This post is an update on my experiments measuring the effect of low-carb foods and dietary supplements on blood sugar.

This week, I have the results from low-carb cereals. Next week I’ll be posting results from my sleep study, followed by low-carb ice-creams.

Testing Queue:

• Baseline:
• Glucose re-test: In queue
• Low-carb foods:
• Meal replacements: 2/3 complete, (Ketochow previously reported)
• Flour replacements: 1/4 complete
• Tortilla: Reported
• Bread: 2/10 complete
• Snack bars: 0/8 complete
• Ice cream: 5/11 complete
• Cereals: This post
• Supplements:
• Vinegar: In queue
• MSG: Reported

Cereals

Summary:

I tested 7 low-carb cereals from 3 categories (nut & seed granolas, milk protein & sweetener blends, and protein & fiber blends). 

The granolas had the lowest blood glucose impact by weight (~7%  & ~15% of glucose for peak BG/g & iAuC/g). By volume, though, all except Catalina Crunch were very similar. 

The highest blood glucose by a wide margin was Catalina Crunch, a protein and fiber blend, with at 28%  & 62% of glucose for peak BG/g & iAuC/g. This likely stems from its use of potato and corn fiber, which are digestible despite being subtracted for the net carb count. 

On taste, all the cereals were good, but sweeter than I’d like (see Table above). Of the granolas, my favorite was the NuTrail, which had a strong cinnamon & vanilla flavor that paired well with the pecans, pumpkin seeds, and coconut. Of the more cereal-like cereals, my favorite by far was the Magic Spoon. The texture was shockingly like regular cereal and it was the only one of the bunch that didn’t taste strongly of the non-nutritive sweetener. For these experiments, I used the banana, but a tried a few of their other flavors and liked them a lot more (my favorite was maple).

Compared with the tortillas, I didn’t get as much direct value out of these measurements. While I liked the cereals, they’re too sweet and too low in nutrition for me to use for regular meals and keeping them around is too much of a temptation to break my diet. I might get a box every once in a while as a treat, but I won’t be incorporating them in to my regular rotation.

Does anyone know any other good low-carb cereals I should try?

Details:

Purpose

• To identify low-carb foods that taste good and have minimal effect on my blood glucose.
• To determine the effect of popular, literature supported dietary supplements on my blood glucose. 

Background

Before I got diabetes, my favorite breakfast was a bowl of cereal with milk. In the last few years, a ton of new brands of low-carb cereals come out, with some even available in supermarkets (some popular press articles here and here). 

Although the net carb counts look good, I’ve become very suspicious of the blood sugar impact of some of the dietary fibers used (see evidence of blood glucose impact of dietary fibers here & here). 

To see if any of available low-carb cereals would hold up, I decided to test them myself.

Design/Methods

Foods

I tested 7 low-carb cereals from 3 different categories:

• Nut- & seed-based granolas
• Milk protein & sweetener blends
• Protein & fiber blends

Full nutrient and ingredient info here. Key nutrition facts in the table below.

Procedure

At 5:00a, I took 4.5u of Novolog (fast acting insulin, duration of 2-4h), then drank a Ketochow shake (website, BG testing) at 5:30a. After that, no food or calorie-containing drinks were consumed and no exercise was performed. Non-calorie-containing drinks were consumed as desired (water, caffeine-free tea, and decaffeinated coffee). At 11am-12 pm, the substance to be tested was eaten as rapidly as comfortable and notes on taste and texture were recorded (before observing any change in blood sugar).

Blood sugar was monitored for 5h using a Dexcom G6. Calibration was performed 15-30 min. before the start of each experiment.

Data Processing & Visualization. iAUC was calculated using the trapezoid method (see data spreadsheet for details). Data was visualized using Tableau.

Medication. During these experiments, I took long-acting basal insulin each evening at 9pm (Lantus, 1.52u) and 2000 mg of metformin and multivitamin each morning at 5am. I did not dose for the experimental food ingested.

Data

• Formatted raw data, peak blood glucose, iAuC, and time to peak data
• Summary data, ingredient details, and taste assessments

Results & Discussion

Figure 1. Left – Change in blood glucose vs. time. Right – Change in blood glucose per g(food) vs. time

Figure 2. Left – Peak change in blood glucose per g(food). Right – iAuC per g(food). All values reported as % of the value measured for glucose.

Figure 3. Left – Peak change in blood glucose per cup(food). Right – iAuC per cup(food).

Changes in blood glucose as a function of time are shown in Figure 1. All cereals show a longer time to initial rise, less steep rise, and longer duration of impact than glucose, consistent with a slower absorption and metabolism. This profile is consistent with a mix of protein, starches, and a higher fat content.

There was a dramatic difference in the blood glucose impact of the different types of cereals by weight. The nut and seed granolas showed the lowest impact at ~7%  & ~15% of glucose for peak BG/g & iAuC/g. The milk protein and sweetener blends were about twice that at  ~15%  & ~35% of glucose. The worst of the bunch was Catalina Crunch, a protein and fiber blend at 28%  & 62%. It looks like the fibers it uses (potato fiber, corn fiber) have a significant impact despite being subtracted for the net carb count. 

The different types of cereals also had significantly different densities. Depending on whether your goals and preferences, you may care more about the impact of a volume of cereal instead of weight. Personally, I tend to eat a full bowl of cereal in a sitting (~1 cup), so that’s what I care about. When you normalize by volume, the differences between the granolas and the milk protein & sweetener cereals goes away, though Catalina Crunch is still much higher impact than everything else. 

On taste, all the cereals were good, but sweeter than I’d like (see Table above). Of the granolas, my favorite was the NuTrail, which had a strong cinnamon & vanilla flavor that paired well with the pecans, pumpkin seeds, and coconut. Of the more cereal-like cereals, my favorite by far was the Magic Spoon. The texture was shockingly like regular cereal and it was the only one of the bunch that didn’t taste strongly of the non-nutritive sweetener. For these experiments, I used the banana, but a tried a few of their other flavors and liked them a lot more (my favorite was maple).

Note: taste and texture observations were recorded when I ate the food. I.e. before I knew its impact on my blood sugar.

Thoughts & Next Experiments 

The food effect studies continue to go well. I’m still seeing very large differences in blood glucose impact, independent of the carb count, bolstering the conclusion that the digestibility of fibers varies widely. After I collect more data, I plan to go back and see if I can find a better metric than net carbs to predict the impact of a given food.

Compared with the tortillas, I didn’t get as much direct value out of these measurements. While I liked the cereals, they’re too sweet and too low in nutrition for me to use for regular meals and keeping them around is too much of a temptation to break my diet. I might get a box every once in a while as a treat, but I won’t be incorporating them in to my regular rotation.

Does anyone know any other good low-carb cereals I should try?

As always, please let me know if you have any thoughts or suggestions.

– QD

This post is an update on my experiments measuring the effect of low-carb foods and dietary supplements on blood sugar.

Testing Queue:

• Baseline:
  • Glucose re-test: In queue
• Low-carb foods:
  • Meal replacements: 2/3 complete, (Ketochow previously reported)
  • Flour replacements: 1/4 complete
  • Tortilla: This post
  • Bread: 1/8 complete
  • Snack bars: 0/8 complete
  • Ice cream: 2/11 complete
  • Cereals: 4/7 complete
• Supplements:
  • Vinegar: In queue
  • MSG: Reported

This week, I have the results from low-carb Tortillas.

Tortillas

Summary:

I tested 6 low-carb tortillas from 3 different brands. The winner on both taste and blood sugar impact was La Tortilla Factory, with about half the impact of Mission and Nutri-Rica tortillas (~13 vs. 25% of glucose for peak BG/g & ~30 vs. 60% of glucose for iAuC).

The key difference between the brands appears to be that Mission & Nutri-Rica use resistant wheat starch, as their main ingredient, which I’ve previously observed to have almost the same blood sugar impact as glucose. La Tortilla Factory, in contrast, uses oat fiber or cellulose fiber, which have essentially zero blood sugar impact.

Overall, I’m really happy with how this experiment turned out. The La Tortilla Factory tortillas have low enough blood glucose impact that I’ve started using them with meals (recipe here).

I looked for other low-carb tortillas that don’t use resistant wheat starch, but except for La Tortilla Factory and Mr. Tortilla (recommended by a commenter on the preliminary data), they all are either too high carb or high calorie (e.g. using almond flour) and I’d rather leave room for more fillings. I ordered the Mr. Tortilla ones, but UPS lost the package so I wasn’t able to test them before this post (will update once I get them).

Does anyone know any other good low-carb tortillas I should try?

Details:

Purpose

• To identify low-carb foods that taste good and have minimal effect on my blood glucose.
• To determine the effect of popular, literature supported dietary supplements on my blood glucose.

Background

I was in the supermarket recently and noticed that they’ve started carrying low-carb tortillas. The macros looked decent (2-5g net carbs/tortilla depending on brand and type).

I love tacos and wraps and it’d be great to have a convenient way to make them, but I noticed that some of the brands used resistant wheat starch, which I’ve previously observed to have a substantial impact on my blood sugar (33% peak BG & 76% iAuC vs. the same amount of glucose).

To see if any of available low-carb tortillas would hold up, I decided to test them.

Design/Methods

Foods

I tested 6 low-carb tortillas from 3 different brands. Full nutrient and ingredient info here & at the bottom of the post.

Procedure

At 5:00a, I took 4.5u of Novolog (fast acting insulin, duration of 2-4h), then drank a Ketochow shake (website, BG testing) at 5:30a. After that, no food or calorie-containing drinks were consumed and no exercise was performed. Non-calorie-containing drinks were consumed as desired (water, caffeine-free tea, and decaffeinated coffee). At 11am-12 pm, the substance to be tested was eaten as rapidly as comfortable and notes on taste and texture were recorded (before observing any change in blood sugar).

Blood sugar was monitored for 5h using a Dexcom G6. Calibration was performed 15-30 min. before the start of each experiment.

Note: I take a long-acting basal insulin (Lantus, 2u at 9pm each evening).

Data Processing & Visualization. iAUC was calculated using the trapezoid method (see data spreadsheet for details). Data was visualized using Tableau.

Medication. During these experiments, I took long-acting basal insulin each evening at 9pm (Lantus, 2u) and 2000 mg of metformin and multivitamin each morning at 5am. I did not dose for the experimental food ingested.

Data

• Formatted raw data, peak blood glucose, iAuC, and time to peak data
• Summary data, ingredient details, and taste assessments

Results & Discussion

Figure 1. Left – Change in blood glucose vs. time. Right – Change in blood glucose per g(food) vs. time

Figure 2. Left – Peak change in blood glucose per g(food). Right – iAuC per g(food). All values reported as % of the value measured for glucose.

Changes in blood glucose as a function of time are shown in Figure 1. The Mission and Nutri-Rica tortillas show a steep rise for the first 1-1.5h, similar to glucose, followed by a leveling off with a peak 2-3h after eating. This profile is similar to resistant wheat starch, their main ingredient (listed as “modified wheat starch”, but same thing). In contrast, the La Tortilla Factory tortillas show a slower rise and lower peak, consistent with their use of non-digestible oat fiber and cellulose fiber.

The difference between the brands can seen even more clearly by looking at the peak change and iAuC per gram, shown in Figure 2 and the blood glucose impact table. La Tortilla Factory tortillas have about half the effect of Mission and Nutri-Rica tortillas (~13 vs. 25% of glucose for peak BG/g & ~30 vs. 60% of glucose for iAuC).

Happily, I liked the taste of the La Tortilla Factory tortillas the best of all those I tested. The flour tastes like a real flour tortilla: chewy, not much flavor, and with a hint of sweetness. The wheat has a strong wheat taste and is slightly gritty, very similar to the taste of oat fiber (main ingredient). I really like the wheat taste by itself, but found it would overpower/obscure the taste of fillings when I used them in meals. The La Tortilla Factory tortillas are also the thinnest of those I tested, letting you put more filling in a taco or wrap.

The Mission tortillas were pretty good as well: thin, with decent texture, though not quite as good as La Tortilla Factory. The spinach and tomato basil were ok, but I prefer a neutral flavor so the tortilla pairs with any filling.

The Nutri-Rica tortillas had good flavor (flax seed), but were way to thick and chewy for me. I prefer a thinner & larger tortilla that can hold more filling.

Note: taste and texture observations were recorded when I ate the food. I.e. before I knew its impact on my blood sugar.

Thoughts & Next Experiments 

I’m really happy with how this experiment turned out. It reinforces my previous observation that resistant wheat starch is not really low-carb and I need to avoid it. But, the La Tortilla Factory tortillas have low enough blood glucose impact that I’ve started using them with meals (will post some recipes soon).

I looked for other low-carb tortillas that don’t use resistant wheat starch, but except for La Tortilla Factory, they all are either too high carb or high calorie (e.g. using almond flour) and I’d rather leave room for more fillings.

Does anyone know any other good low-carb tortillas I should try?

As always, please let me know if you have any thoughts or suggestions.

– QD

La Tortilla Factory – Flour

La Tortilla Factory – Wheat

Mission – Wheat

Mission – Spinach

Mission – Tomato

Nutri-Rica

Of the previous studies I’ve done, the most informative and useful for me was measuring the effect of different macronutrients and low-carb ingredients. 

I was especially fascinated, and disturbed, by the huge variation in blood glucose impact of fibers that are listed identically on nutrition labels. This makes me extremely suspicious of the so-called “net carb” count listed on most low-carb food. I suspect that there’s a wide variation in actual blood glucose impact for foods with similar nutrition labels.

To follow up on that, I’m going to test the effects of popular dietary supplements and low-carb foods to see if claims about them really hold up. For low-carb foods, I’m going to focus on the ones that I like or that are recommended by commenters here or on Reddit. For supplements, to avoid wasting a lot of time chasing after BS fads, I’m going to focus on those that are either reasonably supported in the academic literature or otherwise appear to have solid data backing them up. 

Lastly, it’s widely believed that the relative blood sugar impact of foods varies from person-to-person. If you’re interested in helping me to quantify that, let me know in the comments or send a via the contact form or to quantifieddiabetes_at_gmail.com. 

Testing Queue:

• Baseline:
• Glucose re-test:
• Low-carb foods:
• Ketochow: Reported
• Carbquick: Complete
• Tortilla: Complete
• Bread: In progress
• Snack bars: In queue
• Ice cream: In queue
• Supplements:
• Vinegar: In queue
• MSG: Reported

Week 1: Tortillas

Since this is going to be an on-going exploration, rather than wait for complete sets of data (which could take a long time), I’m going to post each weeks worth of data as I collect it in the hopes of soliciting feedback to guide later experiments.

As always, if you have any comments, suggestions, ideas for new experiments, or want to participate, please let me know in the comments or send a PM via the contact form or to quantifieddiabetes_at_gmail.com.

Ingredient Background

When I went to the supermarket recently, I noticed that they’ve started carrying low-carb tortillas. The macros looked decent (2-5g net carbs/tortilla depending on brand & type), so I wanted to give them a try. 

Procedure

At lunchtime (11a-12p), I consumed the specified quantity of food. No other food, drink, or medication was consumed except for water (ad libitum, but always less than 500 mL). Note: I take a long-acting basal insulin (lantus, 2u/day).

Blood sugar was monitored for 5h using a Dexcom G6. Calibration was performed 15-30 min. before the start of each experiment.

Results

As shown in Figure 1, all of the low-carb tortillas show a steep rise for the first 1-1.5h, similar to glucose, followed by a leveling off with a peak 2-3h after eating. This profile is similar to resistant wheat starch, their main ingredient (listed as “modified wheat starch”, but same thing). 

Most interesting is the difference between brands. La Tortilla Factory tortillas have about half the effect of Mission and Nutri-Rica tortillas (~13 vs. 25% of glucose for peak BG/g & ~30 vs. 60% of glucose for iAuC). They are thinner and have a, to me, better taste and texture as well, so a significant overall win. 

From looking at the ingredient list, the main difference seems to be that La Tortilla Factory uses cellulose fiber instead of Resistant Wheat Starch. I haven’t tested cellulose fiber, but Resistant Wheat Starch has a significant impact on my blood sugar, so that’s presumably where the difference is coming from. I’ll get and test some cellulose fiber to confirm. 

Interim Thoughts and Next Steps

From this preliminary data, it looks like there are meaningful brand differences for similar products with similar net carbs, justifying these experiments. 

I looked for other low-carb tortillas that don’t use resistant wheat starch, but except for La Tortilla Factory, they all are either too high carb or high calorie (e.g. using almond flour) and I’d rather leave room for more fillings. 

Does anyone know any other good low-carb tortillas I should try?

Next, I’m going to do a more detailed analysis of the tortilla results while starting testing on breads, ice creams, and snack bars. 

As always, please let me know if you have any thoughts or suggestions.

– QD

Nutrition & Ingredient Information

La Tortilla Factory – Flour

La Tortilla Factory – Wheat

Mission – Wheat

Mission – Spinach

Mission – Tomato

Nutri-Rica