut or abrasion. Manufacturers try to estimate a particular persons reactions, but it is little more than an intelligent guess particularly soon after first insertion.
They also try to smooth wild variations
Spot checks of individual readings are not well smoothed which is why trends are so much more use, particularly during early days.
So how does this affect Libre v Dexcom?
Libre is preprogrammed with expected curves while Dexcom allows adjustment to fingerprick levels.
Problem with the latter is that it is just an individual level at one moment in time.
It can be wrong. Even the strip manufacturer says only within a few percent for 95% of the tests, at best!
Doctors are warned that even large sample tests in the lab using very experienced testers and expensive equipment have substantial variation.
Airplane pilots, responsible for perhaps hundreds of lives, are not allowed by their licence to use fingerpricks to test their glucose. The must use CGM if taking insulin.
That is mainly because all CGM rely upon many many single readings to provide a much more accurate average and trend.
So can you rely upon a single 25c strip and single drop of blood?
That is why I use Libre rather than Dexcom. But Libre still needs intelligent interpretation.
I will try to show graphs of spot checks against the average trend, both at the start of sensor and at the end. The variation at the beginning when the body is still reaction to insertion is enormous. Much more than a few ml/mol. Taking spot checks at this time is both dangerous for precise treatment and calibration. We have to intrepret readings. We have to understand variation.
Attached is A sensor just installed - the yellow line is continuous spot readings - the red line is the smoothed averages of the trends- and the same sensor after 14 days- readings much more reliable with little variation- but even this one shows a false spot low warning at around 9:30.
