A keto diet liver fat trial published in the Cell Press journal Cell Metabolism on 27 August reports that a ketogenic diet reduced liver fat by an average of 67% in participants with obesity, prediabetes, and fatty liver disease, compared with 45% in both Mediterranean and low-fat groups, despite all three groups losing roughly the same amount of body weight.
What the keto diet liver fat trial actually tested
The trial enrolled 42 people with obesity, prediabetes, and fatty liver disease and randomly assigned them to one of three diets: ketogenic, Mediterranean, or a low-fat, plant-forward plan. Crucially, the research team supplied all food and participants met a dietitian every week. Calorie intake was individually adjusted so that everyone would lose about 10% of their body weight, which took roughly four to five months. That design matters: by equating weight loss across groups, the researchers tried to isolate any metabolic effects specific to diet composition, rather than to weight reduction alone.
The three diets differed sharply in macronutrient make-up. The ketogenic plan provided just 4% of calories from carbohydrates and 73% from fat. The Mediterranean diet drew 50% of calories from carbohydrates and 35% from fat. The plant-forward diet went furthest in the other direction: 70% of calories from carbohydrates, 15% from fat.
By the end of the study, body fat had fallen and insulin sensitivity had improved across all three groups. So far, so expected. The more pointed findings were in the liver and prediabetes measures.
Liver fat and prediabetes: where the differences emerged
Liver fat in the ketogenic group fell by an average of 67%, against an average of 45% in both the Mediterranean and plant-forward groups. The researchers tracked participants’ metabolism over a 24-hour period by repeatedly analysing their blood. People following keto showed the largest rise in glucagon, a hormone that helps reduce fat stored in the liver, and the greatest decline in blood insulin levels, another change associated with lower liver fat.
The prediabetes results were also uneven across groups. About 50% of keto participants no longer met the criteria for prediabetes after the intervention. That compared with 29% of those on the Mediterranean diet and just 7% of those following the plant-forward diet.
Samuel Klein, the Danforth Professor of Medicine and Nutritional Science at Washington University School of Medicine and the study’s corresponding author, offered a measured interpretation. ‘We know weight loss can improve metabolic health in people who are obese,’ he says. ‘Our data suggest that if you have high liver fat content and prediabetes, reducing carbohydrate intake can have important therapeutic effects on metabolism beyond just weight loss alone.’
Max Petersen, the study’s first author, also at Washington University School of Medicine, flagged an unexpected finding on blood lipids: ‘The most surprising finding was that participants in the ketogenic diet group didn’t experience an increase in blood fat or cholesterol levels despite consuming the most fat.’ That is worth noting because elevated blood fats are a concern sometimes raised about high-fat dietary patterns.
Petersen also drew a connection to drug-based approaches: ‘GLP-1-based medications have revolutionized the therapy of obesity. We now have a very effective pharmacotherapy for achieving significant weight loss. But in addition, our study shows that the specific composition of the diet can give you added benefits.’
What the evidence does and does not support
Before the results are taken further than the data allow, the sample size is worth keeping front of mind. Forty-two participants across three diet groups is a small trial. The study was also conducted under controlled conditions (food was provided, dietitian contact was weekly) which is good for internal validity but limits how directly the findings transfer to everyday eating. The trial was supported by the National Institutes of Health and the Foundation for Barnes-Jewish Hospital.
The trial was also designed around a specific population: people with obesity, prediabetes, and fatty liver disease. Klein’s framing reflects that focus. The claim is not that keto is better for everyone, but that for people with high liver fat and prediabetes, cutting carbohydrates may carry therapeutic effects beyond what weight loss alone achieves.
The study has been published in Cell Metabolism and is available with full methodology, which is where any clinician or patient weighing these findings should begin.
