• Guest - w'd love to know what you think about the forum! Take the 2025 Survey »

Carb Counting And Exercise Questions

E

evilclive

Well-Known Member
Messages
502
Type of diabetes
Type 1
Treatment type
Insulin
Hi -

I'm 46yo, fairly fit, T1 for 20 years. Currently injecting Fiasp + Levemir. Blood sugar moderately well controlled, though I can see peaks of up to 20 occasionally (every few weeks), and will hit 11-13 after meals, though that comes down fairly quickly. I'm on a Libre, so can see the peaks. I try and do an hour a day of exercise in the evenings on work days - mostly cycling, though also one run/week, and weekends I'll do a longer walk or bike ride.

My Diabetes team are very keen on me doing more carb counting than I do at the moment. I'm less sure than they are that it's a great idea. Here's my reasons :

I've informally carb counted for years - I've got an idea of how much food corresponds to how much insulin, and will happily change doses to fit what I eat. (though my normal meals are fairly regular, so portion size and hence insulin dose don't change much through the week)

The biggest thing which affects the amount of insulin I need is exercise. On a working day, I can take 20u of Fiasp + 12u levemir with breakfast, do a 2km walk, and be stable through the day sitting at a desk, with a couple of snacks. On a weekend day if going for a walk in the hills (eg 10mi), or on a cycling day, I can take 8u of Fiasp, same Levemir and then have to eat through the day after the carbs from breakfast have gone.

On the second day of little exercise (eg it's raining a lot) sugars start to go a lot higher too - that 20u + 12u might go up, or additional injections, and the snacks have to cut down a lot, despite being hungry.

This tells me that accurate carb counting can be completely irrelevant - there isn't a sensible starting point for accurate insulin/carb calculations, so my informal guesswork is actually good enough.

What I'd like to do is be able to adjust my insulin a bit more dynamically, but obviously that's hard with the basal on a long acting insulin - once that's taken, I can't reduce it, so I've got to eat to correct instead. But my diabetes team insist accurate carb counting is the important thing before we can look at other options.

Any thoughts?

(obviously this is just a brief summary, there's a lot more to say, so I'm not after detailed analysis, but an idea if I'm wildly inaccurate or if others have experienced similar issues would be useful)

cheers,
clive
 
Hi @evilclive, (love the username )!!, As I think you may be inferring, the amount of carbs does not fully correlate with what happens to one's BSL after eating them. I say that having been on insulin for 51 years, the last 7 on an insulin pump, but on MDI Novorapid and Levemir before the pump for some 15 years or so. The following is based on that experience, not as health professional advice or opinion.
If you look up mendosa.com, and Glycaemic values you will see definitions for Glycaemic Index (GI) and Glycaemic Load (GL) ( I have corrected the American spelling)!. The higher the GI (where foods in terms of the rise and length of time of blood sugar in the blood are compared to glucose which is 100%) and higher the GL (like a density figure of carbs in the food ) the greater effect on BSL the food has. And the admonition is to ideally consume foods with GI < 55 and GL < 10. Of course one has to allow for the fact that the overall GI and GL of a mix of foods in a meal is likely to be different to the individual food components of that mix (if you look at Cereal with Milk the GI tends to be lower than for the cereal by itself). Low fat ice-cream has higher GI than full fat ice-cream, because the higher fat content slows the absorption of carbs moreso than with the lower fat content ice-cream.
The other point about diet is seen by reading up on the Low Carb Diet - examples such as in the threads and information available on the home page of this website. The less carbs one has (provided calories, vitamins, minerals etc requirements are met in other ways) the less BSL rise there is. Of course, one's insulin doses etc would need to be modified for such diets and also being aware that some approx. 50% of protein is converted in the body to glucose appearing gradually nearer the 3 hour mark after the protein is consumed. (there are a number of Type 1 diabetics (T1Ds) who are on low carb or nearly zero carb diets). Of course during an illness with rising BSLs ketone formation may be a concern with a low/no carb diet and a doctor may advise not only insulin dose adjustment but some increase in carb content of one's diet also.
With exercise, I have read about and experienced the agonies of trying to exercise if one's BSL at start of exercise is > 13 mmol/l. At those BSL levels the liver seems to put out glucose from storage in response to exercise and that makes BSLs far worse, so I only exercise at levels around 5 to 11 mmol/l ( that allows for an up to 15% error range with use of the glucose meter).
I used to play squash, cycle at top speed etc and found that the adrenaline release pushed the BSLs way up ( and then they would come plummeting down some 6 + hours later as the depleted glucose storage in my muscles, seems almost literally suck up glucose from my bloodstream to replenish itself.
Walking, as you do, is less vigorous and I just reduce my basal rate on my pump 10 to 20% for about 4 hours for the 1 to 1 1/2 hour walk, as well as reduce the bolus dose before the following meal by about the same % and watch BSLs particularly around the 6 + hour mark. I used to do something similar with Novorapid and Levmir. I found eating food before exercising only made me feel bloated and sometimes the rise in BSL was unsatisfactory. Of course, I take some spare food to eat in case I end up walking for longer and find my BSL is dipping a bit.
I hope that answers some of your questions and provides some knowledge with which to improve things, always in consultation with your dsn and doctor. Best Wishes for a more normal BSL experience !!
 
Re GI and GL - I gave the examples I did because they were based on identical breakfasts, but dramatically different insulin requirements due to exercise. So no GI/GL change.
Re exercising on high BSL - I've not really found that a problem, possibly because if I've got a high BSL in that situation, it's normally from food I've just eaten rather than insulin I haven't taken. Eg two days ago, I went out for a run with BSL peaking at 14 just before I set off, corresponding with some food I ate (snack rather than meal), and I finished at about 5 mmol/l. Eating before I went out gave me the food to do the run (7 miles, 187m ascent).
Re low carb diet - I'm not terribly interested in that. I seem to work quite well on a fairly carby diet, though it needs rather more bolus than my DSN really wanted.
It might be worth noting that I consider 1-1.5 hours walking a bit of an amble Longer walks will have rather more effect, as will hillier ones.
Interesting re dropping the bolus after exercise - I don't tend to change the next meal dose that much, though it's possibly hard to say since my evening meals are nearly all after exercise. The bit I'm looking to tweak if I can is the period during exercise for the longer days.

My question was around carb counting and if I'm alone in feeling it's not necessarily going to be as useful for me as for some others because of the variance introduced by exercise.
 
Perhaps if you used a Libre or other continuous BSL monitoring device you could better track what is happening with your BSL during exercise. That might also show you better what happens with a starting BSL >14 mmol/.
Also remember that there may be days where you are not exercising. Do you really need an identical food on those days ? And exercises effect on BSL may vary from one time to the next with factors like air temperature, season, time of day etc.
Yes, I set up my pump bolus program for grams of carbs vs units of insulin BUT for carbs around the same low GL/GI, otherwise I would be changing the program for high vs low GI foods.
There maybe a day where exercise is delayed e.g. until late afternoon and that usually puts people at some risk of hypos in early morning hours.
Running does increase the risk of joint injury, stress fracture compared to walking.
 
Hi evilclive. No you are not alone. If you are eating the same meals with the same doses and getting even blood sugars then I think you are intuitively doing what you need. Do you pay for your FSL? If not then carb counting courses are sometimes a mandatory part of getting funding for a cgm and maybe this is why you are being encouraged to take this option?
I exercise a lot and was given a pump because it allows for much finer tuning of my MDI insulin regime to cover the different types of exercise that I do either by temporarily reducing or increasing my hourly rate for a specific time period or by adjusting the bolus dose by 50% for cardio (just as you seem to be doing manually!). For weights I add 25% though because the stress produces glucose!
I think you are experiencing the seemingly random after effects of exercise with both insulin and glucagon coming into play in response to the demands placed on your body; you can control the insulin and the food but glucagon and gluconeogenesis are obviously not within your control so it has to be guess work but perhaps a combination of pump and cgm would give you the best tools. Again you may need to go on that carb counting course just to acccess funding for a pump.
 
Hi @evilclive - you present an interesting set of questions.

First up a couple of observations regarding what you've posted.
  1. Regularly hitting 12-14mmol/l as a post-prandial glucose level is really too high, even if you aren't staying up there all that long. I'd suggest that you may need to look at changing your bolus time to earlier before you eat than you are now, as regardless of the duration of that number, it is too high.
  2. When you talk about making adjustments, you seem to be focusing on the bolus insulin, and I think that's erroneous based on your comments. You say that:
    evilclive said:
    On a weekend day if going for a walk in the hills (eg 10mi), or on a cycling day, I can take 8u of Fiasp, same Levemir and then have to eat through the day after the carbs from breakfast have gone.
    Click to expand...
    Which is suggesting that it's not your bolus that's the issue (which lasts 4-6 hours) but rather your Basal. You may benefit from reducing your morning basal insulin on days when you know you are going to be exercising like this.
I think your team is trying to help you, and they're correct to do so, but there are a couple of places that you can start from.

Firstly, I think you need to do a basal test to confirm that you're Levemir is really at the correct level. I am suspicious that it isn't for two reasons. Firstly your comments on exercise and secondly, for the majority of people, two doses of Levemir are required daily for best background glucose management. You can find details of how to do this here: https://mysugr.com/basal-rate-testing/

Once your basal is confirmed to be okay, it's worth looking at your Carb ratio and Correction factor and confirming that these are correct. That would make carb counting a lot easier.

In answer to your question relating to exercise, carb counting and variability, I find it a lot easier to manage glucose levels with exercise (and I do plenty) with well set basal levels and carb counting. The simple reason for this is that it gives you a base from which you can then adjust insulin levels up and down in order to manage exercise more efficiently.
 
So is your DSN inferring that you are taking too much bolus for the food you are actually eating, but then eating snacks later to counteract this. You mention that you enjoy your snacks and would be hungry without them. Perhaps what you need to do is bolus correctly for the intake, and if you do fancy a snack have a bit more bolus to allow for this. You may find you do not actually need the snack, but if you do then just go for it and enjoy it. It sounds like you know your own body, particularly for exercise requirements.
 
 
The gluconeogenesis can be triggered by exercising at blood sugar levels > 14 mmol/l , so that at least is preventable. And if one chooses less adrenaline-fuelled activities that can moderate the BSL response.
 
kitedoc said:
Perhaps if you used a Libre or other continuous BSL monitoring device you could better track what is happening with your BSL during exercise. That might also show you better what happens with a starting BSL >14 mmol/.
Click to expand...

As I said in my first post, "I'm on a Libre". When I said what happens with that starting BSL > 14 is that it goes down fairly quickly, not suffering the problems you mention, that was based on the information the Libre is giving me.

Running does increase the risk of joint injury, stress fracture compared to walking.
Click to expand...

Are you suggesting that running is a bad thing? Jim Fixx would probably argue with that, except his heart gave out...
 
"Once your basal is confirmed to be okay, it's worth looking at your Carb ratio and Correction factor and confirming that these are correct. That would make carb counting a lot easier."

I don't currently have a Carb ratio and Correction factor. The reason I don't is that I reckon they vary by a factor of two or more - which is really quite high, and basically renders any calculations using them irrelevant. Instead, I'm hoping to work out a more reactive scheme, working out different basal levels for different activities (sitting at desk, out on the hills, night, day, etc) - but this is quite hard with a long acting insulin, because there's no opportunity for correction.
 
Draco16 said:
Hi - what does this mean in terms of your HbA1c score?
Click to expand...

Nothing too scary - somewhere in the 50s at the moment. The things I'm looking to work on at the moment are the basal levels to avoid lows during exercise and while I'm sleeping.
 
Hi @evilclive, your A1c does as you say suggest moderately good / moderately ok control, but armed with a Libre and as an active person I think you should be getting that below the 48 target. Regularly at 11-13 post meals, with the occasional 20, probably means fairly often in the mid teens given that a1c? That’s not great.

I think you’re taking their carb counting suggestion too literally. They don’t mean exclusively base your insulin on set carb ratios. That would be madness.

For me, insulin dose takes account of
1. Current bs and trend.
2. Carbs (type and nutrition eg fats of meal overall)
3. Exercise coming up and in last day.

Those are the big 3, but also stress, tired or not, alcohol consumed or not, temperature, time of day are significant others. Then as we know there are 100s of others!

So given the above, if I had 50g of carbs in front of me, my dose could be anything from 3.5 to 6 units given the other factors. But usually 5. But counting the carbs is a key part of the equation.

They really mean get better at calculating your insulin dose. So each of those 1,2, 3 factors get good data on each. And better at timing perhaps as mentioned... do you pre-bolus?

Basal - the advantage of levimir is that it can be adapted to exercise and respond over that day or so. I’m on 7 / 7 split normally, but this weekend I’ve 2 days on the hill and will go to 6/6 as they’re going to be long days.

The run where you started at 14 (and fell to 5) I feel is too high to begin. I start similar at say 7 or 8 to give a buffer but then a glucose tab / wine gum every 15 mins or so keeps me topped up and flat.

So I don’t think you’re completely wrong, carb counting in isolation is certainly not the answer, but I do think you need to think about your insulin equation a bit more. Carb counting accurately is certainly part of that.

Apologies if this reads harshly, but as another 40 something male, who loves being active and on the hill I hope you can crack it!
 
The thing is, I do do these calculations, only informally rather than formally. It helps that the meals I'm taking bolus for are pretty steady - it seems my body has a pretty good idea of what the right amount of food is. (eg if I ignore my body and keep eating to empty an accidentally oversize plate, that means things will be high).

Glucose tablet - bleargh Wine gum, maybe for walking (though I'll normally use something slower for that), but hangs around in the mouth a bit long for running. Right now the jelly babies are doing the job - tasty, and dissolve quickly. But I'd rather have the food in me at the beginning for an hours run, because I know it can be done. And this is partly where my problem comes - I'd like to be able to drop my basal for that time (exact time to be worked out by experiment - working out when to drop and when to come back up again), and that's not an option with a long acting insulin, levemir or otherwise (*).

But I'm not going to be allowed anywhere near a pump without going through all the calculations - and as I said at the beginning, those calculations being based on food aren't going to work any better than my current estimates. You seem to acknowledge this by saying 3.5-6 for 50g - you've calculated 5, but changed it by 20-30%. I've worked out 8u-20u for my 145g breakfast. I might well have started at 13u or so when first diagnosed (20 years, no, my breakfast hasn't changed in that time...), and added and subtracted based on experience to get to where I am right now. I do bigger adjustments than you, and possibly need even more sometimes - the adjustments dominate.

I think I'm saying my point is that despite not doing the accurate counting, I'm actually remarkably close to how people end up when they are doing the accurate counting, because I've got the experience. And this means I think I should be able to work a pump and manage doses - but I'm not going to be allowed to give it a go.

Re sounding harsh - I'm ok with you describing what you're doing, that is helpful. And you are engaging with the question I'm asking, which is good.

(* tried toujeo, that was worse, flat through 24 hours is really bad for me)
 
evilclive said:
I think I'm saying my point is that despite not doing the accurate counting, I'm actually remarkably close to how people end up when they are doing the accurate counting, because I've got the experience.
Click to expand...

evilclive said:
(HbA1c) somewhere in the 50s at the moment.
Click to expand...

With the A1C and the bs readings you describe you seem to have reached an impasse on this point with your diabetes team.

Have they said what numbers they would like you to achieve?

Going on the inputs of carbs, insulin, exercise and the general underyling trend that your bs readings are too high, could you broadly:
i) lower carbs but keep others the same?
ii) increase insulin but keep others the same (pre-bolus timing could be a useful opton for you here as well)?
iii) increase exercise but keep others the same?
iv) or combinations thereof?
 
No, I don't have any defined targets.

What I'd like to do is lower basal insulin at certain points in the day, because my main concern is being low, not high. I know you're trying to work on what you see as my high BSLs, but I'm not really interested in discussing those right now. And I suspect they're not as bad as you think they are.

For example you've said a run where I started at 14 and ended up as 5 was bad, and you think I should be reducing the starting BSL then feeding through the run. What I want to do is reduce starting BSL and not have to feed through the run. I know it's possible with the right tools, but I don't have access to those tools right now. I'm not really worrying about a brief peak of 14 (remember, I do have the pretty graph, this isn't a fingerprick, so I do know it's a brief peak).
 
Last edited:
Hi, ok you could try a few things.

1. Do you split the Levemir into two injections a day? Gives you more flexibility to plan for and respond to exercise. And the big advantage of full 24 hour basal coverage.
2. If you move the timing of your bolus forward a touch (experiment 0-10 mins) you can avoid those spikes you mention (and get a smoother profile) that yes may come down quickly, but are spikes nonetheless.
3. For the runs and cycles of 1 or 2 hours - could you eat a light snack shortly before but don't bolus for it, or under bolus for it. eg you start at 8bs reading, before run you eat eg banana, some slow release carbs but no bolus. This would normally raise you up to eg 15 but the exercise offsets it so no need to eat during the run itself, bs stable. I do this. Takes some experimentation.

So something like...

Whole day or half day on mountain - reduce morning Levemir
More intense run or cycle of a one or two hours - eat lighter snack before but don't bolus / under bolus for it
 
1 - Levemir is already split - the tweaks I'm looking at are at a finer level than the split levemir can manage
2 - The spike we've been talking about doesn't come with a bolus
3 - Um, that's what I've been talking about - that spike is me doing precisely that. You've changed from saying it's a bad thing to a good thing now
 

No, 8bs to 8bs is stable.

14bs to 5bs is not stable.

You get the carbs and the timing right to achieve the former trend.

As I say it takes some experimentation to get this right.
 
You must log in or register to reply here.
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn More.…