Of Glucose and Finger Pricks — Is Personalized Nutrition Possible?

I sincerely think that personalized nutrition constitutes the binary that nutritional science should follow to instruct humans on which is the “best diet” for them and how to manage it.

The fact is that personalized nutrition, to be “personalized”, requires a much higher degree of quantification than most people are capable, or more so willing, to do.

In Wired to Eat, the reader learns about how a team of Israelis scientists have collected a sample of DNA and microbiome of several subjects, applying them subsequently on a CGM (Continuous Glucose Monitor).

Thanks to the different glycemic responses of the individuals at various meals (with high variation in macronutrient ratios), they have been able to develop an algorithm that can predict the glycemic response of the person to any given meal, with a surprisingly high degree of accuracy.

On top of that, the algorithm can develop two lists that consist of the top food choices and the worst food choices for the given individual.

Not bad, huh? Being as curious as I am, I would like to have access to that algorithm as well.

However, as for now, I must be content to follow the alternative DIY solution proposed in the book: playing with a glucometer to test the glycemic reactions to different kinds of carb-sources.

On paper, this procedure should help me to have a better understanding of my insulin sensitivity (associated with a lab test) and eventual particular auto-immune reactions triggered by certain foods because we know that stress-levels and inflammation can influence glycemic control.

What’s needed to run the experiment?

A simple glucometer and the food source that you want to test, with a proper amount of net carbs (usually 50g NET for a medium-sized male).

Are we giving anything for granted?

It is assumable that glucometers are reliable by law, right? Therefore two different glucometers should provide similar measures when using the same drop of blood, or at least the trend should be the same when several recordings are taken.

Even better, now that CGMs are available for the general public, if one wants to have an idea of his or her glycemic response, it should be enough to stick one of those expensive gizmos on the arm (or finger) and look at the results, without even have to bother spilling precious blood.

My Experiment

To verify the state of the art of this technology, during this summer I ran several experiments on myself, taking advantage of my regained high degree of insulin sensitivity (that required almost two years of hard old-school “Ketogains approved” dietary and physical work).

I have done so by testing different glucometers with various food sources (even a CGM) and then I have compared the results.

Let’s start with fasted blood glucose, here’s mine:

Confused?

Here’s my fasted blood glucose one hour after waking up:

Let’s test a quick carb source, shall we?

I hear you… which one is the correct?

According to my latest lab blood markers, the Precision Neo by Abbot is the most reliable one in term of taking a “realistic snapshot”.

Let’s compare it with a CGM made from the same producer, the Freestyle Neo.

The reality is that, even if the idea of monitoring blood glucose excursion after a meal — by having a standard scale to compare it with — is rational and enough simple to apply, the devices that are currently available on the market are:

1. Calibrated for people with diabetes, therefore the precision of the measurement is higher, the higher your blood glucose is.
2. Overall, not so much precise, especially the CGM (unfortunately). Moreover, this should be a serious matter of concern for people with type one diabetes, because, considering the huge margin of error between devices, the suggested insulin dose to be taken in relation to a state of hyperglycemia could be noticeably different by using the same drop of blood.

Should we then trash the home-made glycemic testing proposed in the book?

No.

However, given the relative state of immaturity of this technology, it is suggested to look at the results of this kind of home-made experiments with a grain of salt, placing them in the context of a safe and well-researched whole-food based diet.

The protocols that are suggested in Wired to Eat constitute a reliable baseline, and it is not a case that Rob suggests starting from there before running this kind of experiments.

Finally, I want to end this article by launching this question: if something that should be supposedly well established like glucose reading is so “relative” and device-dependent, how reliable could be something like BHB tracking for nutritional ketosis outside from a lab context?

At state of the art, chasing macros and micros (nutrient density) should be more important than chasing ketones or glucose readings.

Have you experienced something similar? Let me know in the comments below.