Please read the previous articles on my CKM use for more background.

1. PART I
2. PART II
3. PART III

My second Sibio CKM sensor was meant to monitor 2 weeks of heavy endurance exercise.  Trying to capture the stress of increased volume and adaptation to it.

Unfortunately that testing came to a quick end because the sensor took the exit quite soon.  As I took off my cycling outfit, the elastic band, that is strapped over the shoulders, slid of my arm as I was pulling it down and got captured by the sensor and ripped it off.  After some major cursing I have to be honest and blame myself.  When applying the sensor I first whipped the area clean with the alcohol tissue.  I think I didn’t wait long enough for the area to dry up, in essence, to let the alcohol evaporate.

But I managed to get enough data to derive some meaningful insights!

Legend:

• * red star: ingestion of food or decaf with cream or regular coffee with milk
• * purple triangle: start or end of exercise
• * green diamond: notable experience
• * blue star: into bed or out of bed

Note in the charts below that the sensor started recording at around 11:20 so a chart is 12 hours and 40 minutes of the previous day and 11 hours and 20 minutes of the next day.

On Day 1, I did go for a run in the morning at 09:00 and applied the sensor when I got back.

What we can see is that during the night, just like with my previous sensor, ketones are barely detectable.  Keep this in mind for the next night.

After the morning run at around 7am and having had breakfast, ketones hover around 0.5mmol/L

On Day 2, the exercise is interrupted with a pause of about 1 hour to drink a coffee with milk and some water.

We now see that night time has a more elevated ketone level after 3 training sessions in a short period of time (2 runs + a bike ride in 36 hours). The only thing I can guess at is that this somehow reflect the repair my body is going through, requiring a higher lipolysis rate to free up fatty acids and replenish my muscle and perhaps repair damaged (mitochondrial) membranes.

My Garmin watch measures stress level 24/7. This is what it showed me from midnight until I got up.

In comparison the following stress chart of a night where I’m well rested and have a low heart rate:

In the morning (end of the ketone graph), after the run and eating, I now experience higher ketone levels for a longer time compared to the day before. As the stressful night indicates, my recovery wasn’t 100% so a higher lipolysis rate is needed throughout the day?

On Day 3, I started cycling after about 2 hours from my lunch (6 eggs, lots of butter, garlic, half union, salt, surimi)

Notable observations:

• * On Day 1 and Day 2, as I started running there’s a drop in ketones and then it levels out for the remaining duration. This seems like a repeatable pattern.
• * On Day 2 and Day 3, starting the bike ride in both cases stabilized the level of ketones towards 0.5mmol/L.  On Day 2 it first  gradually went below 0.5mmol/L and then came up again.  On Day 3, I started with high ketones and it dropped to hang around 0.5mmol/L. Both rides were fairly equal in intensity.
• * On Day 2 and Day 3, in both cases stopping for a coffee break, my ketones increased fast.  Upon continuing the ride, the level dropped again but this time did not drop down to 0.5mmol/L.  In both cases, the green diamond indicates the moment where I felt weak, slightly hypoglycemic.

Hypoglycemic

So what happened?  Here’s a valuable lesson for me.

As I stopped for the break, naturally insulin will increase to stop the lipolysis as it is no longer needed at the sedentary level as it is while exercising.  Now the caffeine enters the body.  Caffeine stimulates cortisol release and cortisol antagonizes insulin action.  

As a result, lipolysis can continue, glucose release from the liver can continue and ketones rise as a consequence. Ketone uptake is reduced because the legs are now not generating so much lactate anymore so in essence the insulin resistance that caffeine is causing, is releasing too much energy into circulation.

What happens next is that a stronger bolus of insulin will be released in order to get all that releasing energy under control and back into storage.  Now I get on the bike again to continue my ride.  I need energy again but I just got extra insulin released so glucose and fatty acids will rapidly decline.  In both cases it took about 30 minutes to experience hypoglycemia and the level of ketones were around 0.9mmol/L and 0.8mmol/L respectively.

On Day 3, you see that after this hypoglycemic event (felt it for about 5 minutes), while riding, my ketones climb up to 1.3mmol/L.  On Day 2 I got home shortly after and ate something.  Finishing the ride and eating cheese with 30% fat shortly after caused an increase to 1.7mmol/L.

McArdle & exogenous ketones

Understanding what happened above, I’d like to refer to one specific study where they concluded that exogenous ketones offered no performance improvement in McArdle patients (GSDV).

As you can see from the protocol picture, they administered a ketone ester drink 25 minutes before commencing exercise.  Pretty close to the 30 minutes where I experienced the hypoglycemic effect.

You can see that this ketone drink caused their 3HB level to rise up to 3.5mmol/L.

Notice in the following chart what it did to the subjects’ insulin level.  Similar to my own experience.

Also pay attention to how low the insulin level stays at baseline for the GSDV placebo group (white squares).  Again, keep in mind that exercise started at 25 min after the drink.

The lack of benefit is not because ketones don’t work, it’s because the protocol used, caused a strong rise in insulin with a consequent inhibition of energy which we can see in the following charts. Insulin has a logarithmic effect, not a linear one so a doubling of insulin isn’t 2x as strong but more like 10x or even more.

For glucose:

Notice the GSDV placebo group maintaining glucose.

For fatty acids & glycerol:

Once again, notice how the FFA & glycerol rise up from 30 minutes onwards, 5 minutes into the exercise for the GSDV placebo group.

The lack of muscle glycogen is met with catabolic hormones which also breaks down protein for energy.  This is evidenced by the rise in ammonia.  A similar rise is noted for both the GSDV placebo and GSDV KE group but we’ve seen that glucose, glycerol and FFA are severely restricted in the KE vs placebo group and the main cause of that is the difference in insulin.

This actually indicates that insulin is a strong inhibitor of lipolysis and gluconeogenesis but is a weak inhibitor of proteolysis versus adrenergic hormones.

“No effect of oral ketone ester supplementation on exercise capacity in patients with McArdle disease and healthy controls: A randomized placebo-controlled cross-over study”

https://doi.org/10.1002/jimd.12484 https://onlinelibrary.wiley.com/doi/10.1002/jimd.12484

Thus, in conclusion what this study teaches us and as I’ve experienced myself.. Make sure you don’t start exercising when insulin levels are high.  You can never expect this to provide a performance benefit.  This may be an obvious idea given all the evidence but we don’t always think about our insulin level that way.

This has also been a valuable lesson to understand why certain studies doing energy interventions find no result or even negative results.  If they would time the ingestion and start of exercise right before insulin would rise

Energy Management

What I’ve learned from my previous sensor and this one and watching how all events influence ketone levels, I’m convinced that my ketone readings reflect on one hand the response to fatty meals (little surprise there) and on the other hand the need for glucose replacement combined with the level of energy required.

Simply put, if you don’t have a lot of glucose in reserve (in the liver), ketones will reflect the need for energy.  Given the lack of glucose, the majority of energy is all derived from your stored fat so naturally the rate of lipolysis determines the rate of glucose production from the glycerol backbone of fatty acids and the rate of ketone production. 

Under conditions of low energy requirement (sedentary), fatty acids themselves cover more of the energetic need so insulin, although very low on a very low carbohydrate diet, will be slightly raised so that it keeps ketone production under control.  As the liver is very sensitive to insulin, it allows glycogen to build up in the liver but this will be a slow process as blood glucose also needs to be maintained.  We’ll notice normal stable blood glucose and ketone levels around 0.4mmol/L

If there is too much energy (ketones and fatty acids) then insulin will manage to bring it down to the appropriate level.  If you add in endurance exercise then there is obviously a higher need for energy.  Both during exercise itself and in the hours after the exercise but also during sleep as your body needs to repair itself.

Training

So what can we do with this knowledge?  How can we make it practical?

Unfortunately I couldn’t follow the evolution across 14 days but here’s what I would like to know.

1. Will my ketones go down again during the night and will that mean I’m recovering well?  Adapting to the training stress? Need to train harder again?
2. As observed during Day 2 and Day 3, while training, could I use the level of ketones maintained as an indication?  For example if ketones would start to drift up, would that mean I have trained long enough?  Would consecutive training sessions increase this baseline, does it mean I’m getting more fatigued?

After the sensor got ripped off and I learned my lesson with the coffee, I changed my behavior.  Knowing what would happen, I made sure I got on the bike again as soon as I finished my coffee because I know it would liberate energy and I have to use that energy to avoid insulin bringing it down.  What I noticed from that is that my power/hr level improved.

Caffeine itself may increase the metabolic rate but having more substrate available is also required to drive a higher performance.  Excluding caffeine, could I measure a higher lipolysis rate somehow and see that as a sign of greater fitness?

I also did not do hard races or rides where I had to give up at some point.  How will that affect my ketones?  Will I be able to note an increase or decrease in ketones?  And will that happen before I start to feel like giving up?

There’s a lot of questions and I hope to answer some of them in the future.  I have 2 sensors left and will pick the moments carefully so I can at least have some of those questions resolved.

– – – – – – – – – T H E – E N D – – – – – – – – – –