Tracking Blood Ketones: Behind the Scenes Data on the Ketogenic Diet
I’ve tried a lot of diets.
I first went vegetarian, then slow carb, then gluten-free, then Paleo. I even did a 28-day Chipotle diet, which is exactly as awesome as it sounds.
Eventually I found the Ketogenic diet.
For me, like for many people in our communities, this all started with a health concern.
I was born with a heart condition. It never impacted my life, but it was there, lingering.
When I was a junior in college, a few classmates and I were out enjoying late night pizza. Out of nowhere, one classmate suddenly jolted upright and fell off his stool. He died. I found out the next morning it was from a lingering heart condition, not too unlike my own.
I started to think about my health a lot more after that.
I read about nutrition and started exploring the confusing world of diets. As I learned more and as I became more involved in Quantified Self, I found myself wanting to quantify these diets.
That’s what drew me to Keto. It’s the most measurable diet.
Quick Summary of the Ketogenic Diet
Keto is a high-fat, very-low-carb diet, usually with 70% of calories coming from fat. The idea is to switch your body from using glucose as its primary energy to breaking down fats into ketones for energy.
You can measure the macros that you eat and you can measure the ketones in your urine, breath, and blood.
In 2013, I did my first experiment with the ketogenic diet.
In that experiment, I tracked everything I ate in MyFitnessPal and compared it to other data I was collecting. I found my energy increased, my sleep quality went up (according to my Zeo data), my cholesterol levels improved, and my food cravings went away.
However, I also found that measuring everything I ate was a pain, I didn’t really know when I was in ketosis or how deep, and I eventually had a friend’s birthday party at a sushi buffet and never recovered.
Fast forward to 2015. I go to Ecuador and get a terrible stomach bug. This is another whole topic (I talked about it here), but one interesting note from that data was that intermittent fasting improved my symptoms.
Since the Ketogenic diet and intermittent fasting often go together, this observation inspired me to give Keto another shot. After that stomach bug, I was eager to do anything that would improve my gut condition.
This time my experiment only lasted a week.
Even though I felt great during that week, I was finding it hard to get over the initial hump where you switch into ketosis and get used to it (the Keto-Adaptation period).
If you’ve ever tried going Keto, you know this feeling:
Every time I tried to start back up again, I failed.
That changed when I moved to Napa and started working with Todd White.
Todd has been Keto for 3 years, and with his guidance, I’ve now been Keto for 3 months.
While I focused on my macros in my first experiment, and I only tracked the number of cans of fish I ate in my 7-day second experiment (it was 16 for anyone wondering), this time I wanted to focus on what was happening inside my body. I wanted to collect blood ketone data and see what it showed me.
My Goal
- Find my optimal ketone level.
I bought a ketone meter and a bunch of ketone strips, and I started tracking.
My first round of measurements was in September. As you can below, the data was all over the place.
I quickly realized I needed to test more frequently and control more of the variables for this data to be useful.
Instead of testing sporadically and once a day, I started testing consistently every few hours so I could see changes throughout the day.
Here are 4 consecutive days. Immediately you can see a pattern.
Ketones were low when waking up, dropped after working out, rose after drinking Bulletproof Coffee, were highest before breaking my intermittent fast, and dropped after eating an afternoon snack and dinner.
Doing some research, I learned that these are all fairly common.
Blood ketones drop after high intensity and strength workouts. The exact reason for this isn’t fully understood, but it seems to be due to the body breaking down glycogen.
They rise after consuming MCTs, which the body can quickly convert into ketones.
And they drop after eating a meal, even a high fat meal. I’m guessing it’s because every meal has a little carb and protein.
To see if this pattern was regular, I graphed my average ketone reading at each hour of the day.
Sure enough, the same pattern was there. Drop after working out. Rise after Bulletproof coffee. Peak in afternoon. Drop after dinner.
My goal, though, was to see how these readings correlated with how I felt, so I also recorded a subjective “energy” score from 1–5 at the time of many of these readings.
Looking at them here, by hour of day, I can see I often felt best in the early afternoon. This is also when my ketones were highest.
Was my energy highest at that time because my ketones were highest, or do I just naturally feel good in the early afternoon regardless of ketones?
I don’t know. I would need more testing to answer that.
To look at this a little differently, I then took each energy rating from 1 to 5 and found its average ketone reading.
There’s a clear trend here. When I felt like a low “1”, my average ketones were 0.5. When I was a 3, ketones were about 1.5. And when I was 5, ketones were closer to 3.5.
This is a small sample size but it shows a possible correlation between higher ketones and feeling better.
Given this correlation, I thought I should be able to predict my ketone level before measuring. I tried this for a few days.
This graph shows the difference between my guesses and actual readings. Red bars are when I guessed too low and blue bars are when I guessed too high.
This shows I’m not very good at guessing my ketone levels, often being off by 0.5 or more. I think I’m still getting used to the feelings of ketosis. I’m curious to see if I will get better at guessing this in the next few months, especially as I continue to become more keto-adapted.
Next, I wanted to see how my blood ketones compared with Todd. I’ve been Keto for 3 months; he’s been Keto for 3 years. So we did an experiment where we did the same things, ate the same foods, and measured at the same times.
We started the morning at the same level, but his ketones quickly escalated much faster than mine and his followed a different shape.
This hints at the idea that we all don’t work the same.
I’m curious to have him do more experiments to see if that shape is his norm or if his graph would average out to look more like mine.
Here’s another small example: Ramzy, another coworker, and I tested our blood at 8am on back-to-back days. We both were at 2.1 on day 1. We then spent the entire day together, doing the same things and eating the same foods. And yet, the next morning, I was at 0.6 and he was at 4.8.
I have no idea why.
It seems there’s still a lot we don’t understand about ketones. Fortunately, that means there’s a lot more to learn!
Here’s what I’ve learned so far.
What I’ve Learned
- My ketones fluctuate throughout the day. They are influenced by many different variables, including exercise, Bulletproof Coffee and ketone supplements, food, sleep, and more.
- I still don’t know what’s optimal for me but higher seems better so far.
- Everyone’s ketone production is likely different. My readings were much different from Todd’s and Ramzy’s.
- Measuring keeps me accountable. In these three months, I’ve stuck with the diet even through a bachelor party (in Vegas!), a wedding, and plenty of travel. The act of measuring my ketones has kept me on course.
- Ketone strips are expen$$$ive! I’m doing intermittent fasting not because it helps with the diet, but because all my lunch money is going to ketone strips!
What’s Next?
- Ketone tracking while fasting. I want to see how ketone levels change throughout the day during a 3-day fast.
- Comparing my data to other people. I found the comparison between my data and Todd’s/Ramzy’s to be fascinating. I’m curious to explore that more. Who else has taken blood ketone readings multiple times throughout the day? If you know anyone, I’d love to compare.
- Finding cheaper blood testing strips. Currently working on this. I’ll let you know when I find a solution.
Video
Here’s my full talk about this experiment:
If you found this interesting, here are two places to go next:
- An Intro to the Ketogenic Diet (with Dom D’Agostino) — Eating Fat, Lifting Cows, and Preventing Seizures (13 min read, plenty of stick-figure drawings)
- A Surprising Thing Happened When I Stopped Eating For 3 Days (8 min read, explains why I care about fasting)
Thank You!!
