Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Possible Reduction in Severe Low Glucose Risk?

19 views
Skip to first unread message

Pro-Humanist FREELOVER

unread,
Jul 25, 2013, 1:38:53 PM7/25/13
to

- - -

The following article, highly technical and over-
reliant on rarely used medical terminology, -but-
the following excerpt may, some day, enable
those with Insulinitis (old name: type 1 diabetes)
to reduce our risk of suffering multiple severe
low glucose events.

Myself, by the way, I've noticed an increase in
adverse low glucose events occurring after my
evening meal at work. I plan to lower my Novolog
dosage to 1 unit less than normal to try to reduce
the likelihood of those happening, while also trying
to maintain the effort to have an HbA1c level as
close to normal as practicable.

One sidenote: Exogenous insulin guessing, a core
aspect of the manner in which Insulinitis (old name:
type 1 diabetes) glucose levels are currently lowered,
and have been lowered ever since insulin injections
first lowered glucose levels in humans in 1922, the
advances in perfecting that side of the equation, such
as are occurring in a wide array of research utilizing
various modalities, are critical, even -if- an ability to
automatically counter low glucose is found, via clinical
trials, to be of unknown (as of yet) value:

- - -
July 24, 2013
http://www.medpagetoday.com/Endocrinology/Diabetes/40676
- - -

Excerpt [with inserts, not part of original
article, included in brackets]:

...

Glucagon, beta-agonists, and other agents are known
to raise glucose levels. However, none of the agents
have demonstrated increased glucose-raising activity
in response to falling glucose levels [none of those can
automatically respond to low glucose events in a preven-
tative manner, a capability lost or severely dysfunctional
in those who have Insulinitis due to the nature of that
disease/condition and its treatment with *all* current ver-
sions of glucose-ignorant exogenous insulin (ignorant in
that exogenous insulin is a guess, and there is currently
no way for the insulin to automatically be released on an
as needed in exactly the right amount basis)].

Five other types of drugs have attracted considerable
interest for their potential to reverse glucose counter-
regulatory defects [i.e., to help counter the exogenous
insulin causality of low and potentially severe low
glucose which transpires in those with Insulinitis]:

selective serotonin-reuptake inhibitors (SSRIs),
adrenergic antagonists,
an opiate-receptor antagonist,
fructose, and
a selective ATP-sensitive potassium-channel agonist.

All five agents have been shown to enhance counter-
regulatory responses when glucose concentrations fall ...

o SSRIs have increased counter-regulatory responses
to hypoglycemia in animal models and in humans.
The evidence is sufficiently compelling to warrant
a clinical trial.

o Combined alpha- and beta-adrenergic blockade pre-
vents attenuation of sympathoadrenal responses in
humans the day after a hypoglycemic episode.

o Naloxone, an opiate-receptor antagonist, has demon-
strated the ability to increase the plasma epinephrine
response, or prevent its attenuation, during hypogly-
cemia.

o Fructose infusion increases epinephrine and glucagon
responses and increases glucose production during
hypoglycemia in humans ...

o Finally, an investigational potassium-channel agonist
has been shown to increase epinephrine response to
hypoglycemia in rats.

...

- - - end excerpts - - -

- - - - - - - - - - - - - - - - - - - - -

- - -
Pro-Humanist FREELOVER
C.ure I.nsulinitis A.ssociation
http://prohuman.net/cureinsulinitisassociation.htm
- - -

--
Posted by Mimo Usenet Browser v0.2.5
http://www.mimousenet.com/mimo/post


Pro-Humanist FREELOVER

unread,
Jul 27, 2013, 9:37:52 AM7/27/13
to

- - -

Follow-up:

On 25 Jul 2013 12:38 PM ,"Pro-Humanist FREELOVER"
<lovefo...@hushmail.com> wrote:

> - - -
>
> The following article, highly technical and over-
> reliant on rarely used medical terminology, -but-
> the following excerpt may, some day, enable
> those with Insulinitis (old name: type 1 diabetes)
> to reduce our risk of suffering multiple severe
> low glucose events.
>
> Myself, by the way, I've noticed an increase in
> adverse low glucose events occurring after my
> evening meal at work. I plan to lower my Novolog
> dosage to 1 unit less than normal to try to reduce
> the likelihood of those happening, while also trying
> to maintain the effort to have an HbA1c level as
> close to normal as practicable.

That happened 3 straight work days, July
21-22 (I was off Tuesday & Thursday) &
24, but I was able to avoid that at work
yesterday. On those 3 events, helped by
co-workers (& glucose tabs & other carbs)
in recovering from those.

Unfortunately, last night, on the floor with
a severe low glucose event while sleeping.

Awoke with a glucose level of 43, so appar-
ently, whatever remains of my ability to on
my own, without any glucose intervention,
to automatically recover (this time, unknown
how long I was 'out') it does kick on, although
overnight, it kicked in after a severe low glu-
cose event sent me to the floor beside my
bed.

The "adverse low glucose" events I refer-
enced in my previous post above, I did
succeed at avoiding that by lowering my
Novolog at dinner by 1 unit at work, but
even though I ate a small carb snack due
to being too low at bedtime, it apparently
wasn't enough to avoid an overnight
severe low.

The "adverse low glucose" events above,
I typically don't include those in my report
of my severe lows, -but- since they were
followed by an indisputable severe low, I'm
including those in my monthly tally of my
severe low history below (July, a record
number of severe lows, what with those
-3- recent adverse lows included in the
following statistical data):

July, 2013 -- 6 (thus far, with
inclusion of 3
adverse low events)
June, 2013 -- 3
May, 2013 -- 1
April, 2013 -- 5
March, 2013 -- 4
February, 2013 -- 1
January, 2013 -- 1

December, 2012 -- 2
November, 2012 -- 1
October, 2012 -- 0
September, 2012 -- 1
August, 2012 -- 2
July, 2012 -- 1
June, 2012 -- 0
May, 2012 -- 2
April, 2012 -- 1
March, 2012 -- 2
February, 2012 -- 0
January, 2012 -- 0

December, 2011 -- 0
November, 2011 -- 1

- - - - - - - - - - - - - - - - - - - - -

(Repeated closing I used last time I
reported on a severe low glucose event,
on July 11th and July 15th)

As always, the high cost of shooting for
a 'normal' glucose level close to or about
at what those without any glucose anomaly
have, the price is extremely high. My post
earlier on July 11th, on ViaCyte, I have
the highest hope that their device will
turn out to be the solution, the holy of
holies, THE CURE, the end to all hypogly-
cemic events, the end of Insulinitis.

Meanwhile, of course, other improvements
in treatment, like a glucose-responsive
insulin, or other advancements in restoring
beta cells or islet cell transplants or in an
artificial pancreas or in noninvasive contin-
uous glucose monitoring, all of that and
more may remedy the severe low glucose
risk that everyone with Insulinitis currently
battles with every insulin decision, every
moment of every day and night.

Pro-Humanist FREELOVER

unread,
Jul 28, 2013, 9:07:54 AM7/28/13
to

- - -

Follow-up 2:
Last night, a recurrence of an overnight
on the floor event. Don't know if the meds
I'm taking for a combination of problems,
eye and dental, are impacting my glucose
levels, but -if- they are, they're lowering
them.

Also, yesterday at work, despite the lower-
ing of Novolog I previously mentioned, I
had another co-worker-assisted response
to an adverse low glucose event, after my
evening meal, despite eating a 16 carb
snack 30 minutes after my meal, and in
addition to the 1 unit lowering of Novolog
before the evening meal, I also lowered my
Levemir by 1 unit that day.

Today at work, I might lower my Novolog
by an additional unit. Also, I'm lowering
my Levemir by another unit this morning.

July, once again, an extension of the record
for severe lows, -if- you include the -4- re-
cent adverse low events in the following
record of severe lows since November of
2011:

July, 2013 -- 8 (thus far, with
inclusion of 4
adverse low events)
June, 2013 -- 3
May, 2013 -- 1
April, 2013 -- 5
March, 2013 -- 4
February, 2013 -- 1
January, 2013 -- 1

December, 2012 -- 2
November, 2012 -- 1
October, 2012 -- 0
September, 2012 -- 1
August, 2012 -- 2
July, 2012 -- 1
June, 2012 -- 0
May, 2012 -- 2
April, 2012 -- 1
March, 2012 -- 2
February, 2012 -- 0
January, 2012 -- 0

December, 2011 -- 0
November, 2011 -- 1

- - - - - - - - - - - - - - - - - - - - -

(Repeated closing I used last time I
reported on a severe low glucose event,
on July 11th, 15th, and 27th)

Glucoii

unread,
Jul 29, 2013, 8:10:02 AM7/29/13
to

It would have been a much more useful post if success at controlling one's
type 1 diabetes roller coaster effect had been reported.

The law of small and large numbers is a respecter of none.

Pro-Humanist FREELOVER

unread,
Aug 2, 2013, 9:35:40 PM8/2/13
to

- - -

Follow-up 3:

On 28 Jul 2013 08:07 AM ,
> [skipped as I updated the list today]

Well, Tuesday, July 30th, day off, my insulin
guess was too high, and as I was working
on the internet, on my computer at home,
I passed out somewhere around 10:00 AM.

When I awoke, I thought it was 10:00 AM,
but later found out it was 10:00 PM, and I
had been passed out for about 12 hours. As
far as I'm aware, that's the longest amount
of time I've been passed out.

Unfortunately, during my passed out episode,
I knocked my computer monitor to the floor
-and- pulled my computer from the table top
to the floor, so that's why I've been out of
touch for the past few days.

I took my computer and monitor to my
younger brother, and he fixed the monitor,
but the 2 hard drives I had, they were obliter-
ated by the fall to the floor. I had a backup
drive, and that's what I'm using now.

It appears that Amoxicillin, which I last
took last night, had a dramatic impact to
cause severe lows and adverse low events,
that totalled -9- in July (5 severe lows, 4
adverse lows), and after taking my last
Amoxicillin last night, I went to the floor
last night.

Obviously, I'm in dire straits, and I've made
an appointment with endocrinologist, Dr.
Osvaldo Alejandro Brusco. Unfortunately,
that appointment is for September 4th,
-but- I plan to present documentation
regarding my severe low history, -and-
perhaps that might persuade him to
move me up to an earlier appointment.

I hope, and fully expect, that Dr. Brusco
will support my effort to get a Continuous
Glucose Monitor, and -if- I can afford it,
I expect to be using that in the very
near future.

August, 2013 -- 1 (thus far)
July, 2013 -- 9, with inclusion of 4
adverse low events

Alan Mackenzie

unread,
Aug 3, 2013, 5:11:35 AM8/3/13
to
Pro-Humanist FREELOVER <lovefo...@hushmail.com> wrote:

> Obviously, I'm in dire straits, and I've made
> an appointment with endocrinologist, Dr.
> Osvaldo Alejandro Brusco. Unfortunately,
> that appointment is for September 4th,
> -but- I plan to present documentation
> regarding my severe low history, -and-
> perhaps that might persuade him to
> move me up to an earlier appointment.

Never heard of the said Dr. Brusco, but I'm sure he'll be good. When
you get to his consulting room, listen to what he has to say. LISTEN!
Really listen. Be explicit with your currently held assumptions, and be
prepared to have these challenged.

> I hope, and fully expect, that Dr. Brusco
> will support my effort to get a Continuous
> Glucose Monitor, and -if- I can afford it,
> I expect to be using that in the very
> near future.

From what you've related over the years, I suspect that a continuous
monitor is not the solution. I hope the good doctor will recommend an
entirely different insulin/testing/food regime from your current one,
which clearly doesn't work - having several hypos a month is not normal
for T1 diabetes. Please have the open mindedness to follow any new
recommendations you get.

I wish you all the best at your upcoming appointment.

--
Alan Mackenzie (Nuremberg, Germany).

Pro-Humanist FREELOVER

unread,
Aug 3, 2013, 10:50:14 AM8/3/13
to

On 03 Aug 2013 04:11 AM ,Alan Mackenzie <a...@muc.de> wrote:

> Pro-Humanist FREELOVER <lovefo...@hushmail.com> wrote:

> >
> > [...]
> >
> > Obviously, I'm in dire straits, and I've made
> > an appointment with endocrinologist, Dr.
> > Osvaldo Alejandro Brusco. Unfortunately,
> > that appointment is for September 4th,
> > -but- I plan to present documentation
> > regarding my severe low history, -and-
> > perhaps that might persuade him to
> > move me up to an earlier appointment.

> Never heard of the said Dr. Brusco, but I'm sure he'll be good. When
> you get to his consulting room, listen to what he has to say. LISTEN!
> Really listen. Be explicit with your currently held assumptions, and be
> prepared to have these challenged.

Well, my currently held assumptions are
that HbA1c levels, the closer they are to
'normal', the better, if by better one is
referring to reduced risk of long-term
adverse complications, and that without
a continuous glucose monitor, the reality
is that achieving that near-normal HbA1c
has a dramatically increased risk of severe
lows.

-If- a continuous glucose monitor does,
indeed, reduce the risk of severe lows,
dramatically, the only reason to not pre-
scribe one would be -if- it's so expen-
sive that I can't afford it.

> > I hope, and fully expect, that Dr. Brusco
> > will support my effort to get a Continuous
> > Glucose Monitor, and -if- I can afford it,
> > I expect to be using that in the very
> > near future.

> From what you've related over the years, I suspect that a continuous
> monitor is not the solution.

I would think that a CGM would not be
the solution -if- it's too expensive, and
-if- so, well then, I'm screwed unless
what you mention below has viability.

> I hope the good doctor will recommend an
> entirely different insulin/testing/food regime
> from your current one, which clearly doesn't
> work - having several hypos a month is not
> normal for T1 diabetes.

It's possible that a higher HbA1c level
would be suggested, but I can't help
but worry -if- that's the case, I'd be
at higher risk of _________ (insert
long-term complications here).

> Please have the open mindedness to follow
> any new recommendations you get.
>
> I wish you all the best at your upcoming
> appointment.

Thanks. We'll see. Perhaps, there's
something I haven't tried that provides
low HbA1c levels without increased
risk of severe lows and without the
need for use of a CGM, but since I've
failed to figure that out as of now, I
can't help but wonder if my "screwed"
presumption is my fate until some-
thing CGM-like (i.e., some non-inva-
sive glucose monitoring that would
resumble what current CGMs pro-
vide, but without the cost) comes on
the scene.

>
> --
> Alan Mackenzie (Nuremberg, Germany).

Presumption

unread,
Aug 3, 2013, 11:32:10 AM8/3/13
to

"Thanks. We'll see. Perhaps, there's
something I haven't tried that provides
low HbA1c levels without increased
risk of severe lows and without the
need for use of a CGM, but since I've
failed to figure that out as of now, I
can't help but wonder if my "screwed""

A cgm will do no good. It will only confirm what you know already.

You have been provided info by which to take control of your roller coaster
effect common in type 1 diabetes. Knowing minuet to minute your glucose
will be of no benefit unless and until you stop whining and take control of
your type 1 diabetes.

Pro-Humanist FREELOVER

unread,
Aug 3, 2013, 1:36:54 PM8/3/13
to

In a previous post, the following segment
of what I wrote was included in a reply
from Presumption:

> [...]
>
> Thanks. We'll see. Perhaps, there's
> something I haven't tried that provides
> low HbA1c levels without increased
> risk of severe lows and without the
> need for use of a CGM, but since I've
> failed to figure that out as of now, I
> can't help but wonder if my "screwed"
>
> [...]

Presumption replied:

> A cgm will do no good. [...]

Many studies of the use of CGMs have
conveyed reduced risk of severe hypos
is associated with the use of a CGM
and is likewise associated with a
reduction in severe lows along with
HbA1c levels which are well within
the acceptable levels (which vary)
suggested by most medical entities
which specialize in such matters.

Unfortunately, I don't think I can
afford such a device, so it may be
a mute point, the use of a CGM,
-if- I can't afford it (i.e., if the
amount provided by insurance
leaves so much out-of-pocket
costs that my low income makes
it an unaffordable option).

Opp

unread,
Aug 3, 2013, 2:48:32 PM8/3/13
to
In a previous post, the following segment
of what I wrote was included in a reply
from Presumption:

> [...]
>
> Thanks. We'll see. Perhaps, there's
> something I haven't tried that provides
> low HbA1c levels without increased
> risk of severe lows and without the
> need for use of a CGM, but since I've
> failed to figure that out as of now, I
> can't help but wonder if my "screwed"
>
> [...]

Presumption replied:

> A cgm will do no good. [...]

" Many studies of the use of CGMs have
conveyed reduced risk of severe hypos
is associated with the use of a CGM
and is likewise associated with a
reduction in severe lows along with
HbA1c levels which are well within
the acceptable levels (which vary)
suggested by most medical entities
which specialize in such matters."

Sure, but adding datum points to the roller coaster curve does not a whit
of advantage when one is unwilling to do something about it.

What you need to do can be tracked by the usual meter use.

You are always looking for the silver bullet solution to your type 1
diabetic problem. This in place of taking control of it you whine and want
to continue lifestyle choices which no longer work for you.

It is common as dirt among type 1 diabetics and you are no exception to it.

This problem existed long before you encountered it and solutions were
derived. Those solutions are at hand, use them.

Pro-Humanist FREELOVER

unread,
Aug 4, 2013, 12:50:19 PM8/4/13
to

On 03 Aug 2013 01:48 PM ,Opp wrote:

> In a previous post, the following segment
> of what I wrote was included in a reply
> from Presumption:

> > [...]
> >
> > Thanks. We'll see. Perhaps, there's
> > something I haven't tried that provides
> > low HbA1c levels without increased
> > risk of severe lows and without the
> > need for use of a CGM, but since I've
> > failed to figure that out as of now, I
> > can't help but wonder if my "screwed"
> >
> > [...]

> Presumption replied:

> > A cgm will do no good. [...]

I replied:

> Many studies of the use of CGMs have
> conveyed reduced risk of severe hypos
> is associated with the use of a CGM
> and is likewise associated with a
> reduction in severe lows along with
> HbA1c levels which are well within
> the acceptable levels (which vary)
> suggested by most medical entities
> which specialize in such matters."

Opp replied:

> Sure, but adding datum points to the roller coaster curve does not a whit
> of advantage when one is unwilling to do something about it.

My glucose levels are not accurately
characterized in that way, and my
HbA1c, well, I'll get back to you on
that when I get the lab results from
my adventure with the endocrinol-
ogist I'm consulting with late August
and early in September.

> What you need to do can be tracked by the usual meter use.

No, bleeding 10 or more times per
day doesn't come close to the 12
times per hour, 24 hours per day,
revelations available via a CGM,
nor does it come clost to the way
in which our bodies handled our
glucose levels (constant monitor,
perfect responses, metabolic
perfection, glucose levels con-
stantly between 70 and 120-130
every second of every day/night,
no matter what) before we came
down with Insulinitis.

> You are always looking for the silver bullet solution to your type 1
> diabetic problem. This in place of taking control of it you whine and want
> to continue lifestyle choices which no longer work for you.
>
> It is common as dirt among type 1 diabetics and you are no exception to it.
>
> This problem existed long before you encountered it and solutions were
> derived. Those solutions are at hand, use them.

- - -
Continuous Glucose Monitoring
http://diabetes.niddk.nih.gov/dm/pubs/glucosemonitor/
- - -

Alan Mackenzie

unread,
Aug 7, 2013, 5:28:21 PM8/7/13
to
Pro-Humanist FREELOVER <lovefo...@hushmail.com> wrote:

> On 03 Aug 2013 04:11 AM ,Alan Mackenzie <a...@muc.de> wrote:

>> Never heard of the said Dr. Brusco, but I'm sure he'll be good. When
>> you get to his consulting room, listen to what he has to say. LISTEN!
>> Really listen. Be explicit with your currently held assumptions, and be
>> prepared to have these challenged.

> Well, my currently held assumptions are
> that HbA1c levels, the closer they are to
> 'normal', the better, if by better one is
> referring to reduced risk of long-term
> adverse complications, ....

I've never seen anybody convincingly argue that HbA1c levels have any
direct influence on long term health. Rather it's blood glucose levels
which are important.

> .... and that without
You need fewer hypos. That means getting rid of low extremes. You can
do that either by increasing the average, or decreasing the variability.
The second of these will have negative consequences on your life style,
even assuming it's possible.

>> Please have the open mindedness to follow
>> any new recommendations you get.
>>
>> I wish you all the best at your upcoming
>> appointment.

> Thanks. We'll see. Perhaps, there's
> something I haven't tried that provides
> low HbA1c levels without increased
> risk of severe lows and without the
> need for use of a CGM, but since I've
> failed to figure that out as of now, I
> can't help but wonder if my "screwed"
> presumption is my fate until some-
> thing CGM-like (i.e., some non-inva-
> sive glucose monitoring that would
> resumble what current CGMs pro-
> vide, but without the cost) comes on
> the scene.

Bear in mind that long term risks are fairly meaningless if the next time
the computer falls in your face and kills you.

> - - -

Pro-Humanist FREELOVER

unread,
Aug 7, 2013, 7:59:02 PM8/7/13
to

On 07 Aug 2013 04:28 PM ,
Alan Mackenzie <a...@muc.de> wrote:

> Pro-Humanist FREELOVER <lovefo...@hushmail.com> wrote:

> > On 03 Aug 2013 04:11 AM ,Alan Mackenzie <a...@muc.de> wrote:

> >> Never heard of the said Dr. Brusco, but I'm sure he'll be good. When
> >> you get to his consulting room, listen to what he has to say. LISTEN!
> >> Really listen. Be explicit with your currently held assumptions, and be
> >> prepared to have these challenged.

> > Well, my currently held assumptions are
> > that HbA1c levels, the closer they are to
> > 'normal', the better, if by better one is
> > referring to reduced risk of long-term
> > adverse complications, ....

> I've never seen anybody convincingly argue that HbA1c levels have any
> direct influence on long term health. Rather it's blood glucose levels
> which are important.

Well, blood glucose levels and HbA1c
levels go hand-in-hand, with HbA1c
levels simply reflecting the average
blood glucose levels over a 1 to 2
month period.
Well, the computer is on the floor (now),
so it's highly unlikely for that to happen.
The monitor, not that heavy, so it's not
likely to kill me if it falls on me. The
speakers, quite heavy, but not likely
that they'd fall on me during a severe
low glucose event -- the one on the
left, I knocked it to the floor when I
damaged my computer & monitor.

The endo lab tests are only 3 weeks
away, with the initial visit to the endo
a week after that. Funny thing, the
endo is right down the street from
where I'm having surgery done to
correct a couple of eye problems
(including cataracts, which have
blurred the vision in my left eye).

> > - - -
>
> --
> Alan Mackenzie (Nuremberg, Germany).

Alan Mackenzie

unread,
Aug 10, 2013, 11:47:13 AM8/10/13
to
Pro-Humanist FREELOVER <love_fo...@hushmail.com> wrote:

> On 07 Aug 2013 04:28 PM ,
> Alan Mackenzie <a...@muc.de> wrote:

>> Pro-Humanist FREELOVER <lovefo...@hushmail.com> wrote:

>> > Well, my currently held assumptions are
>> > that HbA1c levels, the closer they are to
>> > 'normal', the better, if by better one is
>> > referring to reduced risk of long-term
>> > adverse complications, ....

>> I've never seen anybody convincingly argue that HbA1c levels have any
>> direct influence on long term health. Rather it's blood glucose levels
>> which are important.

> Well, blood glucose levels and HbA1c
> levels go hand-in-hand, ....

Only to a limited extent, namely ....

> .... with HbA1c
> levels simply reflecting the average
> blood glucose levels over a 1 to 2
> month period.

The average over a 1 to 2 month period isn't what is giving you hypos,
and it won't be what might cause long term damage. Why do you put so
much reliance on such a second rate average when you've got the actual
raw blood glucose numbers to work with? I can understand doctors doing
so, since they're likely to be lacking a proper BG log, or may be
skeptical about its accuracy. But an actual diabetic, conscienciously
testing?

I don't think high HbA1c causes any health problems. It's just that it's
correlated with high blood glucose, which might. There's a lot of
confusion surrounding this point.

>> >> I hope the good doctor will recommend an
>> >> entirely different insulin/testing/food regime
>> >> from your current one, which clearly doesn't
>> >> work - having several hypos a month is not
>> >> normal for T1 diabetes.

>> > It's possible that a higher HbA1c level
>> > would be suggested, but I can't help
>> > but worry -if- that's the case, I'd be
>> > at higher risk of _________ (insert
>> > long-term complications here).

You should really stop thinking in terms of HbA1c, and concentrate
instead on blood glucose levels. I think it highly likely a higher BG
level will get suggested.

>> Bear in mind that long term risks are fairly meaningless if the next time
>> the computer falls in your face and kills you.

> Well, the computer is on the floor (now),
> so it's highly unlikely for that to happen.
> The monitor, not that heavy, so it's not
> likely to kill me if it falls on me. The
> speakers, quite heavy, but not likely
> that they'd fall on me during a severe
> low glucose event -- the one on the
> left, I knocked it to the floor when I
> damaged my computer & monitor.

It need not be computer equipment which kills you. It could be anything.
Next time it might be your cut-throat razor if the hypo occurs whilst
you're shaving. It could be a motor car, not necessarily your own.
Hypos aren't risk free, and you seem to have neglected their risk in
your calculations to date.

Pro-Humanist FREELOVER

unread,
Aug 10, 2013, 8:15:09 PM8/10/13
to
I recognize the risks. However, I can't
help but note that every single time
I've gone unconscious or otherwise
severely low, I've recovered. Soon,
I'll meet with the endo to ascertain
what he thinks about the severe lows,
as well as obtaining a complete assess-
ment of my endocrine and Insulinitis-
impacted state of health apart from
those severe lows.

I'll let y'all know how all that works
out, and will get back on what altera-
tions to my approach that he recom-
mends.

> > - - -
> > Pro-Humanist FREELOVER
>
> --
> Alan Mackenzie (Nuremberg, Germany).

- - -
Pro-Humanist FREELOVER

Alan Mackenzie

unread,
Aug 11, 2013, 3:42:02 PM8/11/13
to
Pro-Humanist FREELOVER <love_fo...@hushmail.com> wrote:

> On 10 Aug 2013 10:47 AM ,Alan Mackenzie <a...@muc.de> wrote:

>> It need not be computer equipment which kills you. It could be anything.
>> Next time it might be your cut-throat razor if the hypo occurs whilst
>> you're shaving. It could be a motor car, not necessarily your own.
>> Hypos aren't risk free, and you seem to have neglected their risk in
>> your calculations to date.

> I recognize the risks. However, I can't
> help but note that every single time
> I've gone unconscious or otherwise
> severely low, I've recovered.

So far, yes. Most of the time, people get out of burning houses or car
crashes without injury too, but that doesn't mean they're good things to
get involved in.

> Soon,
> I'll meet with the endo to ascertain
> what he thinks about the severe lows,
> as well as obtaining a complete assess-
> ment of my endocrine and Insulinitis-
> impacted state of health apart from
> those severe lows.

You know, in all your reporting of your hypos over the last few years,
you haven't once said you wanted them to stop. It is not clear that you
do want them to stop. Perhaps you could clarify this point, even if only
to yourself.

Also, you report them in a detached impersonal way, as "events" which
occur, much the way a weather event might, rather than as something over
which you have influence. This suggests to me you don't really want to
take responsibility for your hypos.

I sincerely recommend you to explore these two points as deeply as you
can, well before your upcoming appointment with your new doctor.
Otherwise, if what I suspect is right, that appointment will just be a
waste of time and money for everybody. In that case, you'd be better off
just cancelling the appointment and carrying on the way you'd be going to
anyway.

Herman Rubin

unread,
Aug 12, 2013, 12:04:08 PM8/12/13
to
On 2013-08-11, Alan Mackenzie <a...@muc.de> wrote:
> Pro-Humanist FREELOVER <love_fo...@hushmail.com> wrote:

>> On 10 Aug 2013 10:47 AM ,Alan Mackenzie <a...@muc.de> wrote:

................

> You know, in all your reporting of your hypos over the last few years,
> you haven't once said you wanted them to stop. It is not clear that you
> do want them to stop. Perhaps you could clarify this point, even if only
> to yourself.

I have had too many severe hypos, and have YET to lose any control
due to them. However, I take precautions against that happening
"as far as possible"; I would rather have higher levels of blood
glucose than hypos. By severe, I mean 40-60.

I suggest you take Alan's recommendations seriously.

> Also, you report them in a detached impersonal way, as "events" which
> occur, much the way a weather event might, rather than as something over
> which you have influence. This suggests to me you don't really want to
> take responsibility for your hypos.

> I sincerely recommend you to explore these two points as deeply as you
> can, well before your upcoming appointment with your new doctor.
> Otherwise, if what I suspect is right, that appointment will just be a
> waste of time and money for everybody. In that case, you'd be better off
> just cancelling the appointment and carrying on the way you'd be going to
> anyway.

>> - - -
>> Pro-Humanist FREELOVER



--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
hru...@stat.purdue.edu Phone: (765)494-6054 FAX: (765)494-0558

Pro-Humanist FREELOVER

unread,
Aug 13, 2013, 1:53:38 PM8/13/13
to

On 12 Aug 2013 11:04 AM , Herman Rubin <hru...@median.stat.purdue.edu> wrote:

> On 2013-08-11, Alan Mackenzie <a...@muc.de> wrote:

> > Pro-Humanist FREELOVER <love_fo...@hushmail.com> wrote:

> >> On 10 Aug 2013 10:47 AM ,Alan Mackenzie <a...@muc.de> wrote:

> ................
>
> > You know, in all your reporting of your hypos over the last few years,
> > you haven't once said you wanted them to stop. It is not clear that you
> > do want them to stop. Perhaps you could clarify this point, even if only
> > to yourself.

> I have had too many severe hypos, and have YET to lose any control
> due to them. However, I take precautions against that happening
> "as far as possible"; I would rather have higher levels of blood
> glucose than hypos. By severe, I mean 40-60.

My definition of severe lows is loss of con-
sciousness, or (for me) an overnight on the
floor low glucose-caused event, or EMS inter-
vention whether consciousness is lost or not,
or a seizure, or (and without either of the
former, I define the following as adverse
low) loss of cognizance which ends up with
an intervention by someone, but I haven't
done a good job of keeping track of those,
and usually don't report when one of those
transpires.

As for 40 (or lower) to 60, here's a com-
plete record of my glucose level ranges
over the past month (July 14 to right
now):

Severe Lows (as defined
by yours truly above) -- 5
Under 30 -- 3
30 to 39 -- 4
40 to 49 -- 43
50 to 59 -- 43
60 to 69 -- 49
70 to 79 -- 32
80 to 89 -- 26
90 to 99 -- 23
100 to 109 -- 27
110 to 119 -- 16
120 to 129 -- 6
130 to 139 -- 14
140 to 149 -- 10
150 to 159 -- 6
160 to 169 -- 3
170 to 179 -- 2
180 to 189 -- 1
190 to 199 -- 1
200 to 209 -- 0
210 to 219 -- 4
220 or above -- 1 (a test of 278)

Mathematically, if severe lows are arbitrarily
designated as a glucose level of 15, here's
the average of all those readings: 83.1,
remarkably close to the glucose level which
is promoted by Dr. Richard Bernstein (see
below) as "normal", but lower than the glu-
cose level which many medical professionals
deem to be "normal" (that being close to a
level of 100) for those who don't have any
glucose anomaly.

> I suggest you take Alan's recommendations seriously.

> > Also, you report them in a detached impersonal way, as "events" which
> > occur, much the way a weather event might, rather than as something over
> > which you have influence. This suggests to me you don't really want to
> > take responsibility for your hypos.
> >
> > I sincerely recommend you to explore these two points as deeply as you
> > can, well before your upcoming appointment with your new doctor.
> > Otherwise, if what I suspect is right, that appointment will just be a
> > waste of time and money for everybody. In that case, you'd be better off
> > just cancelling the appointment and carrying on the way you'd be going to
> > anyway.

> >> - - -
> >> Pro-Humanist FREELOVER

> --
> This address is for information only. I do not claim that these views
> are those of the Statistics Department or of Purdue University.
> Herman Rubin, Department of Statistics, Purdue University
> hru...@stat.purdue.edu Phone: (765)494-6054 FAX: (765)494-0558

Insulin, exogenously administered in those
who have Insulinitis, statistics demonstrate
that the amount of risk of severe lows is
increased as the HbA1c levels are decreased.

Exogenous insulin, not an optional substance
for anyone who has Insulinitis, required to
continue to live (until/unless a cure arrives,
and many advancements are being pursued,
as referenced below).

Every day/night, I give 6 insulin injections
(although occasionally I only give 5, and infre-
quently, I have to give more due to the glu-
cose going higher than expected) in an ongo-
ing manual effort to guess right at the amount
of insulin needed to keep me alive .

Obviously, severe hypos are not desired, and
new tech (non-invasive glucose monitoring,
glucose-responsive-insulin, artificial pancreas,
new non-immuno-suppressant implants with
glucose-aware insulin production, -or- restor-
ation of insulin production via some non-im-
muno-suppressant method) seeks to 'norm-
alize' glucose levels with minimal or, ideally,
*no* risk of severe lows.

What is my HbA1c level? Last time it was
checked, well over a year ago, it was less
than 6. I fully expect my next HbA1c level
check to be close to normal (which most
medical enties report is 5, with some who
have no diagnosed glucose anomaly hav-
ing an HbA1c level up to 5.6) but the fol-
lowing doctor defines 'normal' as lower
than that).

Dr. Richard Bernstein, he communicates
that for those who are on a low carb/low
insulin regimen, with discriplined/regular
exercise, have less in the way of low glu-
cose -and- severe low glucose. I worry
that the diet is, for the overwhelming
majority, not all, undoable, and as such,
impractical. I also worry that with highly
restricted carb intake, that insulin dosages
that are 'off' by a mere unit increase the
risk of hypoglycemia, as there are insuffi-
cient carbs to raise glucose levels.

Many (like me) have, from time to time,
requirements to do work necessitating
an increased level of exercise, which
(under a severe low carb diet) makes
it impossible to avoid hypos if the un-
foreseen exercise transpires after an
insulin dosing decision has been made,
unless (of course) one exceeds that
severe low carb restriction.

I also worry that the amount of lowering
of quality of life portends a very unlikely
chance that the diet would be able to be
used, by the overwhelming majority, for
other than a short period of time.

Dr. Richard Bernstein has written that
the golden ideal for glucose levels is
83, but independent studies are gener-
ally closer to normal being at or close
to 100.

I expect that the endo I'm about to see
is along the lines of the overwhelming
majority in the medical profession who
recommend treating of Insulinitis with
a moderate carb well-balanced diet,
not a severe low carb one, but we'll
soon see about that.

I hope he sees the value of a CGM, but
don't know if my out-of-pocket costs
would allow me to afford one.

- - -
Pro-Humanist FREELOVER

Alan Mackenzie

unread,
Aug 15, 2013, 5:11:58 AM8/15/13
to
You've got vast numbers of very low readings there. That really isn't
good. If I had that list of values, I'd be doing something urgently to
raise these low values, since they'd be most likely to lead to frequent
hypoglycaemia.

> Mathematically, if severe lows are arbitrarily
> designated as a glucose level of 15, here's
> the average of all those readings: 83.1, ....

"Average"? You probably mean the arithmetic mean, here, and this is
probably not a sensible average to use. Your values are bunched up in
the 40 to 80 range, and each high value you have is "knocking out"
several low values in the calculation. In effect, you are giving your
high values much more weight than your low values. A more sensible
average would probably be the geometric mean, which would give more equal
weight to the low values. This would give you a lower figure than 83.


> remarkably close to the glucose level which
> is promoted by Dr. Richard Bernstein (see
> below) as "normal", but lower than the glu-
> cose level which many medical professionals
> deem to be "normal" (that being close to a
> level of 100) for those who don't have any
> glucose anomaly.

What exactly does Dr. Bernstein promote as "normal". I suspect it is
more likely to be a typical value of 83, not an average.

> Insulin, exogenously administered in those
> who have Insulinitis, statistics demonstrate
> that the amount of risk of severe lows is
> increased as the HbA1c levels are decreased.

Again, why are you placing so much reliance on HbA1c, a crude second
rate average, when you've got the real unadulterated values there?
Rephrasing that statement, it just says that the risk of severe lows is
increased as average blood glucose levels are reduced. That's not rocket
science.

> Every day/night, I give 6 insulin injections
> (although occasionally I only give 5, and infre-
> quently, I have to give more due to the glu-
> cose going higher than expected) in an ongo-
> ing manual effort to guess right at the amount
> of insulin needed to keep me alive .

I think you originally chose this high intensity regime as an experiment,
trying to maintain tightly controlled BG levels around "normal" levels.
Results to date suggest either this isn't possible, or you're going about
it the wrong way. My BG values vary about as much as yours do (although at
a higher, safer level), and I get by with just 3 injections a day. My
having a hypo is a very rare event indeed.

> Obviously, severe hypos are not desired, .....

Aren't they? By whom are they not desired? In your next post, try
writing "I don't want severe hypos" and see how difficult you find it.

> .... and
> new tech (non-invasive glucose monitoring,
> glucose-responsive-insulin, artificial pancreas,
> new non-immuno-suppressant implants with
> glucose-aware insulin production, -or- restor-
> ation of insulin production via some non-im-
> muno-suppressant method) seeks to 'norm-
> alize' glucose levels with minimal or, ideally,
> *no* risk of severe lows.

Pie in the sky. You'll be decades dead before signicant new technology
starts making any difference.

> What is my HbA1c level? Last time it was
> checked, well over a year ago, it was less
> than 6. I fully expect my next HbA1c level
> check to be close to normal (which most
> medical enties report is 5, with some who
> have no diagnosed glucose anomaly hav-
> ing an HbA1c level up to 5.6) but the fol-
> lowing doctor defines 'normal' as lower
> than that).

Why do you care so much about your HbA1c level? It doesn't do anything
to you, for or against. It's merely correlated with "average" BG levels.
I don't know what my HbA1c is (even if it weren't varying constantly) and
I really don't care.

> Dr. Richard Bernstein, he communicates
> that for those who are on a low carb/low
> insulin regimen, with discriplined/regular
> exercise, have less in the way of low glu-
> cose -and- severe low glucose. I worry
> that the diet is, for the overwhelming
> majority, not all, undoable, ....

That's irrelevant. It's undoable for you. It's fine for lots of others,
apparently.

> .... and as such,
> impractical. I also worry that with highly
> restricted carb intake, that insulin dosages
> that are 'off' by a mere unit increase the
> risk of hypoglycemia, as there are insuffi-
> cient carbs to raise glucose levels.

I think the risk of hypos will be less in low carb/low insulin regimes.
The numbers, the swings, the variations, are just less.

> I hope he [PHF's new endo] sees the value of a CGM, but
> don't know if my out-of-pocket costs
> would allow me to afford one.

Consider what impact those costs would have on your quality of life.

> - - -
> Pro-Humanist FREELOVER

--
Alan Mackenzie (Nuremberg, Germany).

Pro-Humanist FREELOVER

unread,
Aug 15, 2013, 11:53:53 AM8/15/13
to

On 15 Aug 2013 04:11 AM ,
Dr. Bernstein, in his own words, discussing
a glucose level of 83. As for HbA1c, Dr.
Bernstein uses a personal HbA1c to glucose
conversion table that is different from the
charts generally available on the net, and
the disparity between the charts -and- Dr.
Bernstein's personal chart widens as HbA1c
levels rise.

- - -
August 10, 2013

Q&A with Dr. Richard Bernstein
http://diabeteshealth.com/read/2013/08/10/7955/qanda-with-dr--richard-bernstein/
- - -

> > Insulin, exogenously administered in those
> > who have Insulinitis, statistics demonstrate
> > that the amount of risk of severe lows is
> > increased as the HbA1c levels are decreased.

> Again, why are you placing so much reliance on HbA1c, a crude second
> rate average, when you've got the real unadulterated values there?
> Rephrasing that statement, it just says that the risk of severe lows is
> increased as average blood glucose levels are reduced. That's not rocket
> science.

> > Every day/night, I give 6 insulin injections
> > (although occasionally I only give 5, and infre-
> > quently, I have to give more due to the glu-
> > cose going higher than expected) in an ongo-
> > ing manual effort to guess right at the amount
> > of insulin needed to keep me alive .

> I think you originally chose this high intensity regime as an experiment,
> trying to maintain tightly controlled BG levels around "normal" levels.

The high intensity regimen was advocated
by my doctor, after I had gone to him due
to my hypoglycemic unawareness causing
severe lows that had become seriously
problematic. His primary interest was
simply in reducing chances of long-term
complications, and really, the only preven-
tative as for severe lows was/is blood
glucose testing, something I resisted
doing until I was, basically, forced into
doing it due to the exigencies of hypo
unawareness.

> Results to date suggest either this isn't possible, or you're going about
> it the wrong way. My BG values vary about as much as yours do (although at
> a higher, safer level), and I get by with just 3 injections a day. My
> having a hypo is a very rare event indeed.

> > Obviously, severe hypos are not desired, .....

> Aren't they? By whom are they not desired? In your next post, try
> writing "I don't want severe hypos" and see how difficult you find it.

I desire ZERO severe lows. I don't want
them, I don't like them, I recognize they
raise my short-term death risk. I do
realize that my chances of having them
are increased due to my HbA1c (glucose)
goals being so close to normal, but I'm
not willing to raise my HbA1c goals, due
to my fear of long-term complications,
and do think a CGM (continuous glucose
monitor) is the best solution possible
given the current technological limita-
tions (if I can afford it).

> > .... and
> > new tech (non-invasive glucose monitoring,
> > glucose-responsive-insulin, artificial pancreas,
> > new non-immuno-suppressant implants with
> > glucose-aware insulin production, -or- restor-
> > ation of insulin production via some non-im-
> > muno-suppressant method) seeks to 'norm-
> > alize' glucose levels with minimal or, ideally,
> > *no* risk of severe lows.

> Pie in the sky. You'll be decades dead before signicant new technology
> starts making any difference.

Unusually pessimistic point-of-view, and
I'm much more hopeful/optimistic than
that. Until 1922, Insulinitis was a death
sentence. With insulin, it's no longer a
death sentence, but it continues to be
a condition that has a huge cost (quality-
of-life-wise, dollar-wise, risk-wise) that
only technological advancement offers
a viable/potentially doable escape for.

> > What is my HbA1c level? Last time it was
> > checked, well over a year ago, it was less
> > than 6. I fully expect my next HbA1c level
> > check to be close to normal (which most
> > medical enties report is 5, with some who
> > have no diagnosed glucose anomaly hav-
> > ing an HbA1c level up to 5.6) but the fol-
> > lowing doctor defines 'normal' as lower
> > than that).

> Why do you care so much about your HbA1c level? It doesn't do anything
> to you, for or against. It's merely correlated with "average" BG levels.
> I don't know what my HbA1c is (even if it weren't varying constantly) and
> I really don't care.

> > Dr. Richard Bernstein, he communicates
> > that for those who are on a low carb/low
> > insulin regimen, with discriplined/regular
> > exercise, have less in the way of low glu-
> > cose -and- severe low glucose. I worry
> > that the diet is, for the overwhelming
> > majority, not all, undoable, ....

> That's irrelevant. It's undoable for you. It's fine for lots of others,
> apparently.

What percentage of those battling Insulinitis
are on Dr. Bernstein's severe low carb and
low insulin regimen for 5 years or longer?

I suspect it's somewhere along the lines of
less than one percent, but if anyone has
any information on that, please share.

> > .... and as such,
> > impractical. I also worry that with highly
> > restricted carb intake, that insulin dosages
> > that are 'off' by a mere unit increase the
> > risk of hypoglycemia, as there are insuffi-
> > cient carbs to raise glucose levels.

> I think the risk of hypos will be less in low carb/low insulin regimes.
> The numbers, the swings, the variations, are just less.

Very close to starvation, the only way
that people survived before 1922, and
I suspect that it's (Dr. Bernstein's severe
low carb/low insulin) undoable for the
overwhelming majority of those who
have Insulinitis, and as such, prepos-
terous and antithetical to human
nature for almost everyone.

> > I hope he [PHF's new endo] sees the value of a CGM, but
> > don't know if my out-of-pocket costs
> > would allow me to afford one.

> Consider what impact those costs would have on your quality of life.

If I can't afford it, I can't have one.

> > - - -
> > Pro-Humanist FREELOVER

> --
> Alan Mackenzie (Nuremberg, Germany).

- - -
Pro-Humanist FREELOVER

Jim Dumas

unread,
Aug 15, 2013, 1:23:07 PM8/15/13
to
On Thu, 15 Aug 2013 10:53:53 -0500, Pro-Humanist FREELOVER wrote:

> Dr. Bernstein, in his own words, discussing a glucose level of 83. As
> for
> HbA1c, Dr. Bernstein uses a personal HbA1c to glucose conversion table
> that is different from the charts generally available on the net, and
> the disparity between the charts -and- Dr. Bernstein's personal chart
> widens as HbA1c levels rise.

In reading Dr Bernstein's webpage with the 83 mg/dl normal average, it
looks like the data is in whole blood assay values, as Dr. Bernstein page
mentions this was quite a while ago in an old office location. The
conversion to plasma assay is 1.12 x whole blood assay and this converts
the 83 mg/dl whole blood assay to 93 mg/dl plasma reference assay that is
used in the labs and in new BG meters. So I would use 93 mg/dl as the
plasma reference average for the normal (non-DM) population.

The change-over to plasma reference was about 1999 and caused a sink in
the DM community. The upper limit for the whole blood to plasma
reference conversion is 15% or 1.15 x whole blood assay value.

So be careful with this 83 mg/dl. It may be an old assay methodology.
--
Jim
Still kick'n! Low tech T1 4/86, no complications. T.75? 7/2011
Email mangled: change SeeSig2Fix to mindspring for utopia. (Where?)

Alan Mackenzie

unread,
Aug 15, 2013, 5:12:40 PM8/15/13
to

Hi, Jim.

Jim Dumas <j-d...@seesig2fix.com> wrote:
> On Thu, 15 Aug 2013 10:53:53 -0500, Pro-Humanist FREELOVER wrote:

>> Dr. Bernstein, in his own words, discussing a glucose level of 83. As
>> for
>> HbA1c, Dr. Bernstein uses a personal HbA1c to glucose conversion table
>> that is different from the charts generally available on the net, and
>> the disparity between the charts -and- Dr. Bernstein's personal chart
>> widens as HbA1c levels rise.

> In reading Dr Bernstein's webpage with the 83 mg/dl normal average, it
> looks like the data is in whole blood assay values, as Dr. Bernstein page
> mentions this was quite a while ago in an old office location. The
> conversion to plasma assay is 1.12 x whole blood assay and this converts
> the 83 mg/dl whole blood assay to 93 mg/dl plasma reference assay that is
> used in the labs and in new BG meters. So I would use 93 mg/dl as the
> plasma reference average for the normal (non-DM) population.

> The change-over to plasma reference was about 1999 and caused a sink in
> the DM community.

Do you mean somewhere where the community washed its hands? Or that the
said community suffered a drop in numbers? I don't get this bit.

> The upper limit for the whole blood to plasma
> reference conversion is 15% or 1.15 x whole blood assay value.

> So be careful with this 83 mg/dl. It may be an old assay methodology.

I remember doing a BG test on a T0, and the result coming out to the mid
nineties. This would have been about ten years ago.

> --
> Jim

Jim Dumas

unread,
Aug 15, 2013, 7:44:44 PM8/15/13
to
On Thu, 15 Aug 2013 21:12:40 +0000, Alan Mackenzie wrote:

>> The change-over to plasma reference was about 1999 and caused a sink in
>> the DM community.
>
> Do you mean somewhere where the community washed its hands? Or that the
> said community suffered a drop in numbers? I don't get this bit.
>
>> The upper limit for the whole blood to plasma reference conversion is
>> 15% or 1.15 x whole blood assay value.
>
>> So be careful with this 83 mg/dl. It may be an old assay methodology.
>
> I remember doing a BG test on a T0, and the result coming out to the mid
> nineties. This would have been about ten years ago.

Hi Alan,

In the late 1990s, there was a switch in BG meters to simulate lab plasma
reference glucose values. So the fingerstick whole blood data was
converted to an equivalent plasma BG value by increasing by 10-15%, with
12% chosen as the average. So data taken on an old Accu-Chek 2, for
example, must be multiplied by 1.12 to get the reference lab equivalent BG
value. This change caused much confusion/arguments at the time.

The major difference between the whole blood assay and the reference lab
assay is the sample is centrifuged. The fibrous components go to the
bottom and the fluid left has a higher concentration of glucose. Thus
the 12% higher value. Since your BG meter can not spin the sample to
remove cells/fiber, etc., the fluid volume appears larger and the glucose
concentration seems lower, like 83 versus 93 mg/dl.

In the late 90s, the Docs were tired of explaining this difference to
their patients and had this change introduced to simulate the lab assay
data. But the fingerstick assay is still whole blood and not plasma so
the computer in the BG meter does this 1.12 x whole blood value for you.

Note that old BG meters before this ~1998/9 or so will read lower.

So I'll send my old meters to Pro to keep him afloat.

Alan Mackenzie

unread,
Aug 16, 2013, 5:58:27 AM8/16/13
to
Pro-Humanist FREELOVER <love_fo...@hushmail.com> wrote:

> On 15 Aug 2013 04:11 AM ,
> Alan Mackenzie <a...@muc.de> wrote:

>> Pro-Humanist FREELOVER <love_fo...@hushmail.com> wrote:

>> What exactly does Dr. Bernstein promote as "normal". I suspect it is
>> more likely to be a typical value of 83, not an average.

> Dr. Bernstein, in his own words, discussing
> a glucose level of 83. As for HbA1c, Dr.
> Bernstein uses a personal HbA1c to glucose
> conversion table that is different from the
> charts generally available on the net, and
> the disparity between the charts -and- Dr.
> Bernstein's personal chart widens as HbA1c
> levels rise.

So what? As I keep saying, HbA1c values are totally valueless when the
actual BG values are available. You test something like 10 times a day.
What does it matter how Dr. B. does his conversion?

> - - -
> August 10, 2013
>
> Q&A with Dr. Richard Bernstein
> http://diabeteshealth.com/read/2013/08/10/7955/qanda-with-dr--richard-bernstein/
> - - -

>> I think you originally chose this high intensity regime as an experiment,
>> trying to maintain tightly controlled BG levels around "normal" levels.

> The high intensity regimen was advocated
> by my doctor, after I had gone to him due
> to my hypoglycemic unawareness causing
> severe lows that had become seriously
> problematic.

Yet this regime hasn't worked. You're still getting far too many severe
lows. This is why your new endo should be recommending some radically
different approach which might work.

> His primary interest was
> simply in reducing chances of long-term
> complications, and really, the only preven-
> tative as for severe lows was/is blood
> glucose testing, something I resisted
> doing until I was, basically, forced into
> doing it due to the exigencies of hypo
> unawareness.

You've been a T1 ~50 years, nearly all of that time without BG testing
(hence almost certainly with high average BGs). If you haven't suffered
long-term complications after all that time, what makes you think you're
in any danger now?

>> > Obviously, severe hypos are not desired, .....

>> Aren't they? By whom are they not desired? In your next post, try
>> writing "I don't want severe hypos" and see how difficult you find it.

> I desire ZERO severe lows. I don't want
> them, I don't like them, I recognize they
> raise my short-term death risk.

HOORAY!!!! That's the first time you've said anything like that on this
newsgroup. Take that attitude with you when you visit your new endo.

> I do
> realize that my chances of having them
> are increased due to my HbA1c (glucose)
> goals being so close to normal, but I'm
> not willing to raise my HbA1c goals, due
> to my fear of long-term complications,
> and do think a CGM (continuous glucose
> monitor) is the best solution possible
> given the current technological limita-
> tions (if I can afford it).

Maybe. I don't think you are wise to give a low HbA1c value absolute
priority over frequent hypos. Maybe a CGM is the best thing if you are
so determined.

>> > .... and
>> > new tech (non-invasive glucose monitoring,
>> > glucose-responsive-insulin, artificial pancreas,
>> > new non-immuno-suppressant implants with
>> > glucose-aware insulin production, -or- restor-
>> > ation of insulin production via some non-im-
>> > muno-suppressant method) seeks to 'norm-
>> > alize' glucose levels with minimal or, ideally,
>> > *no* risk of severe lows.
>
>> Pie in the sky. You'll be decades dead before signicant new technology
>> starts making any difference.

> Unusually pessimistic point-of-view, and
> I'm much more hopeful/optimistic than
> that. Until 1922, Insulinitis was a death
> sentence. With insulin, it's no longer a
> death sentence, but it continues to be
> a condition that has a huge cost (quality-
> of-life-wise, dollar-wise, risk-wise) that
> only technological advancement offers
> a viable/potentially doable escape for.

I've been diabetic since 1965. There have been few, if any,
groundbreaking technologies invented since then. I'm not holding my
breath any longer.

>> > Dr. Richard Bernstein, he communicates
>> > that for those who are on a low carb/low
>> > insulin regimen, with discriplined/regular
>> > exercise, have less in the way of low glu-
>> > cose -and- severe low glucose. I worry
>> > that the diet is, for the overwhelming
>> > majority, not all, undoable, ....

>> That's irrelevant. It's undoable for you. It's fine for lots of others,
>> apparently.

> What percentage of those battling Insulinitis
> are on Dr. Bernstein's severe low carb and
> low insulin regimen for 5 years or longer?

> I suspect it's somewhere along the lines of
> less than one percent, but if anyone has
> any information on that, please share.

You simply don't know, neither do I. But the guy is (or was) a
practicing (?spelling) doctor, so what he's written is presumably
something he's tried out on real patients and found to work.

>> > I hope he [PHF's new endo] sees the value of a CGM, but
>> > don't know if my out-of-pocket costs
>> > would allow me to afford one.

>> Consider what impact those costs would have on your quality of life.

> If I can't afford it, I can't have one.

It may be you can afford it, but only by giving up something else, such
as an annual holiday.
0 new messages