A recent randomized clinical trial found that a very-low-carbohydrate ketogenic (keto) diet produced larger improvements in liver and metabolic health than a Mediterranean diet or a low-fat, plant-forward diet in people with obesity.
Study design and approach
Researchers enrolled 42 adults with obesity who also had prediabetes and fatty liver disease. Participants were randomly assigned to one of three diet groups and received all their meals from the study team. They met with a dietitian weekly to support adherence. Calorie intake for each person was tailored so participants lost about 10% of their body weight over four to five months.
Macronutrients in the trial
– Keto: extremely low carbohydrate (about 4% of calories in this study) and very high fat (about 73%).
– Mediterranean: higher in carbohydrates (about 50% of calories) and moderate fat (about 35%).
– Plant-forward: highest in carbohydrates (about 70% of calories) and low fat (about 15%).
(For context, common ketogenic prescriptions often aim for roughly 5–10% carbs, 50–60% fat, and the remainder protein, but the study’s keto arm used a very high fat proportion.)
Key findings
– All three diets produced similar weight loss and reduced body fat. Muscle benefits—an important driver of glucose disposal—were comparable across groups and related to weight loss itself.
– Insulin sensitivity improved across the board, but the keto group showed the largest gains in several metabolic markers.
– Liver fat fell substantially more on keto: an average 67% reduction in liver fat compared with roughly 45% in the other two groups.
– About half of the participants assigned to keto no longer met criteria for prediabetes by the study’s end, versus 29% for the Mediterranean group and 7% for the plant-forward group.
– Continuous 24-hour metabolic monitoring showed the keto group had the largest rise in glucagon (a hormone that promotes fat mobilization in the liver) and the greatest drop in circulating insulin. Despite consuming the most fat, participants on the keto plan did not show an increase in blood lipids or cholesterol in this trial.
Interpretation and limitations
Investigators and outside endocrinologists noted the study was small (42 participants) and relatively short (about five months), so results may not generalize to all populations or predict long-term outcomes. The trial’s strength was tight control of food intake and matched weight loss, which helps isolate diet composition effects, but long-term sustainability and real-world adherence differ from controlled feeding studies.
What this means for eating choices
– Weight loss itself remains the dominant driver of many metabolic benefits. In this trial, equal weight loss produced shared improvements (for example, in muscle-related glucose disposal).
– For people with obesity, prediabetes, and fatty liver, a ketogenic eating pattern may provide additional, rapid improvements in liver fat and some metabolic markers compared with higher-carbohydrate plans when weight loss is equivalent.
– Keto diets can be hard to sustain and may not suit everyone. The Mediterranean and plant-forward approaches have a longer evidence base for long-term prevention of diabetes and heart disease and may be more sustainable for many people.
Bottom line
When weight loss is matched, dietary composition still matters for the liver and certain metabolic outcomes. For some patients working with their clinicians, a carefully planned keto approach might offer extra liver and metabolic benefits, but considerations about sustainability, food quality, and individual medical context should guide the choice of diet.
