Does anyone have experience with lactate testing for endurance exercise?

[marty313]

I went all in for triathlon since March 2022.
I am 59 and have DM1 since 2004. Since March 2023 I took up lactate threshold testing, both for running and cycling. I need to load carbohydrates before exercising - usually 120 g muesli and 8 IU insulin (some 60 % of the dose with the same meal and no exercise).

Most of the time, my lactate curves look horrible - I start with a lactate of 6 even at low exercise intensity. Is this because I exercise under ketoacidosis? My blood sugar at the start of the exercise is often 14 mmol/l. I mail-ordered some ketone strips for the TaiDoc meter, but I got a faulty batch, so I have not been able to test my hypothesis, but of course I could check ketones in my urine (which did not occur to me until now).

Disclaimer: I read in Olbrecht's book on lactate testing that one should not bother about lactate testing unless one performs at a national level. I live in Norway. I sometimes meet Blummenfelt in our local pool. Since there are not many DM T1s performing triathlon at an age of 60 I decided that I qualify. At the very least structured exercise testing once a month has given me new insights into my diabetes, so it is probably not all a waste of time.

 

[SimonP78]

No, but it sounds interesting, let us know you discover.

I really don't know what lactate levels to expect, do you use a sensor similar to the CGM sensors to measure lactate interstitially? It would be interesting to see a time-profile of exercise intensity vs blood glucose vs lactate.

As a side note, I'm interested that you carb-load so much before exercise, I tend to eat very little to avoid having too much IoB, but I realise that is in direct contradiction to what is suggested for non-diabetic athletes, and as I need to take some insulin these days to cover even a small breakfast (dawn effect), it probably wouldn't hurt too much to eat more. I have some experimentation to do - as ever, I'd not want to get bored

 

[marty313]

No, but it sounds interesting, let us know you discover.

I really don't know what lactate levels to expect, do you use a sensor similar to the CGM sensors to measure lactate interstitially? It would be interesting to see a time-profile of exercise intensity vs blood glucose vs lactate.

As a side note, I'm interested that you carb-load so much before exercise, I tend to eat very little to avoid having too much IoB, but I realise that is in direct contradiction to what is suggested for non-diabetic athletes, and as I need to take some insulin these days to cover even a small breakfast (dawn effect), it probably wouldn't hurt too much to eat more. I have some experimentation to do - as ever, I'd not want to get bored

Dear Simon,
Excellent! I did not expect that anybody would be interested.
I found zero information on lactate testing in diabetics on the net, so somebody has to get the ball rolling. The important thing about exercising for me: It keeps the boredom out of my life!

I bought my first lactate test in September 2022, two weeks after my first triathlon season, at the local professional football team "Brann". The reason I took the test was that my maximum cycling heart rate reproducibly topped 160 /min, when 220 - age was 160. (I have since learnt that this formula is rubbish - maximum heart rate is individual, in my case 172/min measured with Garmin's ECG chest strap ddring a competition.)

The test was at 9 a.m. I had cycled to the test site (6 km) after my standard breakfast (which is 120 g muesli mix without added sugar) + 8 IU Novorapid insulin. The people at "Brann" use the protocol for Norwegian elite athletes ("Olympiatoppen"): I ran on a treadmill with 1° incline. 5 min running, then 1 min pause for measuring blood lactate after pricking my finger. Speed was incremented in steps of 1 km/h every 6 min. As expected, my lactate rose slowly at first and then jumped up to approx. 6 mmol/l at speed between 12 and 13 km/h (see screenshot #1), so my lactate threshold was approx. 12.5 km/h at a heart rate of 158 / min (so 160 /min cycling was still inside my normal capabilities).
(N.B. Some establishments do lactate testing only on certain time slots in the afternoon. Forget it when you have DM1. Everything needs to be standardized for lactate testing.)

Over the winter I went heavily into running, increasing my longest runs from 5 to 10 km and from 5 km/week to 15, and my running speed increased.

In March I took up lactate testing. I train in mesocycles of 4 weeks, increasing the load from week 1 to 3. Week 4 is recovery. I do my running lactate test on Saturdays on a level footpath with roofs from 2 large buildings about 1.8 km apart. The cycling test is on Sunday morning at home. I put my racing bike with Garmin Rally 100 power pedals on a hydraulic bike roll. 180 W warmup for 15 min, then increases of 30 W every 6 min while I prick my fingers after 5 min. (No pause cycling!) I wear a chest strap for reliable heart rate measurements.
I measure from my fingertips using a TaiDoc TD-4289 meter. After pricking your finger wipe off the first drop of blood, then produce some more blood and suck the blood up with the test strip. (We 've done this before, haven't we?) The measurements can be downloaded from the meter after the entire testing protocol is finished. A test strip costs about € 3. At first it is difficult to get enough blood, in the final stages sweat becomes a problem when you do your own measurements without an assistant (which I do).

The problem was that no lactate test since March 2023 (I did more than 3 sets of them, running on Saturdays and cycling the day after) was "normal". I started with lactate values around 3 that quickly increased to 5 - 8 even when I am exercising clearly in my aerobic zone. At first I thought that my meter was defect. But the findings are pretty much reproducible. I attach a screenshot of my combined running/cycling lactate test on 2 successive days in May 2023.

I am quite suprised that I have a pretty high lactate already at low exercise loads. My theory is that hyperglycemia due to my aggressive carbohydrate loading. I eat quite a lot so that my blood sugar shoots up to 14 mmol/l and above because I don't want to experience hypoglucemia when I am swimming in the middle of a cold lake at the start of a triathlon. At the end of my swims, blood glucose is anything between 7 and 14, and I don't want to go below 7. The theory is that my carbohydrate loading induces ketoacidosis that in turn blocks my metabolism of fatty acids so that my entire metabolism is based on glucose so that lactate piles up even at very low loads.

I have not yet been able to test this hypotheses. I ordered ketone strips for my TaiDoc meter, but I was sent a faulty batch. Next time I do my testing, I ought to test ketones in my urine before exercise and after. (I did systematic urine tests for ketonuria before and when I was under dapaglifloxine, both resting and after excercise and NEVER was able to document ketonuria.)

So here is the full documentation - I invite your comments.

 

[Antje77]

Dear Simon,
Excellent! I did not expect that anybody would be interested.

You can trust that many of us are interested and I expect that quite a few of us have hit the 'watch' button for this thread.

To be honest, I don't understand half of it, and it also doesn't apply to me, I think, being a fat couch potato with the occasional short burst of exercising. But I still think it's interesting and will read along!

 

[marty313]

You can trust that many of us are interested and I expect that quite a few of us have hit the 'watch' button for this thread.

To be honest, I don't understand half of it, and it also doesn't apply to me, I think, being a fat couch potato with the occasional short burst of exercising. But I still think it's interesting and will read along!

Thanks a lot!
Lactate testing is the established way how professional endurance athletes find out what level of exercise intensity they can sustain over approx. 30 min. You can use this information to guide your training. 80 % of training of an endurance athlete is supposed to be under that threshold.
When muscles have to exercise hard in the lack of sufficient oxygen, they start producing lactate as an intermediate product. When lactate starts accumulating that's your body's way of saying enough.
I did my first lactate test to find out if a heart rate of 160 / min was within my limits as conventional wisdom (220 - age) suggested that 160 was my maximum heart rate. Fortunately, it wasn't.

When I started lactate testing on my own, I got very strange results and I found absolutely nothing on the topic on the Internet. Then I re-discovered this forum and I decided to post my observations.

 

[kdg]

Hi I am a T1D myself, I will summarise what I know on it later for you. I remember going down this rabbit hole a while back with training and T1D !

(mod edit)

 

[marty313]

Hi I am a T1D myself, I will summarise what I know on it later for you. I remember going down this rabbit hole a while back with training and T1D !

(Mod edit)

Thanks a lot! I am looking forward to your post.
I am glad I found a forum of like-minded people. To be honest, I hate my diabetes and don't want it to limit what I do, but diabetes is stronger than me if I am not smart. And: I am in a much better shape and feel that I have a richer life than most non-diabetics at my age, so I should not complain.

 

[IncogKeto]

Back in Jan 2022 Abbott announced that it was developing products that would be released under the brand name Lingo and which would be aimed at general consumers rather than for the management of conditions such as diabetes.
They're hoping to provide people with continuous measurements of glucose, ketones and lactate. I'd imagine that would be ideal for what you're doing. A smartphone on your handlebars reporting those three metrics in (almost) real time would likely provide you with some great insights into what's happening in your body whilst you're exercising.
I occasionally do a web search to see if there's any news on release dates but my guess is that it won't be until 2024. The ketone aspect of the sensor seems to be progressing through the trial phases. Web pages under the Lingo brand have recently started appearing, but at the moment it just appears to be a Libre 2 CGM on an overpriced subscription model and a gimmick called Lingo Counts.

This sticky biosensor could help you lose weight on a keto diet, or avoid a DUI

Using tech originally developed to help manage diabetes

www.techradar.com

Buy Lingo CGM Biosensor & App - Lingo by Abbott

Lingo is a biosensor and app that offers a coaching programme. Choose your Lingo plan to help build habits and adjust your nutrition.

shop.hellolingo.com

 

[SimonP78]

I've had a read, will read again to try to take it all in!

Where does one start?! I agree it's great to have more information. I can see the appeal of a blood based lactate inflection point, though I do wonder how much better that is than the data generated by e.g. Garmin/determined from analysis of activity HR data? I've not looked into it in any detail, I'd be interested to hear your thoughts @marty313

I don't think the relationship between duration above the lactate threshold and repeatability/recovery is particularly clear - operating above lactate threshold is not sustainable, and if one does need to spend time above this level, the next time one does it it's harder. It would be interesting to understand whether the lactate threshold changes during exercise, in which case being able to see continuous lactate might be useful to keep track of how many "matches have been burned".

I also think it's interesting to understand the relationship between lactate threshold/general lactate metabolism and the presence/absence of insulin and excess glucose and therefore whether aiming for a particular level is desirable from the point of view of lactate threshold on the one hand vs not going hypo on the other.

The same goes for continuous ketone monitoring, as far as I understand the switch from partly fat burning to completely carbohydrate burning is progressive, I can see that it would be useful to understand where one sits on this sliding scale, and if it changes during extended exercise to be able to monitor it in real time. Again understanding the interactions of insulin and excess glucose would be very useful - probably more so than lactate from my point of view as I am (hopefully) well under my lactate threshold when doing long distances on the bike. but it would be useful to know where I sit on the burning-fat vs burning-carbs spectrum to know how much I need to eat, whether I should be doing some bolusing, etc., or to tune power output to suit what I've got left to eat, etc.

 

[marty313]

Dear Simon,
Thanks a lot for considering my posts so seriously.

On Sunday, I performed my first Olympic triathlon. That's what I was training for, but I did not write this as yet, thinking it would have been bad karma. To be honest, the thought of having to swim 1.5 km in open water gave me sleepless nights at times. This despite having had elite swim coaches since Jan 2019. Cycling 40 km did not scare me as I am a life-long cyclist, but running 10 km was also challenging, given that I had started run training only on 27 March 2020 at a time when I still had 10 km overweight. (Now I am 87 kg at 183 cm with a BMI close to 25.)
The official times: Total 3:01:53, swim 36:25 (wetsuit mandatory at water temp. 15°C), T1 2:47 (for once I was not dizzy after the swim, but I had to wheel my bike 70 m until the mounting line), cycle 1:22:58 (could have been better, I cycled 73 km the day before to get known to the course), T2 1:28 (incl. 70 m wheeling my bike), run 58:13 (2 loops of 5 km each). Pace according to Garmin 2:11 swim, cycling 29.3 km/h, running 5:43 min/km. My Garmin registrations (Garmin Forerunner 955, HRM Pro+ pulse belt, Rally power pedals) show that I used constant pace troughout each leg, i.e. exactly the same stroke rate during the entire swim, the same average power and cadence on my bike (216 W, 82 /min), and the same time of 29 min in the first and the last 5 km. That means that I was performing close to my functional treshold in all 3 disciplines.

I go into all these details as the whole point with lactate testing is to find the exercise intensity that you can maintain over 20 - 30 min. So based on Sunday's competition, my functional threshold bike power (FTP) is probably 230 W. Garmin uses some proprietary algorithms. The cycling FTP is based on a time/power analysis: What power can you maintain over 1, 5, 20, 30, 40... min? Carmin claims it measured an FTP of 238 W on 13 Aug 23. Garmin estimates functional pace when running based on heart rate variability. According to Garmin, my treshold pace when running is 5:16 min at a HR of 154/min. We measure a lot of surrogate endpoints here! They are all in the same ballpark, but they are actually different and not the same.

In the light of all these measurements, the results of my lactate testing are probably correct when I assume the treshold intensity at about 4 - 6 mmol/l. (The level of lactate at which you assume lactate treshold 2 is a topic of discussion in itself: Where precisely is the inflection of the lactate curve?)

However, what does surprise me is my HIGH lactate levels at LOW intensities. High lactate at low intensity indicates that your muscles metabolize glucose rather than lipids. My suspicion is that my carbohydrate loading (I will provide details in another post) leads to ketoacidosis, which leads to a relative block in lipid metabolism. But so far, I have not been able to confirm this. Under my last lactate test, I wanted to monitor ketones in my blood (the TaiDoc meter has test strips not only for lactate but also for ketones), it was just that I had been sold expired test strips for ketones and I got an error from the meter every time I tried a ketone measurement. I could measure ketones in the urine, but then the urine I accumulate over a 50 min lactate threshold test will be a mixture of urine produced at low exercise levels and at maximum. (Extreme as I am in my approach to many things in life, I have no intention of cycling with a bladder catheter.)

I did make a literature search on PubMed, but I found no literature on this. I ask a colleague of mine, who is a diabetologist (I am a nuclear physician), if she wanted to do a study on this, but I suppose it is impossible to find 4 more patients with DM 1 in Bergen that do high-end endurance training. There is none that I know of in Bergen Triathlon Club. So this forum is the best platform I know.

 

[SimonP78]

On Sunday, I performed my first Olympic triathlon. That's what I was training for, but I did not write this as yet, thinking it would have been bad karma.

Well done!

registrations (Garmin Forerunner 955, HRM Pro+ pulse belt, Rally power pedals) show that I used constant pace troughout each leg, i.e. exactly the same stroke rate during the entire swim, the same average power and cadence on my bike (216 W, 82 /min), and the same time of 29 min in the first and the last 5 km. That means that I was performing close to my functional treshold in all 3 disciplines.

They look like really good numbers.

I go into all these details as the whole point with lactate testing is to find the exercise intensity that you can maintain over 20 - 30 min. So based on Sunday's competition, my functional threshold bike power (FTP) is probably 230 W. Garmin uses some proprietary algorithms. The cycling FTP is based on a time/power analysis: What power can you maintain over 1, 5, 20, 30, 40... min? Carmin claims it measured an FTP of 238 W on 13 Aug 23. Garmin estimates functional pace when running based on heart rate variability. According to Garmin, my treshold pace when running is 5:16 min at a HR of 154/min. We measure a lot of surrogate endpoints here! They are all in the same ballpark, but they are actually different and not the same.

intervals.icu is quite interesting re power data, with comparisons against age group cohorts, etc. it will pull data from Garmin and/or Strava. Garmin says my FTP is 243W based on something, though looking at the power log interval curve the most I've averaged for 1h is 186W. intervals.icu thinks my FTP is 273W based on one particularly hard 5min effort. It would take some serious motivation (like a non-UK length hill) to make me try to maintain that for an hour!

I don't race though, nor do I try to go out and destroy myself for an hour, as I'm targeting much longer duration rides, so I guess it's a case of me not doing the "right" kind of riding to validate (or otherwise) the model. It's good that it does use short max effort powers (which I do need to do sometimes) along with really long duration power and tries to fill in the gaps as I wouldn't otherwise do a max effort of e.g. 4h, but I would probably want to know what I could theoretically do, so that I can work out where to sit to avoid too much fatigue.

If you're not racing, and even if you are (as there are multiple ways to get an average number, and also because courses are not generally flat), I think being able to determine recovery time for a given power/duration would be interesting, along with any tapering that occurs as the proverbial matches are burned, which are perhaps places where actual lactate testing (real-time) would be really useful to determine how you respond.

The newer xx40 Garmin headunits (I don't have one, but have read the blurb) have some sort of race planner function where it will tell you want power to maintain over the course of a race taking into account the course profile, and presumably an individual power curve, but there must also be some of the recovery stuff in there. GoldenCheetah generates plots that models that calculate things like W' (available work above CP) and tau (W' recharge rate), though I don't believe there's any tapering applied. Perhaps some new research to be done or I need to scour the literature better!

However, what does surprise me is my HIGH lactate levels at LOW intensities. High lactate at low intensity indicates that your muscles metabolize glucose rather than lipids. My suspicion is that my carbohydrate loading (I will provide details in another post) leads to ketoacidosis, which leads to a relative block in lipid metabolism. But so far, I have not been able to confirm this. Under my last lactate test, I wanted to monitor ketones in my blood (the TaiDoc meter has test strips not only for lactate but also for ketones), it was just that I had been sold expired test strips for ketones and I got an error from the meter every time I tried a ketone measurement. I could measure ketones in the urine, but then the urine I accumulate over a 50 min lactate threshold test will be a mixture of

That is interesting, I assume that biochemical reaction is an established fact? Is the lactate escaping from the cells and into the blood stream (i.e. the cause of the inflection point) the cause of fatigue, or just something that occurs at a similar time so can be used as a proxy? I'm sure I've read something about this somewhere, I'll have to do some digging. I ask because I wonder if your high lactate levels at all times are reducing performance?

urine produced at low exercise levels and at maximum. (Extreme as I am in my approach to many things in life, I have no intention of cycling with a bladder catheter.)

Lol, I don't blame you!

I did make a literature search on PubMed, but I found no literature on this. I ask a colleague of mine, who is a diabetologist (I am a nuclear physician), if she wanted to do a study on this, but I suppose it is impossible to find 4 more patients with DM 1 in Bergen that do high-end endurance training. There is none that I know of in Bergen Triathlon Club. So this forum is the best platform I know.

I guess this is perhaps your best bet: https://www.teamnovonordisk.com/research/ though I've not looked for that specific topic so no guarantee I'm afraid!

 

[marty313]

On Sunday, I performed my first Olympic triathlon.

I promised more information regarding my system of carbohydrate loading. I apologize that the next paragraphs are very detailed, but if they weren't, the information would be useless.
I use MDI with Novorapid at meals and Tresiba 10 IU at night. Since I swim s lot and have very sensitive skin I have so far said no to a pump. I use Abbot Libre 3 CGM. I had a reproducible skin reaction to the Libre 1 but not Libre 2 or 3.
I cycle to work every day (except when I run to work once a week). At breakfast I eat 120 g of muesli mix without added sugar and inject 9 IU Novorapid. Cycling is 5 km, 140 m down and 70 m up. My blood glucose peaks at 14 + mmol/l. On some days I have to inject 2 or 3 IU when I arrive, 15 % of the time I get hypo before lunch.
When I swim I need to avoid hypos at all costs. The is no way to use the Abbot Libre under water, so carbohydrate loading is unavoidable.
When I am in control of my swim start, I eat 150 g muesli and inject 9 IU. When I am done swimming 1 to 2 km, my glucose will be anywhere between 7 and 16. It was never lower than 7. It falls quickly when I start cycling.
This is how I did my Olympic triathlon.
Race brief was 9:30 and swim start 10:00. I got up at 7 and had a moderate hotel breakfast with 9 IU aiming for normal blood sugar at 9. At 9:10 I ate 150 g of my normal muesli mix and injected 9 IU. I put on my wetsuit at 9:22. Don't stress and don't forget body glide on your neck! After race brief at 9:30 - 9:45 (which is mandatory to attend) I went swimming for some 200 m to adapt to the water temperature, which was only 15° C. I needed 36 min for the 1.5 km swim. When I came out of the water, my interstitial glucose was only 7.
Fortunately, I had connected my Garmin watch to my Libre 3 with xDrip 3 weeks before.How to keep my blood glucose up using my sport drinks under cycling and running I had trained in the last 3 weeks using the xDrip data field on my Forerunner 955. I thus knew what to do.
On the 40 km bike run I drank up my 2 750 ml bottles with High 5 Energy Drink (without added protein; 1 scoop per 250 ml). For the 10 km run I had prepared 2 sqeezy 500 ml bottles with the same drink. At the start of the run, glucose was 7, so I drank bottles up on the first 7.5 km.
I attach screenshots that show my glucose under the competition.
Two words of caution: Carrying a mobile phone is illegal in triathlon. You need to write the technical director in advance to get an exemption. This has been no problem in my last 4 competition. I also contacted Antidoping Norway: insulin is on the doping list (as it is a potent anabolic). If you are a pro you need a physician's declaration in advance. As age group athlete I can provide one retrospectively in the rare case I was singled out for testing.

Maybe the major take home message from my lactate testing is that I ought to aim for more normal blood sugar under exercising. At high blood sugar levels there is metabolic imbalance with inappropriately high lactate levels (to which I am however adapted; my muscles don't get stiff). The xDrip data field in Garmin lets me manage my glucose while exercising (which needs to be trained!) This has been a major game changer for me.View attachment 62818

 

[SimonP78]

Really interesting to hear the detail there - are those quantities for breakfast the total weight (mass) or are they just the carbs?

My immediate question would be do you need to inject so much basal? My gut feeling is that the combination of large breakfast + large basal dose means that if it's at all unbalanced, you will end up with a large BG rise/drop and you also have to deal with the long tail of IoB.

You may of course want that insulin to be active in your system so you can eat a large quantity of carbs as you're racing? Do you know roughly what your total carb intake is for each part of the event? For comparison on a long ride this weekend I was eating on the order of 30g of carbs per hour once I'd fixed my dawn phenomenon massive overshoot.

Going slightly OT (nothing to do with lactate), but for comparison:

I don't eat much before I go cycling, normally just my standard breakfast of 35g CHO (porridge with milk and a touch of sugar) plus a few white coffees (5g CHO each.) Before the advent of my dawn phenomenon this year, I was able to eat/drink then leave the house without any bolus and wouldn't go too high (say 14mmol/l - I need to look back at my records) and would then remain high for an hour or so while my liver decided to provide glucose. After that point it would start to drop and I'll need to start eating.

My approach to my commute is similar (though I only cycle into work twice a week at most), I would previously have breakfast uncovered then probably arrive at work with my BG at a decent level (either 1h/20km or 2h/40km each way, depending on which site.)

I become super sensitive to insulin when cycling, so when I go to work I have to eat lunch early and only dose for about half the usual quantity of carbs (I'll probably only dose for 1/3 for the 40km commute and accept running a bit higher towards the end of the day before I leave). I'll then have an uncovered snack before I leave for home and on the longer commute will likely need to snack while I ride. I've done a couple of 100 mile rides with lunch half way, and no matter how little insulin I take, it always makes me go low on the return leg (and I go high after eating as I've not had enough insulin to cover it while stationary.) My one and only partial success was on a 200km ride at the weekend where I had very little for lunch (single slice of bread bacon sandwich - ~20g CHO plus a white coffee or two) and didn't hang around for too long to avoid my BG going up too much on account of this.

Dawn phenomenon for me means I now need to take some bolus in the morning, I'm typically taking 3U as early as possible, but it doesn't seem to be enough so far as I've gone very high each time I've started early. I would normally need 7.5U or thereabouts to cover dawn phenomenon, a coffee or two, and 35g CHO of porridge if I were going to work (driving) or taking the kids to school, though if I'm just WFH I might need 6U just to sort out the dawn phenomenon (I should go for a walk or hop on the turbo for 20min).

I think there's a bit of additional adrenaline/nerves involved too as if I'm leaving that early it's going to be a long day going somewhere I've not been before. I'm certainly going to have to take more but have been hesitant as I don't want to overdo it and end up needing to eat more than I need to early on in the ride (I've got to carry it, plus force feeding isn't great for appetite later in the day.)

To sum up, it's really interesting to see the different approaches people have

 

[marty313]

Really interesting to hear the detail there - are those quantities for breakfast the total weight (mass) or are they just the carbs?

Dear Simon,
Our discussion is getting really interesting.
I will split my answers into separate posts.

I apologize for being imprecise: I referred to the weight of the muesli mix in grams.
The label claims 61 g CH per 100 g, so 120 g = 73 g carbohydrates. With milk and yoghurt surely 80 g.

 

[SimonP78]

No worries, I just wanted to double check to understand how much you're reducing the ratio vs normal, which was going to be my next question

 

[marty313]

My immediate question would be do you need to inject so much basal? My gut feeling is that the combination of large breakfast + large basal dose means that if it's at all unbalanced, you will end up with a large BG rise/drop and you also have to deal with the long tail of IoB.

You may of course want that insulin to be active in your system so you can eat a large quantity of carbs as you're racing? Do you know roughly what your total carb intake is for each part of the event? For comparison on a long ride this weekend I was eating on the order of 30g of carbs per hour once I'd fixed my dawn phenomenon massive overshoot.

The answer is that my carbo hydrate loading has evolved over time.
I was an active cyclist since my late teens. However, after I got DM in 2004, anything more than 20 min exercise would give me hypos until I was shaking and my vision was impaired.

I tried a cycle tour of some 30 km on a sunny day in 2008: it was a total disaster. I could eat as much as I wanted, I could not keep up my blood sugar. I have no explanation for this: Did I have too much insulin in my body? (At the time I used MDI with 12 IU NPH insulin at night, none in the morning.) I still had substantial endogenous insulin. Normally, the pancreas "knows" that you are exercising and shuts down insulin production and release immediately. Does the immunologically attacked pancreas listen to the body's signals in the same way or does it produce insulin at a low but constant rate? I never found the answer.

From 2015 I discovered that I could exercise for several hours if I had a regular breakfast without insulin. I took up sea padling.I even became coach. Then I needed 3 units rapid insulin with the breakfast. Now I have increased insulin to about 9, keeping the size of my breakfast constant. When doing endurance training it is important not to have caloric deficit. This can lead to overtraining and/or you break down muscle rather than fat. All active triathletes in my age group (most of them much faster than me) have some belly fat, so I don't care any more. The 4 kg fat that I have left, don't matter.

 

[marty313]

No worries, I just wanted to double check to understand how much you're reducing the ratio vs normal, which was going to be my next question

My rule of thumb is that I use 50 % of my regular insulin when I exercise. This has changed over the years. Initially, I used no insulin at all.

 

[SimonP78]

I assume it's because your endogenous insulin has been tapering off that you have had to increase (from zero) the amount of bolus you take?

I also tend to reduce bolus by ~50% before exercise, but it does depend on duration - so for e.g. an hour or two of badminton that's fine, and for an hour's ride it would also be ok, but beyond that point, I end up with higher insulin sensitivity with the tail end of the bolus (or perhaps I run out of liver glycogen, or both) so need to ramp up food intake quite significantly.

I've been diabetic since '85 and while I did cycle as a child I only took it up in earnest again during lockdown (2020) so I'm almost certainly not producing any of my own insulin, however I feel my response has changed, I now need to eat somewhat less than I previously did (roughly accounting for changes to insulin doses), whether this is down to efficiency or better fat burning I don't know.

When I started I adjusted my bolus down, and found that for rides of say >40km (which would take me >2h30) I would go hypo almost without fail and therefore need to carry vast quantities of food (and try to eat it, which wasn't fun either). I eventually settled on no bolus IoB and reduced basal too, as otherwise I still need to eat more than I want to have to carry with me/want to have to eat (I don't want to get stranded 50km+ from home because I'll run low if I can't eat, and I don't really feel like it due to the long activity).

I'm definitely running a calorific deficit while riding - at the weekend Garmin reckons I used ~4000 kCal (including basal calories) and I ate ~1500kCal from a rough tally (including the food I ate shortly before I left). But, I did probably make up for it later in the evening when I got home!

I guess the question is how many calories are needed during exercise to avoid performance tailing off. Certainly on occasions when I've ridden for 4h+ with elevated blood sugar due to (mis-)adjusting basal, I've felt performance tailing off at that point. I'm sitting mainly in Z1/Z2 while riding, so my glucose requirements will be less than for you racing as I should be doing a fair bit of fat burning (a reason for continuous ketone tests...) though I do feel that my liver stops supplying so much glucose from about 1h30 onwards, but it may still do so but just at a lower rate.

So many interesting things to work out

 

[marty313]

Well done!

I guess this is perhaps your best bet: https://www.teamnovonordisk.com/research/ though I've not looked for that specific topic so no guarantee I'm afraid!

Thanks a lot for the link, I will investigate. I am in the lucky position that there are close ties between Bergen and Denmark including direct flights to Copenhagen. Novo Nordisk is a diabetic-friendly company. (Abbott is in my experience the opposite. They are not at all interested in patient feedback, e.g. regarding their Libre Android apps.)

 

[marty313]

!

I guess the question is how many calories are needed during exercise to avoid performance tailing off. Certainly on occasions when I've ridden for 4h+ with elevated blood sugar due to (mis-)adjusting basal, I've felt performance tailing off at that point. I'm sitting mainly in Z1/Z2 while riding, so my glucose requirements will be less than for you racing as I should be doing a fair bit of fat burning (a reason for continuous ketone tests...) though I do feel that my liver stops supplying so much glucose from about 1h30 onwards, but it may still do so but just at a lower rate.

Joe Friel has written an excellent chapter (#3) on nutrition during exercise in "Your first triathlon" (available via Amazon Kindle). Of course all this needs to be modified for us diabetics.
The important thing is that the ability of the gut to absorb carbohydrates under exercise is limited and needs to be trained up. Untrained the limit is about 60 g CHO per hour. But it can be pushed up with training to 120 g/h.

1 scoop of my favourite sports drink (for 250 ml water) contains 44 g CHO https://highfive.co.uk/products/energydrink?variant=31620326654024
I drank about 4 scoops per hour in the last 2 h in my competition, that's a whopping 176 g per hour! I wasn't aware of this until you asked.

Another issue is salt loss. I was told by Allan Hovda (https://triallan.com/) that the concentration of sodium chloride in your sweat is constant, but varies from person to person. I actually sent in a sweat sample after a cycling lactate test to our hospital lab, so it is easy to measure. When Allan does Ironman triathlon he drinks water from the feeding stations supplemented by gels (which cause me terrible osmosis-induced toothache) and sodium tablets at the optimum rate to top up his system. I have not worked on this problem yet except for analysing my sweat...