Background
I know in prior writings I mentioned how I’ve been tracking my health less and living more spontaneously by intuition. Wearing trackers and doing frequent blood tests for over a decade has given me a strong sense of when my metabolism and energy are off, so now I just mostly pay attention to a few intermittent metrics.
Now having said that, I have recently decided to gear up again not because my health is slipping, but instead because I realize that my insights into my own data help me educate my patients better. I know each individual is different, but just like we can learn by studying the physiology of elite athletes and apply those principles to how chronic disease develops, I consider myself a more realistic model that’s closer to my typical patient’s phenotype.
I’ve had insulin resistance before and can easily slip into that pattern with dietary excursions and periods of inactivity. I’m much more metabolically flexible than I was a decade ago, but the fact that I can revert back to my prior state gives me insights and empathy when I see my patient’s metabolic numbers slide.
With that context in place, I’m going to share some of my recent data below to make some high level points. If you are new to my blog, I’ll include references to prior posts and resources to get you up to speed.
OGTT or Insulin Resistance Training Test
I’ll refer to the oral glucose tolerance test (with insulin) here by its abbreviation OGTT. I often refer to the OGTT as an insulin resistance training test. This is similar to me evaluating your posture and performance doing a lifting exercise like a deadlift (add picture below). Let’s say I chose a load of 200 pounds and asked anyone who came to my clinic to deadlift that weight.
There would be a variety of different responses to lifting that 200 pound load which I’ve boxed into 3 categories below:
- Completely load resistant: Individuals who can’t even budge the weight.
- Partially load resistant: Individuals who can lift the weight with effort, but use poor form that puts their muscles and joints at risk of injury
- Load resilient: Individuals who effortlessly lift the load with perfect form without straining their muscles and joints
The OGTT is a similar type of test that puts your metabolism instead of your muscles under a fixed nutrient load, which in this case is 75 grams of a glucose drink. Every patient who gets this test, regardless of their level of metabolic fitness, gets this 75 gram load, just like the 200 pound weight in my hypothetical scenario for musculoskeletal load. Below is the drink I had for my own OGTT which we’ll discuss in a moment.
In order to assess how well an individual handles this metabolic load, we would need to assess the glucose response to the load, and in this case I also measured my insulin levels concomitantly. A standard OGTT, like women receive to screen for gestational diabetes, just measures glucose alone.
This is where my weight lifting and glucose analogies nicely overlap. At a macro level we depend on our gross muscular strength and function to lift the 200 pound load without injury. At a molecular level we also depend on these same muscles to put away the 75 gram glucose load efficiently.
Our muscles dispose of roughly 80 percent of the glucose we eat and drink. If they don’t do this well, the glucose gets routed to fat (obesity) or liver (fatty liver, high triglycerides, elevated blood glucose).
Insulin specifically is the hormone that does the heavy lifting here, as it binds to receptors on the muscle’s surface to allow the glucose from our meals and from this 75g glucose drink to enter our muscles.
See the short video excerpt (video only without audio) below from my Whole Family Health program where you can see how the glucose car enters the muscle parking lot using the insulin parking pass, representing how muscles clear glucose from the bloodstream.
By the way, with the 200 pound load, if we don’t engage our large muscles and core properly, that load gets disproportionately transmitted to joints, tendons, and ligaments leading to a musculoskeletal strain. Keep doing this repetitively and you end up with an injury. For glucose, the situation is similar.
A single 75 gram drink of glucose won’t trigger acute disease, but repeated glucose loads at even lower doses in someone who lacks metabolic (and aerobic) fitness can cause long-term injury in the form of diabetes, heart disease, cancer, Alzheimer’s disease, etc.
Back to my weightlifting categories mentioned previously, “load resistance” refers to insulin resistance, where you cannot remove glucose effectively with a normal amount of insulin. “Load resilience” is insulin resilience or what we commonly call insulin sensitivity, where you can easily remove a glucose load with a normal amount of insulin.
One key point I’d like to make before we delve into my own results is that insulin is the hormone that is doing the heavy lifting when we encounter a glucose load, but too often it’s an unsung or overlooked hero. I see patients who are surprised at how quickly their diabetes has progressed despite relatively normal appearing glucose and A1C levels for years. How did diabetes all of a sudden rear its ugly head?
The answer is that insulin was silently doing the heavy lifting in the background for years, often decades, to keep glucose numbers looking normal. When I eat a high carb and/or sugar meal or snack and I see a relatively normal glucose level afterwards, I don’t pat myself on the back and say “I’m immune to diabetes!” Instead, I realize that I stressed my pancreas to overproduce insulin to keep my glucose within range, and that if I repeat that task regularly like so many of my patients do, at some point my pancreas will say “I give up,” and underproduce insulin, leading to high glucose levels.
I’m appreciative that insulin is still able to do the heavy lifting every now and then when I do decide to indulge, but I immediately look for opportunities to give insulin a break through proper nutrition and exercise.
For many of us, insulin is like an unhappy laborer who is silently burning out from being overworked day after day, meal after meal, snack after snack.
One day insulin defiantly and rightfully declares… “Enough already! I’m tired of doing the daily dirty work of atoning for your dietary sins, which include even excess amounts of healthier carbs (whole grains, brown rice, oats, etc.) ,and stashing away all that glucose so your blood sugar levels look good….I QUIT!”
Your elevated glucose and/or A1C result may appear like an abrupt and sudden change, but insulin has been overworked for decades without you even knowing it. If you’ve been getting by with normal glucose test results despite eating excessive glucose-containing foods, you can assume that you are overworking insulin and at some point you will likely reach a breaking point that can manifest as a chronic disease like diabetes or some other insulin resistant disorder.
By the way, this breaking point is occurring already in young children and teenagers who are leading excessively sedentary lives and consuming high sugar foods and beverages, leading to obesity, fatty liver, type 2 diabetes, PCOS, etc. Insulin is burning out earlier and earlier in life because we are making it lift loads it has never before experienced during the history of mankind.
My OGTT Results
So now that you understand the concept of the OGTT and the key role of insulin, let’s see how my metabolism performed under a 75 gram load. This test was done by first measuring my fasting blood glucose and insulin, then the lab tech immediately had me chug the 75 gram glucose drink in less than 5 minutes, and then I returned to the lab 1 hour and 2 hours later for repeat glucose and insulin measurements.
The table below has my OGTT results and I also layered on my CGM (continuous glucose monitor) data from my Abbott Freestyle Libre sensor which provided some additional glucose data points to paint a more detailed picture.
My fasting insulin at 8a is nice and low with blood glucose levels that are reasonable. I often wake up in the 80s and think sleep had something to do with this higher than normal value (see later). After the glucose load you can see that as expected, my glucose levels from my CGM data started rising rapidly to a value of 148 within 30 minutes.
At 9a (1hr after glucose load), I returned to the lab for another blood test. You can see that my serum blood glucose already returned back to below my fasting level at 86. The CGM data shows a higher value of 114. There is expected discordance between blood glucose and CGM results given the CGM measures glucose in the interstitial fluid, so we’ll have to trust the serum blood glucose as being more accurate.
The 1 hour insulin level of 65 is well within range by normal standards, but I would have expected it to be a bit lower. The goal is to clear glucose efficiently with the least amount of insulin. One of the reasons my insulin on this test was on the higher side is that I have been relatively low carb in the days leading up to the test. I knew this might interfere with my numbers a bit.
If you are low carb or in ketosis leading up to an OGTT, a glucose load may cause an exaggerated glucose (and insulin response) because your metabolism is adapted to using fat for energy and is not used to seeing such a high carb load. I can’t recall the last time I had a drink sweetened with 75 grams of sugar, so this is analogous to asking me to deadlift a weight that I’m not accustomed to lifting. I’m going to have to recruit more muscle (i.e.-insulin) to lift this 75g load.
Fortunately by the 2 hour blood draw at 10a, my insulin dropped down nicely to 7 and glucose returned to lower than my fasting value. Although I had a bit of a transient insulin spike by my own rigorous standards, the “insulin lift” didn’t last long enough to be of consequence.
The overall results of my OGTT showed that I was able to efficiently “lift” or clear a significant glucose load within a relatively short period of time, using a relatively low amount of insulin. If you monitor your glucose levels, the goal is the same. Minimal spikes in glucose and rapid return to baseline levels. In my insulin resistant and metabolically unhealthy patients, their glucose spikes are far too high and they linger for hours.
Below are a few other factors that can influence glucose and insulin levels. I exercise most mornings and on the morning of this test I didn’t have time to do my usual longer regimen. I instead did a truncated 35 min jog/walk in my work clothes before my test with an average HR of 111 bpm. Below is my Apple watch data.
My sleep was ok but not great. The night before my OGTT I knew I had this blood test followed by a lot of meetings, so my mind was preoccupied during the night. Shorter duration and/or poorer quality sleep can wreak havoc with glucose control.
Glycemic Variability (GV)
Another important insight from this experiment is the low glucose that took place 3 hours after my glucose drink as I was standing at my desk. If I had decided to go for a brisk walk or exercise after that high glycemic meal, it may have brought my glucose crashing down low enough to cause intense hunger and fatigue. Fluctuations in glucose levels throughout the day are referred to as glycemic variability, which I’ll abbreviate as GV.
I want to share another case of extreme GV in real life with you. I went to visit my mother during Thanksgiving and decided to track the impact of some of my activities on my CGM data. I had a relatively high carb lunch followed by a delicious and decadent dessert dropped off by a neighbor. I know this neighbor reads my blog posts so if she is reading this one, I wanted to thank her for the homemade dessert since it was not only delicious, but allowed me to perform my very own glucose tolerance test using something far more palatable than a glucose drink.
You can see that my glucose started rising to 134 mg/dL with an upward arrow next to the reading in the image below, meaning it was still heading up. I panicked and decided to go for a quick jog/power walk. As you can see, my glucose started dropping immediately which is great, but by 1:18p it started moving towards hypoglycemia. So what happened?
As we discussed from the OGTT results, my insulin does a great job of clearing out extra glucose. When I ate the high glucose load meal and dessert, my pancreas pulsed out a strong dose of insulin and in addition to that, I went to exercise which helps further push glucose into muscle.
In other words, I was a bit too insulin sensitive, and that led to an over clearance of glucose into my tissues, particularly muscle, leaving a lower amount of glucose in the blood. There are other causes for hypoglycemia after meals, even without exercise, like reactive hypoglycemia, hormonal insufficiencies (thyroid, cortisol), and rapid gastric (stomach) emptying which we can explore further in a future post.
A larger point here from my OGTT and the high carb+dessert experience, is that we want to avoid abrupt and frequent swings in glucose levels.
In my OGTT, I had an almost 60 point drop according to my CGM in a 1 hour period from 8:30a to 9:30a. With my real life high carb lunch + dessert experiment followed by exercise, in a period of 30 minutes, my glucose dropped 72 points. If you were to look at a graph of these trends over the course of a day, you would see wide fluctuations between highs and lows, despite a daily average glucose that may look normal.
These swings are a marker of high GV which is correlated with higher risk of diabetes, heart disease, inflammation, and a whole host of different medical conditions and symptoms which you can review further in this study.
Studies show that in healthy, non-diabetic individuals, glucose spikes rarely rise above 140 mg/dL throughout the day and are even less than 120 mg/dL. Glucose recovery back to normal usually occurs by 30-60 minutes. Type 2 diabetics may take 2-3 hours or more for glucose to recover back to normal.
My CGM glucose average is typically 85-90 mg/dL and it stays within a preset target range between 70-120 mg/dL 100% of the time. One of the most powerful metrics to measure when wearing a CGM, is to notice if your curve is relatively flat or wildly swerving up and down throughout the day. I personally don’t like to see swings of more than 20 mg/dL in either direction on a typical day. To learn more about CGMs, read my prior post here.
To use an overused pandemic term, given this is being written during the Covid-19 pandemic, my goal personally and in my patients is to “flatten the curve.” Referring to the image below, we’d want our GV to move from the left (wide swings in glucose amplitude) to the right (minimal swings in glucose amplitude). Flatten the glucose curve, which in turn means flattening insulin deviations, which results in optimal metabolic function, and more consistent energy and mood. 
Another key lesson from all of this is for individuals who think they can eat whatever they want since they are exercising. If you are playing the game of eating high carb/sugar meals and then “gaming” your numbers back down through fasting and exercise, you are just trying to cheat your way to a better average glucose number and causing wild fluctuations in glucose that as we mentioned raise the risk of chronic disease.
Your GV is placing a huge stressor on your cells, your mitochondria, your adrenal glands, your thyroid, your heart, your brain, etc. and other than occasionally using this as a bandaid for occasional social and holiday eating, this is not a sustainable daily or even weekend strategy for most.
I’ve also noticed a common nighttime pattern where my patients eat a very starchy dinner, then exercise intensely, and wonder why they can’t sleep at night. They are likely mentally activated by their workout, but they also potentially are experiencing a drop in their glucose at bedtime, causing the stress of high glucose variability and subsequent nocturnal arousal during the night.
My “Elevated” A1C
One test I left out of these results was my A1C test (aka Hemoglobin A1C). This is a blood test that measures your average glucose for the last 2-3 months. The estimate is based on the principle that hemoglobin, the protein in your blood cells that carries oxygen, also attaches to glucose. The higher your blood glucose, the more of it sticks to your hemoglobin and the combination of the hemoglobin linked to your glucose is what we call the glycosylated hemoglobin (aka hemoglobin A1C, HbA1C, or just A1C).
Since hemoglobin has an average lifespan of 120 days, your glycosylated hemoglobin roughly reflects your glucose status for the past 2-3 months. The key variable here is the lifespan of red blood cells. Some individuals have longer red blood cell lifespan which can lead to a falsely elevated A1C, while others have shorter red blood cell lifespan causing a falsely lower number.
In my case, my A1C over the years, has rarely correlated with my CGM or frequent finger stick glucose data which I used to chart before the days of CGM. My numbers have ranged between the mid-5s to low 6s even though my average glucose runs between 90-100 mg/dL. Using the table below or an online calculator like the one found here, you’ll see that my average glucose should correlate to an A1C between 5.0-5.1%.
Despite having an average glucose of 95 for the last 3 months, an optimal glucose/insulin tolerance test, and ideal lipid ratios, my A1C returned with a value of 5.9%, which would be defined as prediabetes! According to the conversion table, that would mean my average glucose value throughout the day is 123 mg/dL which I rarely ever hit.
In fact, about 2 years ago I was on a ketogenic diet for 6 months with average daily glucose levels in the low to mid 80s, which means my estimated A1C should have been 4.6, but my blood test showed an A1C of 5.6 instead!
I used to get anxious and flustered about my A1C result in the past, but thanks to the CGM and years of charting my glucose, I realize the A1C does not correlate with my average glucose or my level of insulin resistance. I have other patients who have the reverse issue. Great looking A1Cs, but multiple other markers of insulin resistance like excess belly fat, high triglyceride-to-HDL ratios, and often high average glucose levels on their CGM or normal averages but wide fluctuations.
Recall that you can “game your average” by counteracting high glucose levels (eating binges) with low ones (periods of fasting and skipping meals), but these wide swings are a risk factor for heart disease, future diabetes, and multiple other chronic health issues.
I’m not discounting the A1C as a useful tool. For most of my patients it does correlate, at least directionally, with their degree of insulin resistance. It’s a cheap and accessible test that doesn’t require arm twisting your doctor to have them order it. It’s also a key marker for monitoring diabetes and there are decades of studies that correlate disease risk to A1C levels.
I’m just highlighting here that there are a significant number of cases like mine, where it doesn’t correlate well to glucose and insulin resistance. Insulin resistance is not a single metric condition. There are multiple other variables that need to be assessed other than glucose and A1C, like waistline, lipids, blood pressure, etc.
The Key to My Own Insulin Fitness
For me personally, achieving insulin resilience involved exercising at the right intensity and frequency since I had my diet dialed in already. Many of my patients keep playing the game of lowering their carbs further and fasting longer in an effort to avoid the exercise time necessary to achieve metabolic flexibility and minimal glucose variance. They often eventually become insulin resistant and may also end up with nutrient deficiencies and a slower metabolism in the long run if they continue this diet-focused pattern without sufficient doses of exercise.
A lack of aerobic fitness eventually catches up to you regardless of diet and manifests as a mood, musculoskeletal, and/or metabolic disorder like insulin resistance.
Excessive high intensity exercise is not the answer and I rarely see a single new patient in my clinic who claims they exercise, actually performing activities at the right intensity and dosage to prevent and reverse insulin resistance. Please refer to my prior exercise posts like the one here so you dial in the right dosage of aerobic exercise. The CGM has personally been more of an activity and exercise tool for me rather than a dietary one, since I see the significant impact it has on flattening my own glucose curve.
Summary
We covered quite a bit of important information in this post. In a nutshell, one of your most important health goals is to strategically incorporate lifestyle strategies like nutrition, exercise, and sleep, in ways that will allow your body to regulate glucose (and insulin) with the least amount of variance.
I don’t order OGTTs in most of my patients and I frankly didn’t necessarily need to do one myself, but my motivation was to try to approach the concept of insulin resistance and metabolic resilience from a different vantage point so you can view a collection of data points that provide a more dynamic view of metabolism, versus the sporadic blood tests most of us get as part of our yearly check-ups.
I know many of you who read my posts are diabetics and please know that I didn’t present my data here in an effort to “glucose shame” you. If you follow your glucose using a CGM or a finger stick device, I realize your peak numbers and slow glucose recovery times are a source of daily stress and anxiety. I was in a similar situation over a decade ago when I used to track my finger stick glucoses closely, and had done a glucose tolerance test over 15 years ago when I practiced in Southern California.
I didn’t access those older results for comparison, but know from recall that my results were not as good as they are today. Despite insulin resistance usually getting worse with aging, my lifestyle changes have allowed me to be far less insulin resistant today than I was a decade and a half ago.
Many longevity experts feel insulin resistance and our ability to efficiently clear glucose from the bloodstream are a major marker for physiological aging, and I completely agree.
From this viewpoint, my results and those of my patients and readers who have also implemented proper lifestyle changes to reverse insulin resistance, have literally turned back the hands of time on aging.
If you step back and look at your data closely and take gradual steps in making meaningful dietary and activity changes, you will slowly or rapidly start seeing improvements in your glucose data patterns. These shifts towards less GV and lower peak glucose levels should serve as golden nuggets of motivation that keep nudging you on the right path back to optimal health.
The good news is that tools like CGMs will continue to become more accessible, so that one day having a real time view of health, rather than intermittent snapshot labs, will become the standard upon which we make decisions around lifestyle and medical management.
