Every Meal Counts
Whether you follow the 8 hour diet, Fast 5, Lean Gains, or my own intermittent fasting protocol, you will have noticed that the principal difference comes down to the concept of an eating window. Simply put, each protocol specifies an interval during which no food is taken, and a subsequent interval during which meals are permitted, the eating window. In the 8 hour diet, this window is unsurprisingly 8 hours, Lean Gains is also an 8 hour protocol, Fast-5 gives you five hours in which to eat, and as for me, I actually don’t have a window, but rather advocate eating a single daily meal.
To the extent that intermittent fasting works at all, it does so by limiting the number of meals one takes in a given day. In general, calorie counting doesn’t play a large role. In fact, to listen to the gushing authors of “The 8 Hour Diet” we can :
I think this is grossly overstating the case for intermittent fasting and landing squarely into the Heisendiet category. Nevertheless, if we give Mr. Zinczenko the benefit of the doubt and assume that the diet works as advertised on the tin, then it is remarkable to think that these weight loss effects are brought about by changing the conventional 12 hour eating window ( from say 7:00 AM to 7:00 PM ) into an 8 hour one. Of course, all meals come with attendant satiety, so for practical purposes, there is a limit to how short an inter-meal interval can be, assuming that you are in the habit of not eating unless you are hungry. In practice, when you remove one third of the time that one has available for eating, the net result is the elimination of one of the conventional daily meals, most likely the meal the we culturally know as “breakfast.” With this realization, even Mr. Zinczenko’s diet is reliant upon meal elimination.
More generally, the only reason to claim that your diet has an X hour eating window is precisely because you intend to have the adherents to your protocol eat at least twice. Your average person will at least eat at time T, the beginning of the eating window, and then at a minimum at time T + X, the end of the eating window. There may also be other meals or snacks in between. As a result, eating windows are conducive to “meal frequency creep”, in fact, this aspect is pretty much baked-in to the diet from the outset.
But, if the secret sauce for all IF approaches lies in the reduction of meals, then it follows that the minimum number of meals that one can have in a given day and still eat is one. This is the intuition behind my recommendation to only eat a single meal. This is the only way to maximize the strategy of meal frequency reduction.
The intuition regarding meal frequency is also consistent with our understanding of metabolism. In order for a meal to be assimilated it needs to alter your hormonal profile to convey the assimilation message to tissues at large. This is principally done via the action of insulin, the metabolic acute energy status signal. Accordingly, all meals, save those consisting exclusively of fat, will cause an increase in plasma insulin, which will in turn prompt systemic nutrient uptake.
More specifically, insulin will increase the activity of adipose lipoprotein lipase, suppress adipose hormone sensitive lipase, decrease the activity of muscular lipoprotein lipase ( this is an isozyme of adipose LPL with a different sequence of amino acids, but which catalyzes the same chemical reaction, hence the opposite response to insulin relative to adipose LPL ), as well as induce translocation of the GLUT4 transporters to cell membranes. The net effect of this is to cause muscles to switch from using lipids as an energy substrate to oxidizing and storing carbohydrate, as well as to cause adipose cells to go into fat storage mode.
Our working hypothesis, then, is that by limiting the meal generated insulin spike to a single episode, we will maximize the amount of time we spend oxidizing adipose tissue. It would be nice if we could find some data to either confirm or refute our hypothesis. It is actually surprisingly difficult to find studies in the meal frequency research literature where the infrequent gorging diets ( the term applied by researchers ) consist of a single eating episode. One meal a day seems heretical, anethma to researchers. Rather, you tend to find studies comparing gorging ( three meals a day ) vs. grazing ( 12 meals a day, an hourly meal intake ). If you’re lucky, you may find a study where only two meals are taken. This will require us to be a bit creative in where we find our data, but I think we can make do with some “off label” usage of the data we do find.
To this end, we can turn to the work of Allirot et. al, a group of researchers interested in studying the effects of meal frequency on appetite and subsequent energy intake.
The researchers had subjects consume a 675 kCal breakfast either at one sitting or in 4 equally sized portions, one portion per hour ( i.e. ~170 kcal per hour ). Their goal was to establish whether meal frequency had any impact on subsequent meals intake, and they managed to prove that if you’ve just eaten an hour ago, then you’re likely to eat less at a buffet than if the last time you ate was 5 hours ago. I wish that the previous sentence were an instance of my tongue-in-cheek rapier-sharp wit, but, no, this is seriously what they concluded from their research. Notwithstanding the banality of their conclusion, they collected some interesting data along the way, data that speaks directly to our hypothesis.
Take a look at the below graph, paying specific attention to the NEFA panel at the bottom. Recall that NEFA are non-esterified fatty acids, a.k.a. fat released from your adipose tissue:
As one might expect, the 675 kCal meal elicits a larger insulin response and with that comes a greater suppression of NEFA relative to the small meals. However, 100 minutes into the absorptive phase, the insulin spike from the single meal is waning and NEFA release is on the upswing again for the subjects who had the large breakfast . In contrast, the subjects eating the smaller more frequent meals are maintaining a consistently elevated insulin level, and this sustains NEFA suppression despite the fact that by this point they have only eaten approximately 340 kCals.
You cannot oxidize that which you do not have, which is to say that fatty acid oxidation depends on a concentration gradient. So, the first step to consuming fatty acids it to increase their concentration in plasma. As the plasma NEFA concentration increases, muscles cells will start extracting the fatty acids to use as an energy substrate. Conversely, when we suppress plasma NEFA concentrations, the net result is that while we cannot say exactly what the substrate was that these subjects were oxidizing, we can be quite confident in asserting that it wasn’t endogenous adipose tissue contents.
Initially, then, this study seems to lend some support to the single meal approach. There is more interesting data here, though, and especially when we consider what happens during the first hour of this experiment. If we calculate the area under the curve for insulin during the first 60 minutes, it turns out that the larger meal, while containing 4 times the calories of the smaller meal, elicits a threefold greater secretion of insulin relative to the smaller meal. This is not particularly surprising. However, when we do the same comparison for NEFA suppression, it turns out that during that first hour, despite secreting nearly three times the insulin, we only increase suppression of NEFA by … 7%.
How do we account for this paltry return on our insulin investment? It all comes down to the fact that adipose tissue is exquisitely sensitive to the effects of insulin, so much so that it really doesn’t take a whole lot of insulin to achieve maximal suppression of NEFA release. A plasma insulin concentration of 120 pmol/l will achieve half the maximal suppression of NEFA and typical meals routinely send insulin concentrations north of 400 – 500 pmol/l. As a result, unless you are taking miniscule amounts of food per meal, you are probably going to achieve maximal NEFA suppression every time you eat. Luckily, insulin spikes are transient, unless you have researchers insisting that you eat every hour, at which point you can see the resulting impact on NEFA release in the graphs above.
It is critical to keep in mind that most research that focuses on the effects of meal frequency comes from an obesity, metabolic syndrome, hyperinsulinemia pathology point of view. In this context, the above results, would be considered something of a home run because they would focus pretty much exclusively on the reduction in insulin area under the curve for an isocaloric meal strategy ( the area that I’ve highlighted in yellow in the graph). For our purposes, those of body recomposition, we might be a bit less enthused due to the effects on lipolysis.
The implications of the sensitivity of adipose tissue to insulin levels for meal frequency are fairly unequivocal. Since it takes anywhere from 3 – 5 hours for insulin to return to baseline levels following a meal, if you eat less than 5 hours after your last meal, what you are doing is maintaining a consistently elevated insulin level with its attendant NEFA suppressive effects, much like the subjects eating their hourly small meals above. By the way, this would also help to explain why the conventional strategy of eating three meals a day roughly 4 to 5 hours apart is particularly effective at ensuring that you retain the body fat that you have been accumulating.
After some digging, I was able to pull up one study that actually compared the effects of a single meal per day vs. three meals. Here is the research protocol:
Unfortunately, unlike Alirot and his colleagues, these researchers did not record insulin and NEFA kinetics, so we cannot look at that data. Nevertheless, we can predict that the response should be similar to that observed by Alirot. The truly interesting thing about this study is the duration, 8 weeks in each phase, which is plenty of time to observe the effects, if any, on body composition. So, to cut to the chase:
Chasing down that reference  tells us that both the “bad” cholesterol, LDL, and the good cholesterol, HDL were increased with the 1 meal per day protocol, but much more importantly, there was an observed “decrease in triacylglycerols … after consumption of the 1 meal/d diet. These changes appeared to be independent of the controlled diets, because dietary cholesterol and the ratio of fatty acids were held constant.” Which is to say that merely by eating one meal a day you probably better your chances against heart disease since circulating triacylglycerols correlate much more highly with coronary heart disease than cholesterol numbers.
But the really interesting part of the results is this: despite the fact that the researchers were actively intervening in the diet in an effort to ensure that “subjects were fed at an energy intake that would maintain body weight so that meal frequency would be the only major change in their diet during the course of the study”, at the end, on the 1 meal per day protocol subjects lost 2.1 kg of fat mass and gained 1.5 kg of lean mass, and all of this even though there was “no evidence … of a significant difference in physical activity after consumption of the 1 meal/d or the 3 meals/d diet!” As an added bonus, the single meal per day protocol halved the circulating cortisol of the subjects, although this may have been an artifact of the different times during the day that the samples were taken. My kind of diet!
Assuming that subjects could maintain these rates of fat loss / lean mass gains, over the course of a year they would be looking at a fat mass decrease of about 12 kg with an attendant increase of lean mass of almost 10 kg, and all of it just by coalescing their meals into one a day! Dare to dream!
How did the researchers explain this weightloss?
Or to put it in my words, if you eat less than 5 hours after your last meal, what you are doing is maintaining a consistently elevated insulin level with its attendant NEFA suppressive effects, much like the subjects eating three daily meals.
When is a 2000 kCal diet, not?
You may be familiar with the concept of the thermic effect of food (TEF), or you might know it under the name of the specific dynamic action of food (SDA), or maybe diet induced thermogenesis (DIT). Whatever the name, this refers to the observation that the act of eating entails first and foremost a metabolic expenditure of energy for the digestion, processing, and storage of food. This energy is not available for general metabolism, it is effectively sequestered, reserved for digestion. The cost of doing business, if you will. Anywhere from 5 to 15% of a meal’s caloric contents will be consumed merely by the digestive and absorption process.
It may strike you that the TEF range of 5 – 15% is rather broad, and you may be curious to know what determines that number. There are a couple of determinants. The first is meal composition. Protein is the most thermogenic of nutrients, followed by fats, then carbohydrates. So the more protein heavy your meal, the fewer net calories you will be ingesting. The second determinant is … meal size. It turns out that the thermic effect of food is proportional to meal size. Or, in other words, as meal size increases, a greater proportion of the caloric contents of the meal is consumed by the digestive process itself ( i.e. the net energy of the meal goes down ). So how could you go about reducing your caloric intake without actually reducing your caloric intake? By eating your day’s calories all in one go. Rather neat, that.
What do our friends, Allirot et al., have to say about TEF? Well :
Apparently, if you eat more frequently (F4) you get the “beneficial short-term effect” of not being as hungry in the buffet line one hour after you last ate, but of course, this comes at the cost of reduced diet induced thermogenesis, and “an inhibition of lypolysis.” I guess its ultimately up to us to determine whether that “benefit” is worth the cost. I’m pretty clear on where I stand.
When all is said and done, I find the evidence in favour of a single meal compelling and I hope that I’ve managed to convince a few of you as well.