There I was in my general physician’s examination room trying to explain why I was eating copious amount of bacon, cheese, and avocado and wanting a more a more advanced cholesterol test to show that it wasn’t really doing me harm. It wasn’t going well to say the least. I think the word “moderation” make more appearances that standing ovations at a State of the Union address. How did I get here I thought to myself?
A couple months ago I wrote that I had switched to a high fat low carbohydrate (HFLC) diet for a number of reasons. I wanted to share my experience, both good and bad about the HFLC. Briefly, the genesis of this was multi-faceted, but stemmed in a large part from the writing of Peter Attia and the performances of Zack Bitter. So be sure to check out their blogs for more insights.
First, let me explain what exactly a high fat, low carbohydrate diet looks like in actual numbers and in practical terms. The goal really is to minimize carbohydrate intake to less than 50g/day, keep a moderate level of protein (max 2g/kg), and the rest is fat, pure fat. After an adaptation period of a few weeks the science shows that you can dramatically increase the energy coming from fat and fat derived ketones. So what did a typical HFLC day for me looked like?
For breakfast it was bacon and eggs, sometimes with avocado or with cheese. Coffee is a must for me and being on a HFLC diet I substituted my traditional cup with coffee, 1-2 tablespoons of coconut oil, and 1-2 tablespoons of butter that is mixed in a small blender. It tastes almost like a latte. Lunch was a salad with plenty of olive oil and normally some sort of meat (bacon, salami, leftover salmon or steak). Dinner was a piece of meat (fish, chicken, steak) and a veggie side. Veggies were always of the low carbohydrate variety, mushrooms, limited tomatoes, greens, onions, broccoli, brussels sprouts, or green beans but NO root vegetables like potatoes, yams, carrots, etc. Typical garnishes are sour cream, cream cheese, or just plain butter. If I did need to snack it’s nuts or nut butter. Alcohol is limited to a glass of dry red wine and if I am feeling frisky some hard alcohol (bourbon or scotch) straight up. Dessert could be diet soda with whipping cream (yum like a cream sisal) and very dark chocolate (>85% cocoa) or some mixture of nut butter, coconut/almond flour, and egg baked as a cookie. And that’s the HFLC day. Everyday.
Except it wasn’t. I knew that this diet was unsustainable (for me). I just like pizza, beer, and pastries too much. So I took an idea from Tim Ferris’s diet in the 4-hour body, I introduced a “cheat day” where I could gorge on anything that I wanted, be it pizza or beer or the worst candy imaginable for 1 day, each week. It was glorious, but the problem is that I had no idea what this did to the long-term effects of the diet. How long did this cheat day set me back from my nutritional fat adapted state? I didn’t know, so I had to measure it.
Measuring Nutritional Ketosis
Of course thats not really surprising that I measured my ketones, I mean as a scientist I could not just believe that I was fat adapted and in ketosis. Show me the data! So every couple of days I pricked a finger and measured my blood ketone levels (see Figure 1). Nutritional ketosis is the state in which you are adapted to using fat and ketones rather than carbohydrate and corresponds to a level of blood ketones above 0.5mM. This procedure can be expensive ($2/stick) and is for certain an uncomfortable conversation starter around friends. “You are eating what? You are making yourself bleed because why? Because you want to eat bacon and eggs everday?” Ha, joke was on them though, I was in nutritional ketosis and pretty consistently at that. After my occasional cheat days it would drop back to near 0 and would take a couple of days of HFLC to rebound. The diet was working, and I was confident I was on my way to super human fat burning capacities and amazing lipid values. So what happened?
One of the major goals of nutritional ketosis was trying to steady out my energy levels during the day and avoid the highs and lows that come with carbohydrate driven fluctuations in insulin level. In this regard, the diet worked. I did not have those afternoon lows in which all I wanted to do was nap. I didn’t feel like I was starving between meals and was looking for my next fix. I could focus throughout the day and just generally felt very good.
Prophetically I said that the one thing I knew for sure was that this diet would improve my lipid biomarkers. Massive studies had shown this, anecdotal evidence from other ultra runners suggested this. Regardless of changes in performance my lipids were going to get better, right? Well of course I would drop $30 to have them tested. (Note: I’ve written about nuance before and perhaps I should have followed that advice more carefully here.)
My first blood test 3 days into the diet showed my cholesterol breaking down this way:
HDL :69 (very good!)
LDL :161(kind of bad)
I wasn’t overly concerned or impressed with these numbers, sure I would have liked to see the LDLs lower as the last time I had them measured they were 120, but that had been 4 years ago. So I was disinterested until I could again.
6 weeks later:
HDL: 74(very, very good)
LDL: 200 (very, very bad)
I’ll be honest, that LDL number of 200, freaked me out. It’s what foolishly caused me to schedule an appointment with my physician. A total cholesterol of 200 is a stated goal, so seeing an LDL cholesterol value that high was just scary and confusing. As nuance will have it, some people respond poorly to the HFLC diet. Still, always the skeptic and optimistic I knew that the thing about traditional cholesterol numbers is that they are actually pretty useless. There are much better markers to track the risk of for cardiovascular disease. If you want to learn more about this phenomenon I recommend the series on cholesterol that Peter Attia wrote about. Inspired by Peter’s posts about cholesterol and my own reading about the complexity of it I went to see my primary care physician to order a VAP Cholesterol test, which rather than measuring total concentration of cholesterol, it measures the size of the lipid particles that carry the cholesterol and Apolipoprotein B, which is a better marker of cardiovascular disease. Well turns out I had the favorable lipoprotein particle size phenotype and apolipoprotein B levels well within the normal range despite my ultra high LDL (which had already decreased to 136 mg/mL in 3 weeks!). So perhaps I over-reacted. Still we all know that the real reason I wanted to go HFLC was to get faster, so how was that going?
The Inconclusive – Training Benefits
First off, it’s hard to know whether something is improving your fitness when you can’t even train normally. Originally I was hoping that now I would be ramping up my training, my hamstring would be on the mend and I would be training at near normal levels. But that stubborn hamstring is just not cooperating, so I am still at about 50% of the volume and severely limited in my hard efforts. It’s better than nothing, but not good enough to get fit to race given the standard of people that I race with. Still I could tell a difference in training on the HFLC diet. I felt horrible. Everything was hard.
It should not be surprising that a carb adapted athlete might struggle on HFLC, but I was running nearly a minute/mile slower than my normal easy pace. Fast and hilly running, when I occasionally attempted them, were near impossible. Now you might think that is just because I was out of shape from taking nearly 6 months off, and that certainly played a role, but I was also supplementing carbs and running hard on my cheat days. Those runs felt normal and much easier. It was night and day. Still I am not going to withhold judgment here. For some people, some events, it might be ideal. I’m just not sure if I’m one of those persons racing one of those events.
So I fell off the wagon. Maybe it was the holidays, maybe the running, maybe the lipid numbers, maybe that I just enjoyed food a lot less when I had to worry about everything I ate and I seemed to have lost my interest in cooking? All these reasons together made my future diet plans clear, drink beer and eat pizza. Well maybe not that exactly, but at least carbs more than once a week was definitely in my plans.
You often hear the phrase, “live high, train low” in terms of altitude training. I think similar principles are applicable to diet, with “high” being full of glycogen using carbohydrate for energy and “low” being glycogen depleted using fat for energy. To me the HFLC shifted the paradigm from live high, train low, to live low, train high. At times I could introduce carbs to high intensity workouts, so I didn’t feel awful and I could push hard. Before I was doing specific workouts, or combinations of workouts to enhance my fat burning capabilities. For me, I think I prefer the latter strategy, for others clearly the former works. In the end the diet that you eat should be highly individualized and designed for you to reach your goal….and the one that you follow.