I had intended to write about something different today, had it not been for this new and exciting study that came out a few days ago. The study, named “Training in the fasted state improves glucose tolerance during fat-rich diet“, adds to the scientific evidence that shows that fasted training may have unique benefits* that are not obtained with fed state training.

* Previous studies had showed that fasted training improves endurance and increases muscle glycogen stores. There has also been work done that suggests an “anabolic rebound” effect from fasted weight training. Theoretically speaking, fasted training should also aid with “stubborn fat loss“, such as lower abs in men and hips and thighs in women.

The new study

Let me briefly summarize the study methodology before I’ll talk about the results, which are very interesting indeed.

The main objective here was to find out whether fasted training had the potential to induce superior and favorable adaptations to fat metabolism, glucose tolerance and insulin sensitivity when compared to fed-state training. The researchers thought this was a hypothesis worth exploring, based on previous studies that showed superior effects of fasted training on fat metabolism and metabolic gene expression. For example, De Bock, et al (2005) and Cluberton, et al (2005).

To test this, they recruited healthy, young males who were reasonably lean and active. The participants averaged 3.5 hours of “physical activity” per week, which I reckon is close to, or less, than what most of my readers engage in. I actually think this is a very strong point of the study. These were not obese couch-potatoes and the results obtained with these subjects should be very applicable to us.

Two weeks before the start of the study, various tests were perform to determine baseline body composition, fitness, diet and health marker values. Then they were randomized into three groups, two of which followed the same diet and training regimen. Only nutrient timing was different.

  • Training: Two 60-min and two 90-min supervised training sessions per week, always between 6:30 and 9:00 a.m. Training sessions consisted of a combination of cycling and running exercise. Intensity was adjusted to each individual and set to 70-75% VO2Max for cycling and 85% VO2Max for running.
  • The C-group: Fed training. C received a carbohydrate-rich breakfast (675 kcal, 70% carbohydrates, 15% fat, 15% protein) ~90 min before each training session. In addition to that, they drank a beverage containing 1g maltodextrin per kg body weight during exercise.
  • The F-group: Fasted training. F received the same “breakfast” as C, and the maltodextrin enriched beverage, but in the mid-afternoon.
  • The CON-group: Control group (no training but the same diet as F and CHO).
  • Diet: 50% fat, 40% carbs and 10% protein. In essence, a diet closely resembling the Standard American Diet (slightly higher fat intake in place of carbs). The diet was also hypercaloric, providing 30% more calories than the subjects required to remain weight stable (range: 3000-4500 kcal). Subjects received supervised lunches, whereas all other meals, snacks and drinks were provided by the investigators as individual take home food packages.

The duration of the study was six weeks. After the study, new tests were perform to study investigate changes in body composition, fitness, diet and health markers.

Fitness Models doing CrossFit
Studies show pictures of hot women increase the likelihood of reading blog posts by 1123%. Irrespective of the subject matter. So if you’re all out of ideas for pictures, that’s a pro tip. I mean it was either this or another lame picture of me running around in the mountains for some sports catalogue. Click the picture to see these women in all their splendor. Then I insist you read on as I report on the (highly interesting) results.


To save some space here, I will not cover the results in the control group. Needless to say, sitting on your ass and overfeeding for six weeks will not result in any favorable changes to any of the parameters covered below (as confirmed by the study). I will only cover the results in the fed (C) and fasted (F) training group and how they compared against each other.

Glucose tolerance and insulin sensitivity: These are basically two sides of the same coin. Glucose tolerance is correlated with insulin sensitivity and vice versa. Both are important health markers when it comes to determining metabolic health and predisposition, or lack thereof, to metabolic syndrome. In summary, F clearly improved glucose tolerance and insulin sensitivity. C did too – but to a much lesser degree than F.

GLUT4: Glucose transporter type 4 is a protein responsible for insulin-regulated glucose transport into the muscle cell. It increased by a whopping 28% in F but only 2-3% in C (not mentioned in the paper but this is my estimate based on the graphs). This partly explains why F saw superior results in regards to glucose tolerance and insulin sensitivity.

Since GLUT4 is triggered by AMPK, which is increased when glucose availability is low, i.e. during fasted training, one would assume the GLUT4 increase could then be explained by an increase in AMPK. This was found to be true: AMPK increased by 25% in F, which correlated closely with the increase in GLUT4 content.

Muscle glycogen and intramuscular lipid stores (IMCL): Much like your muscle mass stores carbs as glycogen, it also acts as a reservoir for fatty acids. From memory, an average adult has approximately 1800 calories (reference forthcoming if I can find it) worth of fatty acids stored in muscle. In contrast to muscle glycogen, which is preferentially used during high intensity activities, IMCL is used during lower intensity activities.

Similar to the study I covered in this article, muscle glycogen stores increased more in F than in C. However, no significant difference was found in IMCL storage.

Metabolic enzymes: Very fittingly, the same group of enzymes that were investigated in the study I covered in “Fasted Training Boosts Endurance and Muscle Glycogen” were looked at here.

Interestingly, changes in citrate synthase and HAD, two markers for fuel-utilization efficiency, were not different between F and C. However, two other important markers for glucose and fat metabolism, FAT/CD36 and CPT1, were increased by ~30% in F. C saw no increase at all.

Body composition: Now this is very interesting. Despite overfeeding the subjects with 1000 calories or more, F only gained 0.7 kg. From a scientific standpoint, this is deemed insignificant. That is, the gain could likely be attributed to chance or, very likely, fluctuations in body weight due to increased muscle glycogen. This lack of weight gain in F could not be explained by the training regimen. They were still overfed by 15-20% when accounting for the extra activity. This left the subjects with a theoretical surplus of 650 calories per day, on average, which should have resulted in weight gain equivalent to ~3.5 kg after six weeks.

What about C? They gained 1.4 kg, twice as much as F, despite doing the same amount of exercise and consuming the same amount of calories and macronutrients.

Exercise capacity: Time to exhaustion increased to a similar magnitude in both groups (+15%). However, C saw a greater increase in VO2Max. On the other hand, “FATmax,” maximal rate of fat oxidation, increased to a much greater degree in F, with no increase in C.

These results are most likely explained both explained by the absence of carbs pre-workout. The greater increase in VO2Max in C is very likely due to the ability to exercise at a greater relative intensity. Maintaining a high intensity is after all dependent on glucose availability, which was abundant in the breakfast-fed and maltodextrin-supplemented group C.

Recall that the fasted group increased VO2Max and peak power output more in the study I covered in “Fasted Training Boosts Endurance and Muscle Glycogen.” What might be the reason for these contrasting results? In the older study, exercise intensity was lower, 65% of VO2Max, an intensity where fat oxidation is maximized. Here subjects were much less dependent on glucose availability to fuel the activity.

In this study, exercise was performed at 75-85% of VO2Max, which greatly increased glucose utilization. When exercising at higher intensities than 65% VO2Max, fat oxidation is progressively reduced and becomes almost non-existent at 82-87% VO2Max.

Simply put, C improved VO2Max more as they could train harder due to providing the proper substrates for fueling the activity. On the other hand, F became progressively more efficient at oxidizing fat at higher levels of intensity as evidenced by the increase in FATmax. This is, in turn, could be explained by the substantial increase in the fat burning enzymes FAT/CD36 and CPT1.

My thoughts

As you can see, the fasted training group beat the fed training group on almost all relevant parameters. More importantly for some perhaps, the fasted training group saw significant improvements in all parameters relevant to improving body composition and health, where as the fed training group saw comparatively lackluster results here.

This study strengthens the theories I outlined here, which is that fasted training may provide some unique benefits to those training in the fasted state, whether it be endurance training, conditioning or bodybuilding. However, this study may be somewhat more relevant considering the higher exercise intensity used, which is closer to, albeit not exactly similar, to CrossFit, kettlebell training and weight training. That is, activities which rely on glucose utilization to a greater degree than fat oxidation.

In addition, I found the lack of weight gain during overfeeding in the fasted group very interesting. Given the improvements in glucose tolerance, insulin sensitivity, metabolic enzymes and maximal fat oxidation rate, it suggests a nutrient partitioning effect that favorably affects how and where nutrients gets stored and burned off.

Carbs may be more likely to be stored as glycogen instead of contributing to fat gain via de novo lipogenesis. Fat that was accumulated in adipose tissue on the day prior may then be more easily released and burnt off during the fasted training session. And so forth.

In summary, a very interesting study, absent any flaws in study methodology, which adds to the mounting scientific evidence speaking in favor of fasted training. It was also fitting to cover it today, as I will be touching on some issues related to the concepts discussed here in my next article.