[Previous Months][Date Index][Thread Index][Join - Register][Login]
[Message Prev][Message Next][Thread Prev][Thread Next]

Re: [IP] sensor approved

In a message dated 2/27/99 1:46:01 AM, email @ redacted writes:

<<"Advisers to the Food and Drug Administration recommended approval of
Inc.'s ``continuous glucose monitoring system.'' The FDA is not bound by its
advisers' recommendations, but typically follows them."

I attended the meeting yesterday -- quite a show all in all.  I had not been
to an FDA advisory panel meeting before:  the panelists are non-government
"experts" brought in to advise, and most seemed to have a real need to have an
opinion even on things that they had not really understood.  (sorry, I'm so
critical)  There were several temporary panel members (mostly MDs, also a
Nurse Practitioner pump trainer and only 1 patient) who had diabetes
connections, in addition to the MDs and PhD scientists who made up the
permanent panel.  The big difference I noticed between it and scientific
sessions I used to attend were the hordes of scurrying people with cell phones
and the people in the back of the room with laptops probably sending it out in
real time.

People generally seemed impressed with the quality of Minimed's presentation
-- they seemed to know what they were talking about and presented it very
well.  It was fun meeting Al Mann and Terry Gregg (I am a minimed user).

My understanding is that what the continuous glucose monitoring system (CGMS0)
has been recommended to be approved for (with conditions about such things as
more study in type 2 and African American populations) is to be used by
doctors (lets not let those patients get any info) for some number of 3-day
periods to identify glycemic excursions.  There is a test every 5 minutes,
making 288 in a 24 hour period, and an alarm that goes off for a low.  You add
meter readings from finger sticks, and events like meals, exercise, and
insulin.  This data becomes available when the doctor prints it out and
analyses it 3 or more days later. 

The meter is attached to tubing attached to an silhouette-like insertion
device with an electrode remaining in the cannula after the introducer needle
is taken out. 

 The difference between  glucose meter results and sensor was compared to
snapshots vs. the whole movie.  Examples were shown of periods of lows
followed by rebound highs that were not picked up with even large numbers of
finger sticks, of finger sticks being taken on the upswing or downswing of an
event and not showing the gravity -- well, most of you can probably relate.
So while it can not currently be the decider of your treatment decisions, it
can point to times to look further with blood sugar measurements.

But it is measuring interstitial fluid, not blood.  There is a lag between the
time of change of glucose concentration in blood vs. interstitial fluid of
somewhere from 5 to 12 minutes.  If you are only looking at trends, then the
concentration per se doesn't matter as long as it has a quantifiable
relationship to blood glucose measurement -- minimed is assuming it does, and
I think that was generally accepted by the panel.

The current method of callibrating the glucose measurement uses 4 points of
finger-stick readings in 24 hours, a 1-point callibration method not having
worked when it was taken at a time of increase or decrease in slope of trend.
Needing several points seems to me to interfere with the minimed plan to have
a sensor by 2000 that will read glucose concentrations in real time.

So, for the real world here....

1) I was told that patients in the trials had been able to get enough data
about ups and downs out of it (without the doctor's printouts) to have made a
significant improvement in their control.

2) The printout at the end of 3 days should in many cases show "glucose
excursions" (see, we don't have to pay to travel) that were previously
unknown, which will give the patient (or doctor, or CDE) landmarks for changes
in treatment or times to watch out for and test.  This could be particularly
useful to those who have hidden reactions followed by rebounds that are
treated by more insulin rather than the lesser amounts really needed.  Those
with high postprandial levels that are not caught in normal testing could be
alerted to treatment changes like timing of injection or bolus, changes in
foods eaten....

3) the thing has an alarm for lows!!!!  but since it is not likely to be used
except very intermittently (and for most people probably just once), it's
probably not the great joy it would be if I could wear it every night and not
have to set that 3 a.m. alarm.

4) fingersticks on the order of 4 a day will still be necessary.

Just thought some might be interested.

Linda Zottoli
Insulin Pumpers website http://www.insulin-pumpers.org/
for mail subscription assistance, contact: HELP@insulin-pumpers.org