The American Journal of Clinical Nutrition published a study last week to investigate claims that ketogenic diet can facilitate greater fat loss than diets relatively higher in carbohydrate.

Commonly known as “keto”, a ketogenic diet is a diet typically characterized by a 4:1 ratio of dietary fat to protein and carbohydrate and was originally used in the treatment of childhood epilepsy. It has been theorized that keto diets facilitate greater fat loss in humans as an absence of dietary carbohydrate forces the body to oxidize fat as its primary energy source.

Taubes and colleagues set out to test this theory. Their study was conducted over eight weeks in a research facility known as a metabolic ward. Seventeen overweight or obese volunteers participated, with no opportunity to eat foods outside of the diets of the study.

For the first four weeks, the subjects were fed a high-carbohydrate, high-sugar diet. Fifty percent of their total calories came from carbohydrate (338g per day), and 25% of their total calories came from sugar. The HCD diet totaled 2,739 calories per day.

For the second four weeks, they were fed a very-low-carbohydrate, low-sugar ketogenic diet. Five percent of their total calories came from carbohydrate (36g per day), and 2% of their total calories came from sugar. 15% percent of the calories came from protein. The keto diet totaled 2,738 calories per day. Note that the caloric intakes were kept close to identical, meaning fat loss could only be attributed to the source of the food rather than its caloric content.

The volunteers spent two days a week inside metabolic chambers, where their calorie expenditure was measured. Once every two weeks, their body composition was measured via a DEXA scan. The researchers also used doubly labeled water to measure the volunteers’ average calorie expenditure during the final two weeks of each diet.

In a fantastic online review of the study, Dr. Stephan Guyenet, a nutritional research expert, notes the study’s thoroughness and that the results, at face value, support the researchers’ initial hypothesis that a ketogenic diet promoted greater fat loss. Compared with the higher carbohydrate diet, the keto diet coincided with increased energy expenditure, meaning the subjects appeared to burn more calories when their carbohydrate levels were cut, thought they were consuming the same amount of calories as they were on the high carbohydrate diet.

But let’s look a little deeper at that. Though the subjects’ energy expenditure increased in the first two weeks of ketogenic diet, this increase eventually plateaued in the final two weeks and fell back down to the study’s baseline. What’s more, though participants lost weight rapidly in the first few days of going keto as a result of decreased water retention, body fat loss eventually also slowed. Further testing also returned that the participants on the keto diet even started to break down lean tissue for energy – not body fat.

Additionally, as also noted by Guyenet, fat mass decreased at a slower rate on the ketogenic diet than on the higher carbohydrate diet. On the ketogenic diet, the volunteers lost the same amount of body fat in one month that they lost in two weeks on the high carbohydrate diet. Guyenet comments:

If you believe the hypothesis that sugar summons Beelzebub to plump up your fat tissue regardless of your calorie intake, the comparison should have been extremely favorable to the ketogenic diet. Yet the effect on fat mass was the opposite of what this hypothesis predicts.”

I’m biased, as I don’t believe that a ketogenic diet is an effective and uncomplicated way to lose weight in the long term. So it’s worth finishing with the point that for many trusted experts, including Guyenet, the effectiveness of the ketogenic diet lies in how satiating fat is. Fat makes us feel fuller, which commonly leads to a cumulatively lower net calorie intake and therefore, weight loss. This study removed the participants’ opportunity to eat according to their subjective hunger levels and thus the only feather in keto’s cap.

As one user on Guyenet’s blog notes, what would be interesting is a look at any data that tracked subjective hunger levels in this study, or a new study with a method that allowed for and controlled a more intuitive consumption of food. But this particular study lends little to no credence to keto.