For adults with obesity who also have prediabetes and fatty liver disease, weight loss may not be the only factor that matters. It can be linked to metabolic improvement as well. In a randomized trial, participants assigned to a ketogenic diet lost about 10% of their starting weight. Their liver fat fell by an average of 67%. About half no longer met prediabetes criteria after the intervention. The results offer a diet comparison, but the trial was small and short.
The study, published online in Cell Metabolism on August 27, 2026, randomly assigned 55 adults with obesity, prediabetes and fatty liver disease to three diets. They were ketogenic, Mediterranean or very-low-fat plant-forward. Forty-two participants completed the study. Researchers supplied all meals and held weekly dietitian consultations. The diets were adjusted to aim for comparable weight loss of about 10% over roughly five months.
The ketogenic group consumed 4% of calories from carbohydrates and 73% from fat. The Mediterranean group had 50% carbohydrates and 35% fat; the plant-forward group had 70% carbohydrates and 15% fat.
In this trial, similar weight loss did not produce identical results across the diets. Keto was linked to a larger average drop in liver fat and more participants falling below the prediabetes threshold. That does not establish that the diet is right for everyone or that the changes will last.
Keto liver fat results compared with other plans
Liver fat fell 67% on average in the ketogenic group, compared with 45% in the other two. Because all three groups lost about the same proportion of their starting weight, the results suggest diet composition may have contributed to the difference. The study compared diets during weight loss; it does not establish how the diets compare without weight loss.
About 50% of participants assigned to keto no longer met prediabetes criteria after the intervention. The figures were 29% for the Mediterranean group and 7% for the plant-forward group. Crossing below the prediabetes threshold is a measured improvement, but it does not prove permanent reversal or rule out later diabetes.
Researchers also tracked blood samples over 24 hours. Glucagon, a hormone involved in glucose and energy metabolism, increased in the keto group and decreased in the other two groups. Insulin levels fell in all three groups, with the largest decrease in the keto group. Despite getting 73% of calories from fat, the keto group showed no greater change than the other groups in LDL cholesterol. The same held for apolipoprotein B and 24-hour plasma triglycerides.
According to study first author Max Petersen, the results show that choice of diet remains important. He said it has an impact on specific health outcomes that go beyond weight loss alone.
The trial did not test keto combined with GLP-1 medication, so it cannot show what that combination would do.
Supplying meals and aiming for comparable weight loss helped researchers compare the diet patterns under controlled conditions. The study’s 55 assigned participants, 42 completers and roughly five-month duration limit what can be concluded about long-term health or safety.
How much did the ketogenic diet reduce liver fat in the 2026 clinical trial?
In a randomized trial, 55 adults with obesity, prediabetes and fatty liver disease were assigned to three diets; 42 completed the study. Liver fat fell by an average of 67% in the ketogenic group and 45% in each of the Mediterranean and very-low-fat plant-forward groups. Participants in all three groups lost about 10% of their starting weight over roughly five months.
The findings describe short-term outcomes in a small, closely managed trial. They do not show whether the differences persist or whether the results apply to people with other health conditions. Keto involves substantial carbohydrate restriction. Anyone considering a major diet change should discuss it with a clinician, especially when managing prediabetes or fatty liver disease.