Discovering how to use continuous glucose monitoring to overcome high sugars and inflammation

This is the first post in my Misadventures in Midlife Metabolism series where I share my experiments in personal nutrition and diabetes prevention from my perspective as a doctor, scientist and patient. In this post, I cover: 

• Why I did this: tackling prediabetes and unanswered diet questions.
• How high blood sugar causes inflammation and impacts many health conditions.
• How you can use continuous glucose monitoring (CGM) to eat better and control your sugars.
• Useful food experiments and challenges- and my results!
• What I discovered from my two weeks of using the CGM.

Please note that this post is my personal perspective with the intention to educate and empower. This is not medical advice, and I do not endorse or receive funds from any device corporations.

Why I did this: tackling prediabetes and unanswered diet questions 

When I first started my primary care medical practice 20 years ago, I screened all my patients for diabetes. I found that a startling proportion of my older adult patients (nearly half!) had undiagnosed diabetes or prediabetes. Now, more than 1 in 3 people in the US over age 18 have prediabetes. Having prediabetes means that our blood sugars are higher than normal but not high enough to be considered diabetes. Prediabetes, like diabetes, is associated with higher risks of stroke and heart disease. 

The other surprising discovery was that many of my patients with prediabetes or diabetes were physically active Asian immigrants whose weight was OK, like me and my family. They didn’t fit stereotypes of what unhealthy eating looks like. They ate the same foods my family did: rice, noodles, fruit, vegetables, tofu, fish and other meat. But we all also ate increasing amounts of processed food after coming to the US, as part of the Standard American Diet, or SAD (a very apt acronym). So I was not shocked when I found out in my 40s that I also have prediabetes.  

As a doctor, I preached the gospel of whole foods, matching most diabetes prevention diet guidance: eat unprocessed whole grains, vegetables, fruit, nuts, proteins! Eat a rainbow every day! Limit your added sugars! Some of my patients dramatically told me that they would starve if they had to eat less rice, so I focused on getting them to replace refined carbohydrates (carbs) with less processed ones. It was like pulling teeth to get my patients to switch from white rice to brown rice, from jook to oatmeal, from fruit juice to whole fruit. I made these changes in my diet too. I wanted to walk my talk. 

Now I regret spending so much time on these whole grain and fruit swaps. Like my patients, I have been tracking my blood sugars with hemoglobin A1C tests (HgA1C), which approximates a person’s average daily blood glucose over 3 months. My HgA1C has slowly gotten worse over the past decade despite eating a healthy-ish diet aligned with national MyPlate guidelines. It would have likely been more beneficial to replace carbohydrates with healthy proteins and fats instead, as I discovered during my time using the CGM. 

Since then I’ve learned more about how much national nutrition recommendations (housed in the US Department of Agriculture) are heavily influenced by corporate food lobbies, including the sugar, grain and meat industries. As a result of these lobbies, nutrition guidelines since the 1980s have advised people to eat more carbs and less fat, a diet that has contributed to our epidemic of obesity and diabetes. Nutrition studies are also rife with bias and funding from diet and supplement companies. We have to think critically about what nutrition guidance we follow. 

OK, so eat less carbs! That’s not rocket science. Low-carb diets have been around for millenia, from ancient Greek athletes to modern-day Atkins, keto and paleo diets. But some of these diets are highly restrictive, with Atkins and keto diets limiting total carb intake to 20 grams per day. That’s limiting all carbs to half a slice of bread over an entire day! This is extra difficult for me because I eat mostly vegetarian and pescatarian. No doubt a super low-carb diet like this would reduce my blood sugars, but it would make my life difficult and maybe also mess up my cholesterol and have a bigger carbon footprint.

Enjoying delicious food, especially with loved ones, is one of my life’s greatest joys. I don’t want to be on a restrictive low-carb diet that I’m not likely to sustain. I love eating fruit and baking. I wanted to find a “middle-way” diet I can maintain for the rest of my life: fun, joyful, not too restrictive and still reduce my blood sugars and inflammation. I decided to experiment with continuous glucose monitoring (CGM) for 2 weeks to find a more individualized way to enjoy the foods I like and eat healthier. 

Specifically, the questions I wanted to explore were:

• How much of my favorite foods can I eat regularly without my sugars spiking above 140? Or 120? I wanted to test my favorite carbs: fruit, noodles, scallion pancakes, sushi, pastries and ice cream.
• How much does mixing in protein and fat to my carbs change my spikes?
• How much does the sequence of eating, such as eating protein and fat first, change my spikes? 
• How much does exercise change my spikes? 

I share what I learned through using the CGM in the discovery section below. 

How high blood sugar causes inflammation and impacts many health conditions

Plenty of existing research shows that high blood sugar triggers inflammatory responses in the body: 

• Immune cells that normally prevent cardiovascular damage instead cause damage with chronically elevated blood sugars.
• Immune system signals get triggered by high sugars and cause widespread immune dysfunction, promoting the development of autoimmune conditions and cancers. 
• The chronic inflammation from chronically high blood sugars results in a higher risk for severe infection outcomes, such as from Covid. 

Controlling my blood sugar is important to me not just because of my prediabetes, but also because I have an autoimmune condition and a strong family history of cardiovascular disease (strokes and heart attacks). In addition, the decline in estrogen during perimenopause leads to poorer blood sugar response and systemic inflammation. In a vicious cycle, high blood sugars trigger inflammation, which in turn triggers worse blood sugar control. It is now clear to me how important it is for my health and aging that I need to control my sugars. 

I’ve also learned that national nutrition recommendations (housed in the US Department of Agriculture) are influenced by corporate food lobbies, including the sugar, grain and meat industries. As a result of these lobbies, nutrition guidelines since the 1980s have advised people to eat more carbs and less fat, a diet that has contributed to our epidemic of pre/diabetes. Even people who have served on the US nutrition guideline committee, such as renowned food scientist Dr. Marion Nestle (who is unrelated to the company Nestlé), are critical of the national guidelines. 

Nutrition studies and medical journal articles, including this one on CGM, are also often paid for by a corporation or industry selling the foods and devices studied. I could have learned how to eat healthier from scientific studies such as the OmniHeart randomized trial studying higher protein and fat diets to reduce cardiovascular risk (good OmniHeart diet summary here), but even this decently-conducted study took donations from the nut industry. I am also skeptical of fad diets (Atkins, paleo, keto, fasting, etc.). I’ve seen many many patients and friends yo-yo on fad diets and fail because the diets are too restrictive and unsustainable. We have to think critically about what nutritional guidance we follow. 

So, given how much nutrition misinformation and bias is out there, what can we do to be healthier? I had spent  years following the standard advice – exercising daily and focusing on a whole foods diet, and yet my HgA1C did not budge. I decided to take a deep dive with my own data to figure out how to get my glucose under better control. I needed my own sugar results to knock sense into me: showing me precisely and personally how my carb intake leads directly to high sugars. I needed to see how my seemingly small bowl of noodles was still 2-3 times more than an OmniHeart serving of grains and resulted in a big spike in my sugars. And I want to develop lifelong healthy eating habits that are still joyful and delicious. Tall order? Yes… I’m optimistic we can figure it out. 

Using continuous glucose monitoring (CGM) to eat better and control your sugars

There is a good amount of scientific evidence supporting the use of continuous glucose monitoring (CGM) for people with type 1 and 2 diabetes. Some insurance companies will now cover CGM prescribed for people diagnosed with diabetes. As of this writing, I don’t know of any health insurance plans that cover CGM for people with prediabetes. Currently there is some but not enough data that shows that CGM can reduce HgA1C or prevent diabetes and other complications, and this makes some doctors understandably skeptical of its use for people with prediabetes. 

There are two direct-to-consumer over-the-counter CGM kits that have been approved and became more affordable this year: $49 for the Lingo sensor for 14 days and $99 for the Stelo sensor for 30 days. If you can pay for these kits and are motivated to use them and learn skills to understand your glucose data (and/or have a knowledgeable medical provider or friend or family member who can help you), I think you can use CGM to gain insight on how you metabolize carbohydrates in the food you eat and make effective changes.  

While I used the prescription-level first-generation Libre sensor CGM in the $294 Zoe kit (with $35 discount here), my partner purchased the Lingo sensor for a much cheaper, $49 do-it-yourself CGM experience. I took what I learned from the Zoe program and applied it to my partner’s DIY experience.  The sensors, applicators and application process were identical. Both store glucose data in 15-minute increments but can read minute-to-minute. The first-generation Libre sensor required me to hold my phone within a few inches of my sensor and sync to get readings (this was quite annoying); the Lingo app read sensor data automatically from several feet away.

Here’s how we made the most of the 2 weeks we had the sensors on:

• I determined my goals using the CGM sensor:
  1. To reduce the number of glucose spikes over 140 and dips under 70. Frequent and large spikes and dips are signs of insulin resistance and inflammation. Spikes and dips make me feel foggy, sleepy and then hangry. No bueno. Want to avoid that! 
  2. Find a yummy breakfast meal I can enjoy every day without raising my glucose >120.
  3. Find 2-3 yummy lunch meals I can eat each week without raising my glucose >120.
  4. Identify the amount and combos of my favorite carbs (fruit, noodles, scallion pancakes, sushi, pastries and ice cream) I can eat in a meal without spiking >140. 
  5. To find foods that help me stay satisfied and less hungry so I snack less, especially at night, so I can keep my eating window within 12 hours each day to give my body a break from insulin, glucose and food processing. 
• I made a plan for the foods and scenarios I wanted to test during my 2 weeks with the sensor. 
• I made a paper document so I could jot down notes on what I ate and my CGM results. You can also create a document or sheet on your phone or other device. This is important so you do not rely completely on proprietary apps which use or remove your data (like the Zoe app).  The most useful data points I recorded were:
  • Food I ate and quantity/amount 
  • Highest glucose reading after I ate and before the next meal (peak/spike)
  • Optional and maybe helpful/interesting data points:
    • Date and time of eating, peaks and dips
    • Dips below 70
    • Exercise: type and timing
    • Sequence of eating during a meal: order of carbs, proteins, fats 
    • Combos: how much proteins and fats were mixed in with carbs
• First 3 days: I started with my current diet and usual meals: my typical daily breakfast, common lunches, dinners and snacks I often eat. 
• Next 5 days: I ran breakfast experiments to test different proportions, sequences, combinations and exercise. See experiment details below. 
• Last 6 days: I used what I learned so far to refine and test more meals. I LOVE noodles, so I tested my glucose response to different noodles in similar quantities. 
• I made a note of the glucose peaks and time after each meal and assessed whether they were:
  • Optimal glucose range: 70-120
  • Normal glucose range: 70-140
  • High but within diabetic target glucose range: 141-180

Useful food experiments and challenges

Below are the breakfast experiments I did to test how I change my glucose response to food I typically eat in different proportions, order, combinations and with exercise. Zoe has a helpful library of more “blood sugar challenges” you can freely access here and here.  

My breakfast experiments: 

• Carb only: Steel-cut oats (cooked) alone, 180 grams
• Protein/fat only: Peanut butter alone, 50 grams
• Mix: peanut butter 50 grams mixed with oats 180 grams
• Carb first: Eat oats 180 grams first, wait 10 minutes, then eat peanut butter 50 grams
• Carb and exercise: Eat oats 180 grams, then take a brisk walk for 30 minutes

Based on my results and what I learned, I tweaked my daily breakfast and tested it for a few days to see if my glucose reliably stayed under 120 until the next meal.  

• My old breakfast: egg, fruit 185g, steel-cut oats 90g + peanut butter 30g + ground sesame seeds 5g + ground sunflower seeds 5g + unsweetened soymilk 60g. This led to glucose spikes around 130. 
• My new  breakfast: egg, steel-cut oats 50g + peanut butter 30g + my “super seeds mix” 20g (see below) + chia seeds 5g + unsweetened soymilk 80g, fruit 80g. The net carbs of this meal is 31 grams. My glucose stays <120 after, and it keeps me satisfied for 5+ hours.
  • My new super seeds mix includes 4 seeds, which I grind once a month by pulsing them in a food processor in equal proportions. I store the mix in a jar to scoop out each morning. Each 20 gram serving of super seeds mix consists of 5 grams each of ground sesame seeds, sunflower seeds, pumpkin seeds and flax seeds. 
  • Chia seed needs to be soaked in water/liquid to be properly digested. Soaking also makes chia more like tapioca- I like the chia seed texture and taste so much better after soaking. Once a week I use a 1-cup glass jar and put in 40 grams of chia seed, fill the rest of the jar up with unsweetened soy milk, mix it together and refrigerate it overnight. It’s ready to eat by the next morning.
  • You can make this completely grain-free and reduce carbs further by omitting the oatmeal, which I now often do. For the grain-free version, I increase the amount of soaked chia seeds to pull the seeds together and give the mix a cohesive heft. 

Here are my breakfast experiment results: 

• Green bars are post-breakfast glucose results that stayed under 120 until my next meal. These are primarily the breakfasts in which I ate only peanut butter or limited my oats/fruits and mixed in more seeds and nuts. 
• Blue bars are post-breakfast glucose results that were 120-140 until my next meal. These are primarily the breakfasts in which I ate more than 120 grams of oats or fruit first or by themselves. 
• Orange bars are post-breakfast glucose results that spiked to 140 and above. These are primarily the breakfasts in which I ate 180 grams or more of oats or fruit first or by themselves. Even a brisk walk after eating these meals was not enough to keep the spike under 140.

What I discovered from my two weeks of using the CGM 

The following are the most important insights that I gained from using the CGM for two weeks. Since each of us metabolizes in unique ways, your carb and fat limits and response to combos, sequences and exercise are likely to be different from mine. While there are some more universal findings (limit starchy and sugary carbs, mix in proteins and fats, spikes and dips suck), the specifics of carb limits and response to certain foods are individual. That’s the beauty of doing the CGM for yourself: you get to see your own unique glucose metabolism in action with the unique set of foods you like to eat! 

• How much of my favorite foods can I eat regularly without my sugars spiking above 140? Or 120?
  • I discovered that I still ate way too much carbs! 
  • Whole food carbs, such as steel-cut oats and whole fruit sadly also led to spikes. 
  • If I limited my grains to <60 grams and whole fruit to <80 grams per meal and otherwise ate non-starchy vegetables, proteins and fats, I kept my glucose under 120. 
  • That’s way more than the low-carb diet limits of 20 grams of carbs per entire day, but still way less than what I had been eating. This is totally a do-able compromise for me. 
  • My partner and I had different glucose responses to the same foods. I suspect that each person’s limits are unique to their own metabolic profile. Figuring out our own individual limits helps us not be overly restricted with a cookie-cutter low-carb diet.  
• How much does mixing in protein and fat to my carbs change my spikes?
  • Mixing seeds and nuts into my meals reduced spikes and kept me satisfied for longer.  
  • Now I put way more nuts and seeds into my breakfast oats and lunch salads. Yummy! 
• How much does the sequence of eating, such as eating protein and fat first, change my spikes?
  • Starting a meal with protein and fat instead of carbs helped reduce my spikes. 
• How much does exercise change my spikes?
  • Exercise (brisk walking, hiking) maybe reduced my spike, but mostly I found that after a carb-heavy meal, exercise delayed a spike until I was resting. 
• I also discovered that night-time snacking led to overnight hypoglycemic dips. Time to stop that!
• I’m also developing my own “internal CGM” by paying close attention to how I’m feeling. This ability to sense what’s happening in our bodies is called “interoception.” While using the CGM, I found that during spikes above 140 I felt foggy, and during the dips after, I felt tired and hangry. No wonder I feel strangely hungry after eating a huge carby meal: I learned that I don’t need to eat more food… I’m actually experiencing a glucose dip after a high-insulin response. I now have a better sense of when I have spikes and dips after eating particular foods without needing to wear a CGM all the time. If you have already developed this sense, you may be able to determine your own limits without using a CGM. Cool, huh? 

None of my discoveries are shocking, but there were a few surprises, especially around foods I ate to supposedly be “extra healthy.” I was surprised to find that steel-cut oats, beautiful organic whole non-tropical fruit and beans led to spikes. I was surprised at how much I needed to reduce the quantity of grains and fruit I ate, not realizing how little 60 grams was until I weighed everything out. I was surprised that exercise mostly delayed my spikes instead of reducing them in a sustained way. Happily, I also learned that I can still eat the grains and fruit I love (and less is still very enjoyable)… and that adding lots of seeds and nuts (more protein and fat) is quite tasty and satisfying. 

Using the CGM for two weeks helped me gain these insights and make sustainable changes to how I eat on a daily basis. The immediate glucose results from testing these changes make me hopeful that the next time I test my 3-month glucose average (hemoglobin A1C), I will finally see some real progress and reduction after all these years of trying.