My n=1 Cyclical Ketogenic Experiment

Mini life-update:

On all accounts, this summer could not have been better, and thanks to UCSD’s quarter system, it continues for another week before classes resume. My research group finished designing a pilot experiment now awaiting funding and IRB approval. Run conditioning is improving, and I finished my first continuous half marathon injury-free a couple of weeks back. Collegiate triathlon season starts next weekend at UC Berkeley’s Bearathlon where I will be competing in both the draft legal and classic sprint triathlons (my legs ache just thinking about that). I’ve decided to heed the advice of my coach and not compete at Ironman Maryland this year, but I have not yet ruled out competing in my first in 2018.

Besides research and training, I had the opportunity to travel back east for a pair of weddings, between which I was able to visit home and spend time with family and friends, part of the reason for the delay rolling out this post. The other part is my mistaken belief that it would be better compiling all my findings on the cyclical ketogenic diet into a single entry instead of a series of shorter posts. Hence, what follows is longer than a typical blog post, so to make it more reader-friendly, I’ve broken it up into three parts:
1. An introduction to cyclical keto, for those who have no idea what this post is about
2. A review of the relevant research, for my particularly curious readers, and for me to quickly locate resources later on
3. A bloggy part, for those wanting to read about my results so far.

– –

In-N-Out 4×2 Animal Style, Protein Style. Ohhhh yeahhhh!

If you’ve shared a meal with me over the past few months, you’ve probably observed something strange in my eating habits. Depending on the day, you might have noticed me avoiding carbohydrates like the plague – no breads or buns, pasta or potatoes, starch or sugar. Or, if you joined me on a carb refeed day (otherwise known as “Faturday”), you likely stared half puzzled, half amazed as I inhaled sushi like I hadn’t eaten for weeks or polished off an entire quart of ice cream by myself.

Below, I bring you aboard my nutrition expedition that began as a casual search for improved performance and turned into a borderline obsession spending every free minute dissecting articles, listening to podcasts, and reading about people’s experiences with ketosis. For the first time in my life, I’ve taken an interest in something biology-related (sorry med school friends).

 Disclaimer: Talk to your doctor before making any substantial dietary changes. I am not a doctor, nor am I qualified to give medical advice. This post is for entertainment informational purposes only.

Cyclical Keto: Why, What, How

“Put good in, get good out.”
“You are what you eat.”
“Let food be thy medicine and medicine be thy food.”

Diet matters – we all know that. What we don’t know is which diet enables us to perform at our best, though it’s probably something with vegetables and probably not Bojangles/Cookout/Chinese (i.e. my undergrad diet). Every body is different, and what is best can only be determined by experimenting. “Best” depends on our goals. For me, I want my diet to:

1. Improve Ironman triathlon performance
2. Improve cognitive performance

However, I have no interest in increased performance if I’m left feeling deprived and “hangry” all the time. And I’m not willing to sacrifice long term health for short term performance gains, so my diet needs to both leave me feeling satisfied and maintain or improve my health biomarkers (e.g. cholesterol, triglycerides, etc.).

It’s unwise to use research alone to craft something as individual as one’s diet, but my review of recent nutrition studies suggests that the cyclical ketogenic diet (CKD) may be promising for my purposes. And now, five months in, here’s some anecdotal evidence:

  • Swam 7 miles in the ocean after an overnight fast, biked 50 – 100 miles multiple times without consuming any carbohydrates, and ran a half marathon after consuming only one serving of UCAN Superstarch, all without bonking on any occasion.
  • 3 PM slump is gone, mental fog is nonexistent, and hunger-induced irritability is a thing of the past.
  • Triglycerides and total:HDL ratio are lower and far below any appreciable level of heart disease risk, blood pressure dropped from borderline high to normal, bodyfat (measured by DEXA scan) is just under 11%.
  • Typical keto day is 4,000+ calories of satisfying meals, snacks, and dessert – tonight’s was chocolate mousse with coconut and macadamias.

What is the cyclical ketogenic diet?

Most of the time I follow a ketogenic diet: a very high fat, low carbohydrate, moderate protein diet – see the pie chart below. On occasion (currently once every two weeks), I’ll cycle/cheat/refeed and eat anything I want, including pie. Most likely Chinese food and ice cream too.

Low carb?! Shouldn’t endurance athletes be piling on the carbs?!

While it’s true that athletes need glucose or glycogen (stored glucose) to perform high intensity exercise, endurance athletes who operate at a moderate intensity for long periods of time (e.g. Ironman competitors) can power their workouts using mostly fat if their bodies have been adapted to burn fat as the primary fuel. This is where the ketogenic diet comes in. It triggers complex changes in the body that raise the rate at which fat is oxidized, enabling one to use more bodyfat and depend less on consumed food, which is especially helpful when carrying or stomaching enough food is difficult. I will call these changes “fat adaptation,” which takes weeks to begin and months to plateau.

Fat adaptation begins with glycogen depletion. Step 1 is restricting carbohydrates and somewhat protein (since the body can turn protein into glucose), which forces the body to use its glycogen stores. The body can only store 2,000 or so calories of glycogen. Step 2 is burning off that stored glycogen. Once the stored glycogen is gone, if not replaced, the body will enter a state of ketosis in which the liver turns fatty acids into ketone bodies – the chemicals Acetone, Acetoacetic acid, and Beta-hydroxybutyric acid – to be used for energy. Ketosis is an adaptation that has enabled humans to survive starvation. Fortunately, one does not have to starve to enter ketosis. Instead, one can achieve ketosis by eating a high fat, low carbohydrate, moderate protein diet.

I glossed over the biology and chemistry involved in ketosis, but the following three videos will give you a better understanding. The first covers the nuts and bolts, the second is a bit more in depth, and the third might as well be a medical school lecture.

This video covers the basics of ketosis:
Peter Attia, MD, does an excellent job briefing the biology of ketosis:
This video describes what’s happening chemically during ketosis:

Everyone has ketone bodies in their bloodstreams, but physicians only consider those with concentrations greater than 0.5 mmol/dL to be in ketosis. One can check ketone concentration using a ketone meter, which works like a glucose meter. I use the Novamax ketone test kit since the strips are the cheapest (about $2 each) among the reputable brands. Other options are urine and breath tests, which are reportedly less reliable. When I tried the urine strips, I tested negative for ketones even though my Novamax meter read 1.7 mmol/dL (below).

Novamax ketone meter 
 
Example urine ketone test taken immediately after above blood ketone test

Don’t you need carbs to live?

There are essential fats (omega 3s and 6s), there are essential amino acids (from protein), but excluding fiber, there are no essential carbohydrates. The body can be almost completely fueled by fat, and the few cells that require glucose ( e.g. red blood cells, thyroid) get enough from the conversion of glycerol (from fat) and some amino acids (from protein) into glucose. Thus, one can survive without consuming any carbohydrates other than fiber. Fiber is the exception since it’s fuel for your gut flora and promotes proper gastrointestinal (GI) function, but because fiber does not generate an insulin response, fiber need not be limited on a ketogenic diet. Instead, fiber consumption is encouraged, especially diverse sources of fiber. More on this below.

What do you eat on this diet?

One key to success on this diet is lots and lots of vegetables…coated in fat. Before this diet I struggled to eat enough veggies. Take any vegetable, cover it in butter or cheese, and suddenly something hardly palatable becomes delicious! Remember that on this diet, fat is good. And since I’m not interested in losing weight, I am vigilant about getting enough fat. Vegetables, dense with fiber and micronutrients, are my fat shuttles of choice.

I eat a similar breakfast almost every morning: 2 eggs and 2 cups of spinach sautéed in olive or avocado oil with a side of olives, avocado, or tomato, sometimes with cheese. Occasionally in the morning I’ll have a double espresso with cinnamon and either heavy cream, butter, or MCT oil. I’m still not sure how I feel about putting butter or oil in my coffee…

A typical breakfast at home 
 
I was pleasantly surprised by the keto-friendliness of the hotel breakfast buffet

Lunch and dinner vary, but I consistently include some form of leafy green vegetable. I’ll often make an enormous salad (bigger than my head) with mixed greens and romaine topped with cheeses, chopped nuts, hardboiled eggs, avocado, bell pepper, cucumber, olives, tomato, broccoli sprouts or florets, high-fat low-carb dressing or EVOO, salt and pepper, and some form of protein, ideally a fatty fish like salmon. When I have a busy week ahead, I’ll batch cook cheese and cream-based casseroles with veggies and some protein.

So good I forgot to a take a picture before diving in
 
Cheesy, creamy goodness 

When I’m craving a particular dish, I can usually find keto-friendly recipe online. One of my favorites is a keto-friendly General Tso’s meatballs, which I served over cauliflower rice once and Miracle Rice another time.

It ain’t China Queen, but it hits the spot

The ketogenic diet is a macronutrient balancing act. Too little fat and one feels lethargic, too much and one gains weight. Too little protein and one loses muscle, too much and one is knocked out of ketosis (since protein can be turned into glucose). Too little carbohydrates and one is probably being too strict with vegetables and not getting enough fiber, too much and one is knocked out of ketosis. A good rule of thumb is 70/20/10 fat/protein/carb ratio, but this is only a guideline. A better plan is to first determine protein and caloric need, which varies depending on activity level. Phinney (2004) suggests eating at least 1.2 grams of protein per kg of body weight (0.55 g protein per lb) and no more than 1.7 g/kg (0.77 g/lb) – that’s between 75 and 130 g of protein per day for someone my size. Next, plan foods that provide enough fiber (I aim for 30+ grams per day). Finally, try to meet the remaining calories with fat and as few carbohydrate as possible.

For more meal ideas, check out this 14-day keto-friendly meal plan.

Aren’t you going to clog your arteries with all the fat you’re consuming?!

I could write an entire post on this topic alone (and plan to), so I’ll only briefly comment here. If you’re afraid of fat, read this article, which explains how the sugar industry has propagated a pro-carbohydrate public perception by selectively funding research aimed at making fat the enemy. Fortunately, public perception is shifting as conflict-of-interest-free research continues to reveal that fats are essential to a healthy diet. More and more frequently I come across the phrase “healthy fats” in pop-health articles. However, even proponents of avocado, salmon, and walnuts seem to demonize the saturated fats common in meats and animal products. Indeed, the American Diabetes Association, American Heart Association, Cleveland Clinic, Food and Drug Administration, Mayo Clinic, and World Health Organization all recommend limiting saturated fat intake. Convinced?

But “saturated fat is bad” does not tell the whole story. First, consider that there are ten types of saturated fatty acids (SFA) processed differently by the body. Researchers have demonstrated that some medium-chain SFAs like Lauric acid may lower the total cholesterol to HDL ratio, correlated with reduced risk of heart disease (Mensink et al 2003). Other SFAs like Palmitic acid, the most commonly consumed SFA in the U.S. (Ervin et al 2004), may raise the “bad” LDL cholesterol (Zock et al 1994). SFAs are more nuanced than the single blanket category “saturated fat” seen on nutrition labels.

Second, consider the endogeneity problem with saturated fats. Who eats a cheeseburger without a bun? Besides a large dose of SFAs, one also consumes a ton of refined carbohydrates when eating a typical American burger. The body is complex, and the interaction between simple carbohydrates and SFAs in the body may drive the ill health effects presumed to be caused by the latter. More research is needed to determine the health impacts of SFAs in isolation and in the context of low carbohydrate diets. One recent study, Forsynthe et al (2010), found that when carbohydrates are restricted, participants with diets high in saturated fats (from dairy and eggs) experienced no change in inflammatory markers. Though promising, more research is warranted before one can make claims about the safety of saturated fats when following a low carb diet.

How could such a radical diet be healthy?!

The ketogenic diet is only “radical” when compared to the modern Western diet. Compared to the diet of our hunter/gatherer ancestors, the Western diet is far more radical than the ketogenic diet (see table below). If evolutionary theory is correct, it follows that our ancestors’ genes, which we carry today, were crafted by natural selection to thrive on the diet available at the time. The implication is that we should expect to be healthier eating like our ancestors once did.

From Simopoulos (1999), modified from Simopoulos (1992)

Certainly our ancestors did not eat a ketogenic diet or even the “Paleo” diet as worshiped by the Crossfit community (here’s what they actually ate). But our ancestors, without grocery stores, restaurants, or even refrigerators, did periodically starve and enter ketosis. Compared to the typical Westerner who eats three or more high carb meals per day and never enters ketosis, those who follow a ketogenic diet likely experience metabolic states closer to that experienced by our ancestors. In addition to providing energy, ketones function as signaling molecules capable of altering gene expression, therby altering metabolism and risk of some diseases (Newman & Verdin 2015). Some scientists believe the ketogenic diet may prove useful for treating/preventing some neurodegenerative diseases and perhaps cancer (Seyfried 2014).

To be clear, there is evidence that low glycemic diets that emphasize fiber-rich whole grains and minimize sugar and refined carbohydrates can be healthy and well-tolerated (Barclay et al 2008). Additionally, grains are extremely cheap, so both your waistline and your wallet may benefit from a diet that allows carbs. Hence, I’d only recommend the ketogenic diet if you seek the unique performance advantages of ketosis, which I present next.

Ketogenic Research

Below, I present some of the research that prompted me to experiment with the ketogenic diet. The medical literature is rich with observational studies that fail to establish causality. While it is tempting to use observational studies to support arguments, I try to stick to clinical trials whenever possible, which do a better job testing causal relationships. Ethics, time, and cost can make human trials infeasible, so researchers often test animals (usually mice) instead of humans. Again, the results should not be ignored, but one must take caution when generalizing the results of animal studies.

Keto for endurance performance

An Ironman athlete of my size burns 600-800 calories per hour during a race lasting anywhere from 8 to 17 hours. An athlete who averages 700 calories per hour for 10 hours will have burned through 7,000 calories by the time he finishes. Our bodies store about 2,000 calories of glycogen in our muscles and liver, leaving a deficit of 5,000 calories that must come from sources other than stored glycogen, namely bodyfat and food, otherwise…bonk!

A person my size carries with them more than 75,000 calories of bodyfat. That’s enough to fuel almost ten Ironmans without eating anything. The problem is that the rate at which most athletes burn fat (less than 30 grams per hour) is not enough to sustain race pace intensity. Consider an Ironman athlete who burns 700 calories per hour and eats nothing. He can fuel 270 calories per hour from bodyfat and the remaining calories using glycogen. Sadly, his limited glycogen stores will deplete in under five hours, about halfway into the 112-mile bike segment.

To avoid bonking, this athlete must consume around 300 calories, or 75 grams, of carbohydrates per hour. That’s about 8 ounces of Gatorade and a gel every half hour, which sounds to me like a recipe for GI discomfort.

I’ll borrow an analogy from Peter Attia that reveals the irony of carb-dependent bonking: imagine a tanker truck that has run out of fuel. The tanker carries a virtually unlimited supply, but alas, it can only access the fuel in its smaller, primary tank. Likewise, most athletes carry tens of thousands of calories as bodyfat that cannot be accessed quickly. Instead, they have to continuously eat carbohydrates to sustain effort.

If only it were possible to tap into that endless supply of fuel…

Enter fat adaptation.

Research suggests that sustained carbohydrate restriction increases fat oxidation rates, i.e. induces fat adaptation and reduces need for consumed carbohydrate. Consider the graphs below, created from data from the FASTER study out of the University of Connecticut, and the different fat oxidation rates between highly trained endurance athletes who eat low carbohydrate diets (LCD) versus high carbohydrate diets (HCD).

The data indicate two suggestive phenomena. First, peak fat oxidation is higher for those in the LCD group (published numbers are 1.54 +/- 0.18 versus 0.67 +/- 0.14 g/min, a 2.3-fold difference), enabling them to use more fat at all intensity levels than can those in the HCD group. Second, peak fat oxidation occurs at higher intensities for the LCD group compared to the HCD group (published numbers are 70.3% +/- 6.3% VO2 max versus 54.9% +/- 7.8%). Ironman athletes race at about 70% – 80% of VO2 max, which is right on the money for peak fat oxidation in the LCD group. Hence, a LCD may reduce an athlete’s dependence on consumed carbohydrates during endurance races.

Back to our previous athlete example. If he managed to increase his fat oxidation rate from 30 to 60 grams per hour, he would be able to supply just shy of 700 calories per hour from bodyfat. Now, instead of 8 ounces of Gatorade and an energy gel every half hour, he could sustain exercise on just 6 ounces of Gatorade every half hour.

Theoretically, our athlete could go 10+ hours fasted and not bonk, but his starting glycogen may be reduced. For a factor of safety, it would be smart to add some exogenous carbohydrates. Additionally, actual fat oxidation rates and caloric needs are not necessarily linear (I assumed 60 g/hour of fat oxidation and 600/700/800 calories/hour for the swim/bike/run, respectively), but the model provides a ‘good enough’ approximation of fuel needs to illustrate the point.

Want to learn more about the FASTER study? Check out Volek’s presentation here:

The FASTER study does not prove that LCD causes increased fat oxidation since the recruited athletes were not randomly assigned to either diet. There could be something omitted among those who chose to go low carb, perhaps genetic differences, that is causing the observed difference in fat oxidation. Fortunately, other scientists have tried to demonstrate the causal effect of LCD on fat metabolism.

In a study by Phinney et al (1983), five cyclists were first given a balanced diet with a 1:1 ratio of fat to carbs by calories. Then, for four weeks, the cyclists maintained a diet with a 9:1 ratio of fat to carbs. The researchers compared a host of performance measures before versus after the dietary change. Most notably, they found that the cyclists, after adapting to the new diet, had no significant change in submaximal performance. Under the surface, however, the cyclists experienced reductions in muscle glycogen mobilization by more than four-fold and blood glucose oxidation by three-fold. Instead of glucose providing most of the cyclists’ energy needs, fat became the primary source of fuel.

Granted, this study’s sample size was small, and the effect on each cyclist was highly variable. But the magnitude of the change is large enough that self-experimentation is merited. I’m not the first athlete to try restricting carbohydrates for improved endurance performance. Tim Olsen is a keto-adapted ultramarathoner who holds the course record for the Western States 100, a 100-mile trail race in California. Zach Bitter is another keto-adapted ultrarunner who holds the world record for furthest distance run in 24 hours. Other keto-adapted high performers include Kevin Grabowski, Ben Greenfield, Sami Inkinen, Mike Morton, and the list goes on.

Peter Attia, MD is a marathon swimmer/cyclist and self-experimenter who conducted elaborated tests after switching from a HCD to LCD. In his lecture below, he explains his experiments and how researchers figure out how much of each macronutrient is consumed during exercise.

Ketosis is not without its performance downsides. Multiple papers and anecdotal evidence suggest that all-out, top-end power is reduced while following a ketogenic diet. Hence, it would not make sense for sprinters or shorter distance triathletes to follow a ketogenic diet. Considering that humans of my size store about 2,000 calories of glycogen and burn under 1,000 calories per hour during endurance exercise, athletes would probably not benefit from a ketogenic diet when competing in events lasting under two hours.

Other areas of ketogenic research

Starting in the 1920s, doctors began proscribed the ketogenic diet to patients with epilepsy to help prevent seizures (Kinsman et al 1992). Since then, researchers have examined the diet for treating metabolic diseases, neurodegenerative diseases, and cancer (Seyfried 2014). Additionally, there is evidence that ketones are the preferred fuel for the brain and that ketone production increases while exercising in ketosis, thus enhancing focus (Volek et al 2015). The opposite is true for carb-dependent athletes who begin to feel foggy as circulating glucose diminishes during exercise.

I’m already pretty deep in the literature for this entry, so I’ll hold off the remainder of my findings for a future one.

The “cyclical” part of the diet: carb cycling

Most existing research has focused on strict ketogenic diets, not cyclical ketogenic diets like the one I am following. As such, it remains a question whether I will reap the full benefits of the ketogenic diet when including regular carb-cycling. I would consider carb-cycling less frequently, but I love bread, pasta, and other carbs too much to sacrifice them ad infinitum. Additionally, adding carbs once per week replenishes glycogen stores and aids in hormone production and hence may improve short-run (and possibly long-run) performance when consumed strategically.

Because carb cycling in the context of ketosis is not well studied, carbohydrate “dosing” remains something to be fine-tuned as I continue with the experiment. I hope to figure out how often I can eat high-carbohydrate meals and still benefit from ketosis. My current carb consumption method is the following:

  • Refeed day: eat enough carbs to exceed glycogen storage saturation, i.e. more than 500 grams of carbohydrate. 500 grams would only be necessary if glycogen stores were totally depleted, which if not the case may lead to weight gain. However, I can live with moderate short term weight gain for the improved diet adherence on low carb days.
  • First few days after refeed day: eat very low carb, moderate protein and complete higher intensity workouts (e.g. interval training, hill repeats, sprints, lifting), ideally while fasted, to burn off excess glycogen and resume ketone production. If followed correctly, ketone levels should be back to above 0.5 mmol/L within 3-4 days.
  • Intermediate days before next refeed day: keep carb consumption low and only supplement with targeted, low quantities of slow-release carbohydrates when needed (e.g. for race days)

I need to experiment further, but now that I’ve become fat adapted, it seems that cycling once every two weeks does not impair my ability to oxidize fat. More to come soon.

Special considerations on keto

Microbiome

Although much is still unknown, evidence suggests that a diverse microbiome plays a part in health including immune function, hormone balance, and weight management (Flint et al 2012). Most GI experts recommend a lifestyle that promotes diverse gut flora including eating an array of fiber sources, fermented foods, and avoiding oral antibiotics.  Kimchi, pickles, sauerkraut, and most vegetables are keto-friendly and make your gut happy. Personally, I’ve had experienced better GI comfort while on keto than ever before, but only after I made a concerted effort to increase my fiber intake. Even my desserts contain fiber when I include 100% cacao (nibs or powder).

Sodium

When following a low carb diet, lower levels of insulin cause one’s body to excrete more sodium than usual through a process called natriuresis (DeFronzo 1981). Low sodium can lead to headaches, dizziness, low blood pressure, cramping, and poor athletic performance. Phinney (2004) recommends that those following a ketogenic diet consume upwards of 5 grams of sodium and 3 grams of potassium per day to maintain healthy electrolyte levels. Before concertedly adding salt to my food, my blood pressure dropped to 100/60. Now I consume at least 5 grams per day, more when exercising, and no longer experience symptoms of low sodium.

Hypothyroidism


Research has shown that carbohydrates play a role in thyroid function, and when carbohydrates are substantially restricted, serum T3 levels may decline (Spaulding et al 1976). Fortunately, reducing T3 hormone may slow the rate of aging (Fontana et al 2006). Unfortunately, reduced thyroid hormone can negatively affect athletic performance, so for now, I’m prioritizing having sufficiently high T3 levels. I’m hoping the intermittent carb influx is enough to keep the thyroid happy, but I will continue to monitor through regular blood panels. So far so good.

Results so far

My diet pre CKD in one word? Garbage. Here’s a typical week of lunches while a senior at NC State:

  • Monday: General Tso’s chicken and white rice (gotta start off the week right)
  • Tuesday: 3 Wendy’s chicken sandwiches (bunned and battered) plus a Frosty
  • Wednesday: Cookout tray with Mint Oreo & PB milkshake (don’t deny it ’til you try it)
  • Thursday: Chipotle burrito bowl (healthy? Not really – my go-to bowl has 141 g of carbs (50 from the tortilla alone) and 1,200 calories…and that’s before the chips! And likely an underestimate since I employ some Chipotle hacks)
  • Friday: An entire medium Pokey Stix from Gumby’s. Or General Tso’s chicken again because you can’t beat perfection.

Friends often commented on my food habits, but why should I care? So long as I regularly exercised, I could stay fit…but body composition only tells part of the story. Under the surface, my blood pressure was high enough that my doctor suggested reducing sodium intake. Additionally, I had to eat every three hours or so, lest I release The Hanger (duh duh duh!). To stave off the hanger, I would always leave the house with some snack on hand to spike my blood glucose if needed. Not exactly ideal.

My training also revealed a need for a dietary shift. During an event, I had two options: 1) consume sugary gels and drinks and feel ill but have energy to finish the race, or 2) eat less, have a happy gut, but bonk. Neither option seemed suitable for finishing an Ironman, let alone being competitive, so I opted to add a third option.

First Weeks

I began following CKD in May after collegiate triathlon season ended. The first two days on the diet were hard, to say the least. I crrraaaaved carbohydrates, especially chocolate toaster waffles for some reason. But after getting over the two-day hump, the cravings subsided. And then the weirdest thing happened – I stopped feeling hungry, at least not the same ravenous “I could eat an entire cow” hunger common before. I could skip breakfast or work through lunch and not think about food incessantly, a wondrous change from my high-carb past.

Four days in, feeling psychologically strong, I bonked on my three-mile bike commute to campus. I could hardly pedal at half the speed as usual. “What the hell?!” Hill repeats the day before had left my glycogen stores mostly depleted, and only four days in to CKD my body still needed glycogen to function. I felt like I was fighting a headwind regardless of the direction I traveled. Reading about others’ adaptation phases, I knew I’d experience some temporary performance decline, but I had no idea it would hit so hard.

At last, my first “Faturday.” I started the morning with a bagel, waffle, toast, and a pain au chocolat. Lunch was (you guessed it) Chinese food at a buffet that had both General Tso’s chicken and sushi, and I finished off the day with half a large pizza and a pint of mint chip ice cream. Glycogen stores replenished.

Time to train. I felt like superman on the bike: averaged 22.5 mph on a 30-mile route with stop lights. Next day, same thing: crushed it. But by the next day, I once again felt as if I were pulling a ball and chain, shackled by glycogen depletion.

So the cycle went: tired muscles and sucky workouts most of the week, until refeed day and the day after when I’d feel unstoppable. The performance decline that would inevitably ensue was discouraging, but I continued the experiment hoping fat adaptation would set in eventually.

Three weeks into the experiment, I was still struggling physically, but commutes started getting easier. I was no longer stuck in the granny gear and could manage a somewhat faster pace. Psychologically, I was pleasantly surprised by my focus and lack of hanger. The best part: I could eat a huge lunch and have zero 3pm slump – extremely helpful when your job involves reading papers on complex economic models.

Several weeks later

The morning of my carb-cycling day, I decided to skip breakfast and go for an ocean swim. At that point, I had restricted carbs for seven straight days, and I wanted to see how far I could swim before bonking. I started north from La Jolla Shores, hugging the coastline should I need to stop. I felt like I could hold a moderately-high pace forever and ended up swimming nearly seven miles. Despite moderate dehydration, I felt great and could’ve continued a while longer. Breakthrough! I celebrated with three plates of Indian food for lunch.

Fast forward to today

While on keto, top-end, all-out power is still limited, but my sub-threshold endurance is stronger than ever, even without eating before or during workouts. And when I do eat carbohydrates, I gain back my ability to climb and sprint, which begs the question: do I eat carbs before or during a race, or am I better off avoiding carbs and staying in ketosis?

I’m attempting to find the answer by experimenting with “strategic” carb consumption. I purchased a sample pack of Generation UCAN Superstarch, which contains hydrothermally treated waxy maize, a carbohydrate absorbed so slowly that insulin response is minimal. The product was initially developed to treat Glycogen Storage Disease (GSD), which requires patients to consume a near constant flow of glycogen (Ross et al 2016). While sleeping, patients with GSD must wake every few hours to consume a source of glucose. With Superstarch, however, patients could sleep nearly eight hours on average since the slow absorption provides a constant source of glycogen throughout the night.

Later, Generation UCAN began marketing Superstarch to endurance athletes who want to replenish glycogen stores without spiking insulin and shutting off fat oxidation. Roberts et al (2011) conducted a clinical trial with superstarch and demonstrated its efficacy for providing a continuous source of glucose without spiking insulin or reducing fat oxidation compared to consuming the same calories as maltodextrin.

Peter Attia explains the science behind Superstarch in the video below (he claims no affiliation with Generation UCAN):

Looking forward

I have a host of experiments in mind to test my body’s response to particular foods or lack thereof. I’d like to test the effect of MCT oil, artificial sweeteners, and moderate carbohydrate foods (yogurt, carrots, chickpeas) on my blood glucose and ketones. I’m also interested in how intermittent fasting and larger, less frequent meals affect ketone production. Another future experiment involves exogenous ketones, which may help me get back into ketosis following a day of carb-cycling.

When I reexamine my diet in December, I imagine I will stick with some variant of this diet. At the very least, I will continue to restrict added sugar and refined grains to one day or less per week. Since I will be racing shorter triathlons January through April, which require more high intensity, anaerobic efforts (and more glycogen), I may add some whole grains, beans, and fruit back to my diet. Even then, I may experiment with restricting carbs a few days before a race and then adding carbs back on race day.

Concluding Thoughts

You might think after reading this post that I’d recommend trying CKD, but it’s is not for everyone, and good thing too – it’s not a globally sustainable diet. Maize, rice, and wheat, none of which are keto-friendly, provide about two-thirds of the world’s food energy intake (Cassman 1999). What I would encourage you to do is to be mindful of your nutrition choices and how they align with your goals – educate yourself, read labels, track your diet for a week with an app like MyFitnessPal. Even if you’re in excellent physical shape, adjusting your diet may give you the edge you need to compete with the best in your sport. Or perhaps you’re interested in reducing your risk of developing neurodegenerative diseases, which may be accomplished by reducing intake of added sugars (Cherbuin et al 2012).

Too often we think of a “diet” as something temporary. When you ‘go on a diet’, you risk erasing all the work you put in as soon as you go off the diet. Instead, make your diet part of your lifestyle, and keep adherence in mind when setting dietary goals. When a diet is too restrictive, adherence is nearly impossible (are you really never going to eat cake, cookies, or ice cream ever again?). The psychological factors of one’s diet are as important as the physiological factors.

This is not a post about weight loss, but I recognize weight loss is a top dietary goal for many. In general, if permanent weight loss is your goal, I urge you to acknowledge that temptation will always be present. Schedule refeed days as outlets for temptation when you permit yourself to eat whatever the heck you want. Otherwise you will feel deprived, eventually cave, and gain back the weight you’ve lost. When only once per week or less, a binge day makes up in psyche what it loses in weight loss progress. I’ve found I rarely feel deprived allowing myself carbohydrates once every two weeks. If I have cravings, I write them down and eat those foods the next refeed day. I also look for keto-friendly recipes of foods I crave including pizza bake, cheddar biscuits, chicken parmigiana, and ice cream.

Pizza pizza
All the taste of Red Lobster biscuits with none the carbs
This batch included egg yolks, but I found the recipe tastes just as good using only cream, stevia, and vanilla extract

Recommended podcasts

If you’re a podcast junkie like me and want to learn more about the ketogenic diet, check out the following:
Dom D’Agostino on the Tim Ferriss show #1
Dom D’Agostino on the Tim Ferriss show #2
Dom D’Agostino on the Tim Ferriss show #3
Dom D’Agostino on FoundMyFitness
Peter Attia on the Tim Ferriss Show
Peter Attia on FoundMyFitness

 

 

4 Replies to “My n=1 Cyclical Ketogenic Experiment”

  1. When you refer to carbs, are you talking about breads, grains, etc only? Or does that also include fruits? Or, would fruits be lumped in with veggies?

    1. I count anything that provides a glycemic response to be a carbohydrate, whether that be the grains in bread, sugars in fruit, or something else. Fiber is not counted as a carbohydrate because it provides no glycemic response, i.e. fiber does not increase blood sugar levels. Some sugar alcohols also have very little glycemic response, which is why you see sugar alcohols often subtracted from total carbs when calculating “net carbs” listed on some low carb products.

      I try to avoid fruit for the most part because of the large dose of sugar. Fruit is “healthy” for its micronutrients and fiber, which I instead supply with vegetables. Starchy vegetables (potatoes, beets, carrots) are also high in net carbohydrates and should be avoided or restricted. Most legumes (beans) should be avoided as well.

  2. Zach, super interesting self-experiment. I have a couple of other podcast recs talking about keto and low carb, with the focus being reversing insulin insensitivity (eg for pre-diabetic or diabetic populations) if interested. I’m curious what your plan is for the shorter races coming up in collegiate season, and how you’ll be able to modify and jump in-between being endurance-race ready and sprint-season ready. Also, I 100% agree with your statement that “adjusting your diet may give you the edge you need to compete with the best in your sport.” In my own self-experimentation, I’ve found that consciously changing my diet allows for better fueling for training and racing , and avoiding that afternoon slump, though I didn’t do keto.

    1. Thanks Clare, let me know about those podcasts! I’m not yet sure what will be the fueling strategy for sprints – I’ll have to experiment and see what works! For Berkeley, since I’m racing twice, plan is to try a mix of fueling strategies (second race higher carb than the first).

Leave a Reply

Your email address will not be published. Required fields are marked *