Posted by on Thursday 10 March 2016

Your Diabetes May Vary (again!) - BG variation after food

Two different carbohydrates, yesterday. (CC)
Which would hit your bloodstream faster, a banana or a biscuit? Pure glucose or a slice of white bread?

Well, in news that will come as a bit of a shock to some (and not at all to others), you can't actually know without checking for yourself.

Anyone with diabetes who has spent much time monitoring blood glucose levels before and after meals (especially if they have then compared their results with anyone else), may well already be familiar with this conversation: Person A: "I find porridge is great in the morning, it releases really steadily until lunchtime"
Person B: "Really?!? I can't go near porridge - it hits me like a train. All breakfast cereal does. Which is odd really, because Mars bars cause me no BG problems at all" etc etc. Rinse. Repeat.

Well a recent study published in 'Cell' by the Weizmann Institute of Science has demonstrated once and for all what we pancreatically-challenged types have suspected for a long time. That blood glucose responses to different foods are infuriatingly and often bewilderingly individual.

The study took 800 people without diabetes, around 54% of them were overweight and 22% classified as obese (with a BMI of over 30 kg/m2). They were connected to a Continuous Glucose Monitor for a week at a time, but the CGM was 'blinded' so participants had no way of seeing what was happening to their levels. CGM consists of a small sensor placed under the skin which records interstitial glucose values every 5 minutes, 24 hours a day. These values generally lag behind true blood glucose values by 10 minutes or so, but give a complete picture of what is happening before and after food and during sleep. People in the study recorded their food intake, levels of activity and so on using a smartphone app. They followed their normal routine, and ate as they normally would with the exception of breakfast, where they were assigned one of 4 standardised meals containing 50g of carbohydrate.

If you have spent much time online, sharing experiences with people with diabetes you may get a nice warm fuzzy feeling of "Aha! I *knew* it!" at the results. Here are a few things that came out of the research that caught my eye:

Responses to different foods were highly individual. Many people's BG rose rapidly after a standardised glucose meal as you would expect, but others were relatively untroubled by pure glucose, while eating bread sent their BG levels through the roof.

A graph comparing two participants shows an almost exact inverse response between, for example, cookies and bananas. In the light of this, any lists of 'foods which release slowly' can only ever be viewed as a general guide. Your own response to any food could well be very different.

In general, people who had higher BG responses after eating carried more weight than those with lower responses. The paper doesn't offer any thoughts as to whether these higher BGs make people put on weight, or whether the excess weight causes the elevated post-meal BGs, but in either case this association did not just occur at the extreme ends, but as a continuous range across the various weights.

The highest post-meal responses 'significantly correlated' with elevated (but still non-diabetic) HbA1c, waking glucose level, BMI and also age - all known to be risk factors for developing Type 2 diabetes. It looks to me like these are people whose metabolisms are already beginning to struggle.

Perhaps unsurprisingly, post-meal responses were shown to be very different to the same foods if eaten after resting/sleeping vs after exercising. Apparently the Pope is also Catholic.

A 100 people took part in a further study which allowed the researchers to develop an algorithm that successfully predicted post-meal BG responses from a variety of clinical, physical and 'microbiome' (eg gut bacteria) factors. Personalised diets were then able to reduce post-meal BGs effectively. In the Diabetes Online Community we simply call this ‘eat to your meter’.

The scientists wonder if working directly on reducing post-meal BGs would, over time, reduce some of the other associated risk factors including reducing weight, HbA1c and lowering risk of fatty liver disease.

What do I think this means for me?
Well first of all, it helps me realise that it's not just me being 'weird' after all. Different people really do react differently to different foods. Sometimes in completely inexplicable ways. I spent almost 20 years eating things that had been recommended as 'slow release' before beginning to systematically test my own responses to foods and discovering a few surprises and several absolute shockers that I had always believed were 'pretty safe'.

In general, it is easy to see that the proportion of carbohydrate in a meal could have a fairly direct impact on post-meal BGs, but this research goes some way to explain many of those 'Huh??!?' moments, and demonstrates that there's a lot more to it than that for each individual.

People make a lot of noise over 'low carb' vs 'high carb', but in truth, those definitions are of little interest to me. What I'm after is a varied, enjoyable, sustainable, LOW BG SPIKE diet that suits *me*. This research encourages me to continue looking for it.

"Your Diabetes May Vary", and all that.


  1. What is the full reference of this paper?

  2. Hello Anon

    The full text is linked above -

    Personalized Nutrition by Prediction of Glycemic Responses

    David Zeevi, Tal Korem, Niv Zmora, David Israeli, Daphna Rothschild, Adina Weinberger, Orly Ben-Yacov, Dar Lador, Tali Avnit-Sagi, Maya Lotan-Pompan, Jotham Suez, Jemal Ali Mahdi, Elad Matot, Gal Malka, Noa Kosower, Michal Rein, Gili Zilberman-Schapira, Lenka Dohnalová, Meirav Pevsner-Fischer, Rony Bikovsky, Zamir Halpern, Eran Elinav, Eran Segal

    Cell , Volume 163 , Issue 5 , 1079 - 1094

    Published: November 19, 2015

  3. Then there is the machine learning paper that demonstrated that gut bacteria had a large part to play in the variance. I'll try and find the link.