The Porridge Experiment: Using a Continuous Glucose Monitor (CGM) to test my blood glucose after eating Oats and Peanut Butter for breakfast.

The Porridge Experiment.

Breakfast is either eggs or oats. And coffee.

I’ve tried all the other breakfast trends too; fasting, bulletproof and nutrient-dense smoothies, etc.

But, just like the experiences of many other people, these trends have dropped by the wayside in favour of simple, quick and consistently satiating food.

With so many people concerned about the negative health consequences of high carbohydrate diets, I thought it would be interesting to see what the effect of my daily bowl of porridge would be on my blood glucose. Does a ‘traditional’ and ‘normal’ breakfast of oats simply melt into sugar?

Furthermore, I wondered about two other common factors that might affect blood sugar. Would added fat and calories, in the form of peanut butter, affect my post-prandial blood glucose? Would exercise after my meal make a visibly significant difference to my readings?

Read on to find out more…

What are the goals of this experiment?

  • Goal 1: Assess any differences between ‘Low Fat’ porridge and ‘High Fat’ porridge on post-prandial blood glucose.
  • Goal 2: To assess the influence of physical activity or sedentary behaviour after ‘Low Fat’ and ‘High Fat’ porridge.

Warning! This is a N=1 experiment. My data is for your interest only. Your results may vary!

Outline:

  • Day 1 = ‘Low Fat’ Porridge, cycle to work.
  • Day 2 = ‘High Fat’ Porridge, cycle to work.
  • Day 3 = ‘High Fat’ Porridge, sedentary.
  • Day 4 = ‘Low Fat’ Porridge, sedentary.

The Porridge Recipe:

65 grams of organic rolled porridge oats (protein: 8g, fat: 3g, carbs: 42g)

  • Soaked overnight and cooked on the hob.

Toppings (consistent between ‘High Fat’ and ‘Low Fat’ versions)

  • 10g Sunflower seeds (protein: 2g, fat: 5g, carbs: 2g )
  • Half an ‘average’ sized banana (protein: 1g, fat: 0g, carbs: 13g)
  • Half an ‘average’ sized apple (protein: 0g, fat: 0g, carbs: 13g)

A hefty serving of Peanut Butter (added to make my ‘high fat’ porridge)

  • 90g of MyProtein crunchy peanut butter (protein: 23g, fat: 45g, carbs: 20g)

Unflavoured whey protein powder

  • One scoop when eaten with high fat/peanut butter porridge
    • (protein: 21.5g, fat: 1.5g, carbs: 1.5g)
  • Two scoops to balance out protein intake when excluding peanut butter
    • (protein: 41g, fat: 3g, carbs: 3g)

Total macronutrients and calories of each bowl:

  • Low Fat porridge (protein: 52g, fat: 11g, carbs: 73g, calories: 597)
  • High Fat porridge (protein: 55g, fat: 55g, carbs: 92g, calories: 1032 )

-> There is already a very significant difference in the fat and calorie content of these meals!

Other Factors:

  • As mentioned in my previous blog, I didn’t add any cinnamon or other spices.
    • Cinnamon can act to lower blood glucose after eating.
  • I drank decaffeinated coffee.

The Morning Commute versus The Morning Couch

  • From my house to the GP practice at which I was working at.
  • About 15-20 minutes of moderate cycling to cover a 5km uphill route.
  • On my day off (Day 3) I spent the morning after eating breakfast in the flat, either sitting or standing at my computer.

The Data Begins…

In all honesty, this bit is difficult to get through. It comprises rough graphs, rougher calculations, inferences and demonstrates my working. You can scroll to the end for the conclusions if you like.

Day 1

These numbers and the trace represent my glucose across the day.

After having breakfast at 07:15, my glucose started to rise until I got onto my bike. When I left the house at 07:40 my glucose was 5.4. A short 20 minute cycle later, it was 5.0.

When I stopped exercising, the blood glucose concentration appeared to rise again, until it hit a peak of 7.4 at 08:50.

→ This drop then rise around exercise makes sense – the activity of pedalling my bike will have consumed some glucose, shuttling it into my working muscles to fuel me up the hill, attenuating any postprandial rise in glucose. It also might represent diversion of blood from my gut to my working muscles, reducing the digestion of my breakfast and hence limiting absorption of glucose from my gut to the bloodstream. Resting after exercise might improve digestion, allowing greater absorption of food, and reducing consumption of glucose by my muscles.

My peak value of glucose was 7.4mmol/L at 08:49, taking 1 hour 15 minutes to reach it. 3 hours after eating my glucose was 5.5mmol/L.

It’s reassuring that these numbers are normal, but it would be pretty tough to go on writing and reading like this. These numbers are hard to make sense of, and even harder to make interesting.

Furthermore, as my previous blog post described, total values for glucose as measured by a CGM aren’t very valuable.

What about a way to gauge the general trend of my glucose?

1) Draw a graph comparing the different glucose traces.

2) Calculate an ‘Area Under the Curve’ (AUC)

What is ‘Area Under the Curve (AUC)’?

As discussed in a previous blog, CGMs can’t measure fine variations of blood glucose, hence finding a measurement tool that avoids spikes and prefers trends is better.  Calculating an ‘Area Under the Curve’ generates a numerical value representing the concentration of glucose in the blood over a time period.

I thought that AUC might compliment the use of graphs to assess overall trends.

Using a spreadsheet to map my glucose numbers against their time points, it’s possible to calculate an AUC value. I calculated the AUCs for this experiment across the 3 hour period following breakfast.

Porridge, Peanut Butter and Cycling

Here is the graph comparing the level of glucose against time for both ‘Low Fat’ and ‘High Fat’ porridge followed by cycling to work:

Hour 1

On the ‘Low Fat’ day (blue) I started cycling about 20 minutes after eating.

On the ‘High Fat’ day (red) I started cycling about 25 minutes after eating.

During these two cycles, the glucose level of the ‘High Fat’ porridge appears higher than the ‘Low Fat’ porridge. Although these plots appear different, the AUC for the hour following breakfast was 0.24 for ‘Low Fat’ and 0.23 for ‘High Fat’, which are remarkably similar readings despite the difference in appearance of the tracings. This AUC figure means that the amount of glucose in the blood during the same 1 hour time frame was very similar.

Hour 2

The intersection of the two graphs at ~50 minutes fits with my arrival at work, getting off the bike and into a shower. It signals the start of the working day. It also signals some subsequent differences in glucose!

It appears that the ‘Low Fat’ porridge continued to increase my blood glucose before it tapered downwards towards the two hour mark. On the other hand, the ‘High Fat’ porridge shows almost the inverse – an immediate drop in glucose followed by a rise.

Comparing the AUCs of each meal during the period between 60 and 120 minutes shows that the ‘Low Fat’ porridge had an AUC of 0.27 while ‘High Fat’ had an AUC of 0.24. Despite the dramatic differences in graph appearances, the numbers show only a small difference of about 11%. This small difference may suggest that during this hour I had more glucose in my blood after eating ‘Low Fat’ rather than ‘High Fat’, despite that each bowl of porridge had the same amount of carbohydrate.

Hour 3

In the final hour, between the 120 to 180 minute mark, both graphs appear to level out to a similar figure. The AUC results for both meals was 0.23. I take this to mean that both meals had been fully digested and that my body was returning to a state where it relies on its own supplies (e.g. fasting, using glycogen or fat).

AUC

Low Fat + Cycle

High Fat + Cycle

0-1 hours

0.24

0.23

0-2 hours

0.51

0.47

0-3 hours

0.74

0.71

Between 1-2 hours

0.27

0.24

Between 2-3 hours

0.23

0.23

Overall, both bowls of porridge had very similar AUCs throughout the entire 3 hour post-meal period, resulting in 0.74 for the ‘Low Fat’ bowl and 0.71 for the ‘High Fat’ bowl. I’d hope that they were similar as both bowls contained the same amount of carbohydrate!

The 0.3 difference might be accounted for by recipe error, failure to absorb the carbohydrate and/or failure to measure the blood glucose properly; it might have been absorbed too quickly into the muscles if I had cycled a bit faster, or perhaps was assimilated into the bloodstream too late, after the 3 hour sampling period had finished.

Porridge, Peanut Butter and Sitting on the Couch

Here is the graph comparing the CGM traces between porridge with and without peanut butter followed by a morning of ‘rest’.

Here are the AUC numbers for these rest days:

AUC

Low Fat + Rest

High Fat + Rest

0-1 hours

0.26

0.23

0-2 hours

0.48

0.44

0-3 hours

0.71

0.66

Between 1-2 hours

0.22

0.21

Between 2-3 hours

0.22

0.23

Comparing these two traces, it would appear that the ‘Low Fat’ porridge results in a greater rise in glucose within the first hour. This ‘Low Fat’ trace then drops back towards the ‘High Fat’ trace. In contrast, the ‘High Fat’ trace remains steady. The AUCs in the first and second hourly segments and overall reflect that the ‘Low Fat’ porridge resulted in higher levels of glucose.

Again, a difference in the total AUC at three hours is seen. Where is the disappearing glucose going?

Finally, this graph compares all of the experiments; peanut butter, no peanut butter, rest, exercise.

For completion, here are all of the AUC values:

AUC

Low Fat + Cycle

High Fat + Cycle

Low Fat + Rest

High Fat + Rest

0-1 hour

0.24

0.23

0.26

0.23

0-2 hours

0.51

0.47

0.48

0.44

0-3 hours

0.74

0.71

0.71

0.66

Between 1-2 hours

0.27

0.24

0.22

0.21

Between 2-3 hours

0.23

0.23

0.22

0.23

The goals of this experiment were to:

  • Assess any differences between ‘Low Fat’ porridge and ‘High Fat’ porridge on post-prandial blood glucose.
  • To assess the influence of physical activity or sedentary behaviour after ‘Low Fat’ and ‘High Fat’ porridge.

Did we meet these goals? What can we infer?

Peanut Butter on Porridge Delays Glucose Absorption

The differences in the 01:00-01:30hrs segment between High- and Low Fat after exercise appear striking on the graph; the ‘Low Fat’ increases while the ‘High Fat’ plummets, with a resultant difference in the AUC values across the hour time frame. In the following 30 minutes, their paths cross again, suggesting delayed absorption of the glucose from the high-fat porridge bowl.

The pattern is not as striking on the rest days, however, while the ‘Low Fat’ porridge results in a glucose rise, the ‘High Fat’ does not. The AUCs during the first two hours remain similar, suggesting that despite the rise and fall of the ‘Low Fat’ bowl, the overall glucose absorption was similar. Furthermore, the AUC of the ‘High Fat’ bowl exceeds that of the ‘Low Fat’ bowl in the final hour – suggesting that adding peanut butter to porridge allows blood glucose to play ‘catch-up’ later in the day.

Low Fat Porridge results in Faster & Higher Glucose.

Between 0-1 hours, 0-2 hours and 1-2 hours, the ‘Low Fat’ bowls have a higher AUC than ‘High Fat’ bowls, whether resting or exercising. This suggests that more glucose is absorbed from the gut into the blood during the early period of digestion.

Furthermore, the total AUC from 0-3 hours for ‘Low Fat’ porridge is higher than the ‘High Fat’ porridge, suggesting that more glucose is absorbed from meals without peanut butter compared to with peanut butter, despite both bowls having the same amount of carbohydrate.

Where did the missing Glucose go?

It’s apparent that both ‘High-Fat’ bowls of porridge have a total AUC less than those of the ‘Low Fat’ porridge. I don’t know why this is, but it could be due to:
1) error when I made the porridge;
2) differences in exercise/how my body has used the glucose;
3) measurement error by the CGM;
4) slowed absorption of the porridge, resulting in absorption that continued beyond the 3 hour mark and hence wasn’t measured or;
5) failure to fully digest the porridge.

Exercise Increases Glucose more than Rest.

The glucose measurement recordings following exercise are all over the place. They rise and fall much more readily than the measurements following the same meal but on a rest day. This isn’t what I had expected – I had expected that exercise would attenuate the glucose rise, rather than be associated with a greater glucose rise than resting alone.

Eating the same food on a rest day does not increase my glucose in the same way, suggesting that it isn’t food that creates the turbulence of the exercise graphs. Instead, perhaps exercise is what creates the rise in glucose, rather than food.

This would make sense – the porridge eaten for breakfast hadn’t had enough time to be absorbed, nor would its absorption be at a great enough rate to fuel ongoing exercise. Instead, the energy needed to pedal myself to work comes from my current stores of glucose in glycogen, or is created via the process of gluconeogenesis. The resultant glucose curve is less likely to be from the porridge I ate, but from the powerful effects of hormones on metabolic physiology during exercise; lowered insulin with elevated glucagon and cortisol.

In Summary:

  • Peanut butter appears to slow the digestion/absorption of glucose from porridge
  • Low Fat Porridge absorbed more quickly than High Fat Porridge
  • Low Fat Porridge absorbs more completely / we lost some glucose from the High Fat Porridge.
  • Exercise increases blood glucose significantly, perhaps more than porridge can alone.

Was this experiment worth it?

As I wrote in my previous blog, the CGM device can be a life-saving tool for an individual with diabetes.

For people with intact blood sugar regulation, a CGM is a tool of novelty rather than one that collects reliable hard data. It was fun and novel to use the device at the time, but trawling through the data has been hard work, while the conclusions, although demonstrating interesting physiology, have not been ground-breaking.

So, while I haven’t been able to shock and awe, I am reassured that:

  • The rules of physiology still stand
  • My physiology is normal
  • I’ve had a chance to think, plan and negotiate spreadsheets
  • Whether it’s a 500 calorie bowl of porridge or a 1000+ calorie peanut butter porridge bomb, my body handles it pretty nicely.

 

Thanks for reading

This was a marathon post but you made it.

Check out Part 1 and 2 in the series:
My attempt to correlate lifestyle factors with blood glucose using Continuous Glucose Monitoring (CGM).

Practical Experiences, Advantages and Limitations of Measuring Glucose using a Freestyle Libre CGM.

Feel free to share, comment or follow the page!

 

3 comments

  1. Really ! Surely no significant difference at your age. But me? I think my glucose would rise with carbs but not porridge.

    Sent from my iPhone

    >

  2. […] That’s a lot of output for a device that monitors only one input of glucose. Of course, you could argue that there is useful extra information from the raw glucose measurement; the absolute glucose value, its variability and its rate of ascent or descent would be some extra data that could be derived – as I tried in my previous blogs. […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s