This is Part Two of the three-part series on glucose and metabolic health. You can find Part One here.
Perhaps you have seen people wearing continuous glucose monitors (CGM), and you were interested to understand what they were. Or, you may have seen advertisements for products and services that claim to improve your metabolic health through the use of continuous glucose monitoring (CGM) technology. Current nutrition guidance and dietary requirements are published based on research averages, but none of us is average. Let's look at how glucose response is measured today and then explore the emerging research and technology that point to a future where a more individualized approach to improving glycemic response is possible.
How is glucose response evaluated?
Blood glucose is typically evaluated by measuring fasting blood glucose. Unfortunately, this method doesn't measure any post-meal glucose levels, even though most people spend most of the day in a post-meal state.
For those with known or suspected disordered glucose control, HbA1c is also measured. HbA1c calculates how much sugar has accumulated on a hemoglobin (Hb) blood protein. Sugars are more likely to bind to the Hb when blood sugar levels are high. Based on the turnover rate of blood cells, HbA1c reflects average accumulation over three months. While both fasting blood glucose and HbA1c do correlate with likely long-term disease outcomes, they don't provide insight an individual can use to take action to improve their glucose response.
The future of glucose control begins with continuous glucose monitoring-CGM technology. So, what is it?
CGMs were originally introduced in the early 2000s to help people with type 1 diabetes. People with this autoimmune disease can no longer produce insulin. CGMs helped people better understand their glucose levels so they could administer insulin more accurately. The earliest versions were worn for only a few days and required that the data be shared with the physician before being released to patients. Current versions, which allow users to see their data in real-time, were approved for use in the US around 2015. At this time, studies and therapeutic usage of CGMs for prediabetics, type 2 diabetics and even outwardly healthy patients began to emerge.
CGMs are wearable medical devices that continuously measure glucose levels, providing regular glucose data to the patient. The current devices are typically attached to the upper arm or lower abdomen and worn consistently for 10–14 days, depending on the device. They sample interstitial fluid, a clear fluid that surrounds the body's cells and is the conduit for nutrients like glucose between the blood and the cells. Glucose in the interstitial fluid reflects the levels in the blood, though the timing is slightly delayed.
CGM technology opens a new window into discoveries for metabolic disorders because it provides frequent insight into what glucose levels do throughout the day under varying circumstances and for different people. It has the potential to help us understand how lifestyle choices such as nutritional and eating patterns, exercise, sleep and stress affect long-term metabolic health. Although the technology is relatively new, it has already begun to provide significant new insight.
Key findings from emerging research Currently, glycemic response is predicted to be the same for all people and is based on the concepts of glycemic index and glycemic load, which are rooted in the carbohydrate content of foods. The glycemic index is a measure of how quickly a carbohydrate-containing food is digested and results in glucose in the bloodstream (based on a 50-gram serving of carbohydrates). Each food has its own glycemic index, and it is applied equally for everyone. Glycemic load takes the glycemic index and modifies it to reflect an appropriate serving size for a particular food. Emerging research using CGM technology has shown that the one-size-fits-all approach is not the best predictor of glucose response, and more individualized methods are possible. Some key early findings are discussed below. Individuals can have vastly different glycemic responses to the same foods or meals One study in Israel with 800 subjects found that, for the same meal of bread, the average rise in glucose level was 44 mg/dl (milligrams of glucose per deciliter of blood) but that it varied +/- 33 mg/dl—with the top 10% increasing more than 79 mg/dl and the lowest 10% rising less than 15 mg/dl. It also found highly interpersonal responses to different foods. For example, one person had almost no glucose rise in response to consuming a cookie but a roughly 15 mg/dl rise after eating a banana. A different person had almost no rise in response to a banana but an approximately 25mg/dl rise after eating a cookie, as seen in Figure 1 below.
Carbohydrates tend to elicit a lower rise in glucose earlier in the day
Recent studies have found that, for most people, having carbohydrates earlier in the day causes a lower rise in glucose than when the same meal occurs later in the day. A study conducted in the UK with 1,002 healthy adults found that when the same meal was eaten for lunch versus breakfast, it elicited an almost two-fold increase in glycemic response.
Meal composition affects the glycemic response
Meal composition refers to the make-up of a particular meal in terms of proteins, fats, carbohydrates and fiber. Studies have found that meals containing the same number of carbohydrates but different amounts of protein, fat and fiber may cause differing glycemic responses. One such study of 24 healthy adults used four standardized meals with 50g of carbohydrates to test the impact of different meal compositions. One meal, called meal 4, with lower fat, fiber and protein, had an average glucose rise of 54.8mg/dl, while another meal, meal 3, with higher protein, fiber and fat, had an average increase of just 20.2 mg/dl, as shown in graphs below. Although these meals demonstrate an overall standard pattern of glucose rise based on meal attributes, the individual rises varied over a wide range, as shown in the lighter lines above and below the average line.
Models built using biometric data, including microbiome composition, have the potential to better predict individual glucose response to a meal than the traditional methods of using carbohydrate content alone (Note: Biometric data is any measurable biological information unique to an individual, from your height, weight or lipid levels to the makeup of your microbiome.) The two large-scale studies mentioned above in the UK and Israel created complex mathematical models to test if individual biometric factors such as microbiome, age, current glycemic biomarkers (e.g., HbA1c and fasting glucose) and meal context factors (e.g., meal composition, timing, exercise, sleep and previous meals) could predict an individual's response to future meals. Both studies found that their models' predictions correlated more accurately with actual glycemic response than models built on carbohydrate content alone. This research indicates that the one-size-fits-all nutritional guidelines and standard requirements are not predictive of individual glucose response, and it opens up the possibility that biometric-based personalized nutrition is on the way. How can you begin to understand your unique glycemic response? Try a CGM and experiment! For those with prediabetes or diabetes: CGMs will likely be covered by insurance with a doctor's prescription. CGMs such as Abbott's Freestyle Libre or Dexcom G6 come with phone-based applications that allow you to see your real-time response to foods and track trends. Experiment with the lifestyle modifications that will be discussed in next week's post in this series and see what works for you. For those without diabetes or those who want more structured assistance in learning from a CGM: In the last 18 months, various companies have begun offering unique phone-based applications that pair CGM technology and user-input food-intake data with other health data available—such as exercise, heart rate and sleep metrics—to provide the user with a more integrated understanding of their glycemic response. Some applications also include access to online dietitians. A few companies in this emerging market are Levels Health (in controlled release) and NutriSense. Other companies such as Zoe Global, DayTwo and January AI include predicted meal response tools based on machine learning models. Even if you don't use a CGM, you can improve your glucose response through diet and lifestyle modifications. Our final post in this series will provide concrete actions that you can take to help you achieve better glucose control.
Resources
Berry SE, Valdes AM, Drew DA, et al. Human postprandial responses to food and potential for precision nutrition. Nat Med 2020 266. 2020;26(6):964-973. doi:10.1038/s41591-020-0934-0
Freckmann G, Hagenlocher S, Baumstark A, et al. Continuous Glucose Profiles in Healthy Subjects under Everyday Life Conditions and after Different Meals. Journal of Diabetes Science and Technology. 2007;1(5):695-703. doi:10.1177/193229680700100513
Zeevi D, Korem T, Zmora N, et al. Personalized Nutrition by Prediction of Glycemic Responses Article Personalized Nutrition by Prediction of Glycemic Responses. Cell. 2015;163:1079-1094. doi:10.1016/j.cell.2015.11.001
Jill Goldring, MNSP, MSIE
Jill Goldring is a nutritionist, engineer, avid gardener, beekeeper and healthy food enthusiast. Nutrition is a second career for Jill after a successful Silicon Valley career managing high-tech projects. Jill is interested in the intersection of diet, glycemic response, microbiome and metabolic health outcomes. Her goal is to make positive changes to food systems and services that provide healthy food, nutrition education and nutrition technology to underserved communities. She is a volunteer with the Samaritan House San Mateo. She manages a pilot program that pairs nutrition education focused on glycemic response and CGM (continuous glucose monitor) technology for the free medical clinic’s food pharmacy for diabetic patients. She has Master’s in Nutrition Science and Policy from Tufts University Friedman School of Nutrition Science and Policy, a BS in Industrial and Systems Engineering from USC and an MS in Industrial Engineering from Stanford University. guthappynutrition@gmail.com Instagram: @guthappynutrition