Hi Paul,

As you know exercise greatly improves insulin sensitivity for several hours following vigorous exercise. Unfortunately, even if we could devise a means of measuring insulin resistance (versus blood glucose levels) after exercise, it's a transient condition that, for the most part, disappears by the next day. Therefore, I doubt that we could accurately infer any long-term effects by taking short-term measurements. I think that the long-term improvements in insulin sensitivity (as mesured by HbA1C) produced by exercise are related to changes in body composition, i.e. a decrease in fat, particularly abdominal fat, and an increase in overall lean body weight.

Larry

Hum, this email has made me realize that we have a different understanding of insulin sensitivity.

I thought insulin sensitivity was not something that varied day-to-day at all, but could only be altered over the long-term by diet and exercise. I was thinking it was to do with the structure of cell membranes or other places that accept signals from insulin. When the cell membranes and other places are ignoring signals from insulin, then glucose sits in the bloodstream and the result is T2. But I thought these were things that could not be changed immediately with exercise.

But you are saying something different - that insulin sensitivity improves immediately after exercise, and then the effect fades. Is that something that has a physical explanation in terms of cells and the liver etc?

I understand your final point - visceral fat is particularly bad, while lean muscle is good presumably because it's very good at accepting glucose.

(Incidentally, I've always assumed 'insulin sensitivity' and 'insulin resistance' are antonyms, not even quite sure if that is true.)

Finally, yes, maybe it's too optimistic to think there are ways to infer long-term effects, and identify best types of exercise, just with a glucose meter. Still what a great technology!

Paul.

Hi Paul,

I'll see if I can find an authoritative source to support my statements.

I am under the impression that immediately following vigorous exercise our metabolisms are closest to being normal. No doubt this is due to our bodies' need to quickly replenish glycogen stores in muscle tissue and the liver, and to supply glucose to power our vital functions. During recovery from exercise we have observed that blood glucose is lowered more rapidly than any other time. As we know it is insulin that is the mediator in the transfer of glucose from the blood into mucle and liver tissue. So if glucose is being transferred into tissue faster after exercise, and insulin is the means by which that is accomplished, then we may infer that our insulin receptors are working more efficiently at that time.

Larry

Hi Paul,

Here's one scientific study supporting my point.

Effects of exercise on glucose tolerance and insulin resistance.



Larry

Thanks for the reference. I need to chew that over even though I read it through a few times...

Hi Paul,

You're welcome.

Here's a couple of more studies demostrating short-term improvements in insulin sensitivity due to exercise.

Insulin-stimulated muscle glucose clearance in patients with NIDDM. Effects of one-legged physical training.



The effect of prior exercise on oral glucose tolerance in late gestational women

Ah thanks. A fair part of this is escaping me, but I have taken a good step forward. So there are two measurements that are relevant to measuring insulin resistance - glucose measurement e.g. in response to GTT, and the amount of insulin in the blood. In the last two sentences of the last abstract, for example, the glucose measurements have stayed the same, but less insulin was required to get that same glucose level, hence insulin sensitivity has increased. So a BG meter alone is not enough to get a measure of insulin sensitivity.

Hi Paul,

I think it's fair to say that even these scientists are inferring improved insulin sensitivity without really having any fundamental insights as to the exact mechanism(s) whereby exercise improves sensitivity. Sort of like my inference in my earlier posting.

They are measuring total glucose levels over time or glucose clearance rates while keeping the incoming glucose load constant, and determining whether or not the amount of insulin required to clear the glucose from the blood is decreasing. Since they see a decrease in the amount of insulin needed, from that they are inferring that insulin sensitivity has increased, but they still don't know exactly what the mechanism is. (Exercise certainly is not bringing beta cells back from the dead. ) They see some contributing factors, increased blood flow could bring more insulin in contact with more tissue cells, increased glycogen production in recovery certainly will clear the blood faster, etc. They also can rule out some non-contributory factors. Since they know that there wasn't any changes in body composition in the short time of these tests, they can rule out reductions in abdominal fat or increase muscle mass as being a contribution factor, even though we know in the long term these are factors that improve insulin sensitivity.

Larry

Yep right, it's a very complicated system and there are only limited ways to look inside to see what's happening. It's only a week ago that it was announced that osteocalcin is a significant hormone affecting BG, so clearly the basics are still in flux.

A different issue, but something that this thread made me think about - internet forums like this one prove there are people doing all kinds of experiments with diet and exercise to try to improve health. And medical researchers no doubt have an endless list of experiments they want to do. I wonder why there are no clinical trials over the internet, where researchers email diet sheets, exercise instructions etc and interested lay-people do the experiments, getting blood tests or other tests as required at a local clinic. It's probably not as ideal as recruiting locally and have indiividuals actually attending a clinic where the researcher works, but it seems like it could be workable. For one thing, if a clinical trial was a subject of internet discussion for everyone involved, like the discussions here, it might be very motivating for people taking part.

On that bit of hiking I did on Mt Rainier one of my reasons for using Dove dark chocolates (I realize that glucose would have been more scientific) was to see if I could use a lot of carbs. I didn't know if I had enough insulin/insulin sensitivity to keep my glucose levels between 80 and 110. Would I go low, would I go high.

When I fasted for a colonoscopy at about hour 21 I dropped down to about 40. I normal person, as I understand it, would have stayed above 70 in that situation.

One of my further suspicions is that a number of these extreme tests just with fasting/glucose/exercise might differentially allow one to indirectly answer a number of questions about one's entire glucose metabolism, duplicating in some broad way all the expensive tests Dr. Bernstein recommends.

Hi Paul,

No doubt there are a number of clinical trials on diabetes. Apparently that's how they discovered insulin in the first place.

Clinical Trials

However, there might not be many trials that fit your concept. I have seen some long-term studies in which the subjects were given a diet to follow over a long period and results monitored. However, as you say these types of studies are not ideal because they have loose controls, and a great deal is riding on the compliance of the subjects. They are generally not as highly regarded as other studies where the control is more rigid because of the clinical setting.

An other factor is that many of the blood tests being conducted are difficult enough to setup even in a clinical setting, and are simply not available at conveniently located "walk-in" labs. Sometimes catheters are installed in the subjects and glucose, insulin and other hormones, or some other medication is both continually monitored and automatically administered depending on the requirements of the study.

Larry

Hi Rob,

I'm not sure I'm following your statement. Are you saying that by measuring your blood glucose after certain events such as fasting for a fixed duration, eating a controlled portion of carbs, taking a measured duration and intensity workout, etc., that we should be able to predict our glucose response for different situations? If so, I agree within limits. After all, my suggestion to measure glucose response throughout a workout and adjust rest periods long enough to minimize spikes, is based on this premise of repeatability. By the way, in fairness to Dr. Bernstein, he recommends testing our carb tolerance in a similar manner.

However, in the real world I find it may not be possible to always avoid unexpected interactions between multple glucose altering events.

For example, today I had a blood glucose surprise that was no doubt caused by unintended interations between events.

I measured my BG before going to the gym, it was 112. I did one set in 30 minutes, rested 20 minutes and then did two more sets in about 45 minutes. After resting about 30 minutes I took my readings and was happy to see that my BG was 113 after exercise, just about the same as when I started exercising. So far, so good. My resting routine seemed to be working as predicted.

Then out of curiousity I decided to take an other reading after 45 minutes to see how far down my BG had gone. I still hadn't gotten around to eating yet. I was astonished to discover that my BG had spiked to 175 while resting! I'm guessing that this glucose reaction was caused by fasting. It turns out that all this working out and resting, and other distractions, had greatly delayed my lunch, so I had been fasting for about 7 hours. In addition, during this time I had a rather strenous workout. So I figure that my BG continued to decrease rapidly from the 113 reading until my liver said, "Wait a minute here! We need some more glucose to keep thing running!" So it dumped a bunch of glucose bumping it up to 175. So I go ahead and finally eat my lunch, a large portion of tuna salad and a large salad and 500 mg of metformin. One hour later the 175 reading is down to 97.

In summary, the whole point in this long-winded discussion is to highlight that even if we go to the trouble of precisely measuring "standard blood glucose altering events", unfortunately real life has a way of imposing combinations of non-standard events that make it a little more challenging to predict outcomes. This of course doesn't mean that we should abandon talking measurements. Without knowing how we react to these standard events we wouldn't be able to have a even modicum of predictability.

Larry

Larry - that was a fascinating incident. I don't think these sorts of singular events could replace the Berstein Battery (and in fact wish my docs would start doing some of the most pertinent). But it would be likely if one collected one's regular responses with some extreme responses that someone like Dr Bernstein would have some pretty good suspicions of what is going on.

Incidentally I have had only 3-6 readings over 150 in the 8 months I have been monitoring. All relating to eating a MODEST amount of carbs. Exercise has never put me over 125, I think. I need to do some charting. But I suspect my docs would not be interested in analyzing what I found.

That lack of interest is probably familiar to many people. It relates to a point that I was trying to make earlier about using the internet, which I'm not sure I communicated very well. My guess is that somewhere out there in the world are diabetes researchers who really would like to get such measurements. And in places like this forum, we are motivated to make and communicate such measurements. It seems like a missed opportunity that the internet is not used to get data to the researchers.

(OK this process would be completely different to traditional clinical trials, and probably any analysis and conclusions would be judged less reliable because the data is being collected from anonymous sources on the internet - still my guess is that it would be useful.)

Well what do I know. If I was doing research on diabetes and exercise though, I might just come to this forum and ask if interested people would keep a journal for me,

Paul.

Hi Robb,

Make sure that you take your measurements following an intense workout (> 80% VO2max). Don't wait to do all of your workout before taking a reading. If you are still doing HIIT a good test would be to take a reading after 15 minutes of grueling exercise, then continue to take readings while in recovery. Make sure to take a reading about 30 minutes into recovery.

I've attached a chart of a study of the glucose response of fit, non-diabetic subjects. FIG. 1. Comparison of responses during 40 min of moderate intensity exercise (50% VO2max) (white data points) and 15 min intense exercise (87% VO2max) (black data points) in normal young male subjects.

Notice that for the intense exercise the pre-exercise plasma glucose is less than 5.0 mmol/l and rises rapidly to about 7.8 mmol/l in about 30 minutes following the end of exercise. Converting mmol/l to mg/dl (what our meters measure) we get a pre-exercise reading of about 90 mg/dl rising to about 140 mg/dl. I should emphasize that these readings are the mean reading of an entire group of fit, non-diabetics. It is likely that fit, type II diabetics would experience even higher spikes.

Larry

Attached Images

Exercise Glucose.jpg (37.6 KB, 5 views)