A small randomized trial found that matched weight loss on a ketogenic diet reduced liver fat more than weight loss on Mediterranean and very-low-fat diets. The result is intriguing—but it is not a universal verdict on the best way to eat.
A ketogenic diet cut liver fat by an average of 67% in a small randomized trial of adults with obesity, prediabetes and excess fat in the liver. Mediterranean and very-low-fat diets also produced substantial improvements, underscoring that weight loss itself did much of the work.
Researchers at Washington University School of Medicine randomly assigned 55 adults to one of three eating plans: a very-low-carbohydrate ketogenic diet, a Mediterranean diet or a very-low-fat plant-forward diet. Forty-two participants completed the intervention, leaving 14 people in each group for the final analysis.
The study was designed around an unusually useful comparison. Every participant received all of their food, met with a dietitian each week and continued until losing approximately 10% of starting body weight. By matching the amount of weight loss, researchers could examine whether the macronutrient mix changed the metabolic response.
All three diets improved several important measures. Muscle insulin sensitivity rose by about 50% across the groups, and body fat declined. The ketogenic group, however, recorded the largest reductions in liver fat, liver glucose production, 24-hour glucose and insulin exposure, and glycated hemoglobin.
- The trial enrolled adults with obesity, prediabetes and fatty liver; 42 people completed it, with 14 in each diet group.
- Participants received all meals and lost approximately 10% of their starting weight over roughly five months.
- Liver fat fell by an average of 67% in the ketogenic group and 45% in each of the other two groups.
- Muscle insulin sensitivity improved by about 50% across all three diets, suggesting that weight loss was the main driver of that benefit.
- The trial was small, short and tightly supervised, so it cannot establish long-term superiority or real-world sustainability.
How a Ketogenic Diet Cut Liver Fat in the Trial
The ketogenic plan supplied approximately 4% of calories from carbohydrate and 73% from fat. The Mediterranean diet supplied about 50% from carbohydrate and 35% from fat, while the plant-forward diet supplied roughly 70% from carbohydrate and 15% from fat.
Those are not minor variations. They represent three fundamentally different fuel environments. The very-low-carbohydrate plan lowered glucose and insulin exposure and increased reliance on fat and ketones. Researchers found that the liver became more responsive to insulin and reduced its production of new fat more sharply in that group.
After matched 10% weight loss, liver fat declined by 67% on the ketogenic diet, compared with 45% on both the Mediterranean and very-low-fat diets. Hepatic insulin sensitivity improved two to three times more in the ketogenic group.
The distinction matters because fat stored inside the liver is not simply a cosmetic measure. It is closely connected to how the liver regulates glucose and responds to insulin. Still, the study measured a specialized clinical population—not healthy athletes, recreational exercisers or the general public.
Weight Loss Helped Every Group
The headline result should not obscure the trial’s broadest finding: losing a meaningful amount of weight improved metabolic health regardless of which diet produced it.
All three groups lost about 10% of body weight. Muscle insulin sensitivity improved by approximately 50% in every group, and no diet produced faster weight loss. Those similarities are important because they separate the effects of diet composition from the effects of becoming lighter and carrying less body fat.
Independent experts reviewing the paper emphasized this point. The ketogenic diet created additional liver-specific changes, but most of the shared health improvement appeared to come from weight loss itself.
That interpretation also fits a larger practical truth. A theoretically impressive diet does little if a person cannot follow it safely and consistently. In the study, meals were supplied and adherence was supported by weekly dietitian visits. Everyday life rarely provides that level of structure.
What Else Improved on the Ketogenic Diet
The ketogenic group experienced the largest declines in 24-hour blood glucose and insulin. Average 24-hour glucose fell by about 20% from baseline, compared with roughly 8% on the other diets. The university’s report also noted that about half of the ketogenic group no longer met the study’s definition of prediabetes at the end of the intervention, compared with 29% in the Mediterranean group and 7% in the very-low-fat group.
Those percentages sound dramatic, but the group sizes were tiny. With only 14 completers in each arm, a difference of a few people can move a percentage substantially. The result deserves further study; it should not be treated as a guaranteed individual outcome.
The primary paper also reported larger reductions in glycated hemoglobin, liver fat production and day-long insulin exposure on the very-low-carbohydrate plan. These findings help explain why researchers concluded that macronutrient composition can influence the metabolic pathway through which the same amount of weight loss is achieved.
The trial suggests that diet composition can change the metabolic route to the same weight loss. It does not crown a universal best diet.
Fitness Living Magazine™ Editorial Analysis
Where Keto Did Not Clearly Win
The ketogenic diet did not outperform the other approaches on every measure. The researchers reported no significant between-group differences in LDL cholesterol, apolipoprotein B or total 24-hour triglyceride exposure.
That is a crucial counterweight to the most enthusiastic interpretations. Lower fasting triglycerides can look favorable, but one expert review noted that higher post-meal triglycerides offset that advantage across the full day. Overall glucose tolerance after an oral glucose challenge was also comparable among the diets.
The Mediterranean group showed numerically larger reductions in several cardiovascular measures, including LDL cholesterol, apolipoprotein B and systolic blood pressure, but the study was too small to determine whether those differences were real rather than chance.
Long-term cardiovascular outcomes were not measured. Neither were exercise performance, strength, muscle retention, hunger over years, food enjoyment, cost or the likelihood that participants would continue the diets without supplied meals.
Why Experts Are Cautious
Researchers conducted a carefully controlled randomized trial, which makes the comparison stronger than an observational survey. Its control also creates the main limitation: the conditions were far removed from ordinary eating.
Only 42 people completed the study, the intervention lasted roughly five months and every meal was provided. Participants met with a dietitian weekly. Experts described the findings as promising and hypothesis-generating, while calling for larger studies lasting at least a year.
Sustainability is especially important with a diet that limits carbohydrate to such an extreme degree. A clinical ketogenic diet is not simply “eating fewer sweets.” It can sharply restrict grains, legumes, many fruits and other foods that are common in established healthy eating patterns.
The Mediterranean diet, by contrast, has a much larger body of long-term cardiovascular evidence. This new trial answers a focused question about matched short-term weight loss and liver metabolism; it does not erase that broader evidence base.
What This Means for Fitness and Body Composition
For readers focused on fitness, the study offers useful context rather than a new universal program. It shows that two people can lose the same percentage of body weight while experiencing somewhat different metabolic changes depending on how the diet is structured.
It does not show that ketogenic eating improves training performance. The participants were selected for obesity, prediabetes and liver fat, and the trial did not test athletes, lifting progress, endurance, recovery or competition outcomes. Carbohydrate availability can matter during high-intensity training, so a liver-health result should not be casually converted into a performance claim.
During intentional weight loss, the foundations remain familiar: an achievable calorie deficit, adequate protein, progressive resistance training, recovery and a dietary pattern a person can maintain. Our recent report on once-weekly interval walking and body fat made a related point—an intervention works through its actual dose and structure, not through a simplified headline.
People choosing between dietary approaches should also consider medical history, food preferences, cultural fit, budget and the kind of exercise they perform. A plan that improves laboratory measures for five months may still be the wrong long-term fit for a particular person.
People who use glucose-lowering or blood-pressure medication, are pregnant, have kidney, liver, gallbladder or pancreatic disease, have a history of disordered eating, or are managing abnormal blood lipids should not begin a ketogenic diet without guidance from a qualified clinician. Medication needs can change quickly when carbohydrate intake falls.
How to Read the 67% Headline
The most accurate interpretation is narrow: among adults with obesity, prediabetes and fatty liver who received every meal and lost approximately 10% of body weight, the ketogenic diet reduced liver fat more than two higher-carbohydrate diets over about five months.
The trial does not prove that everyone should eat keto, that carbohydrates inherently cause fatty liver, or that Mediterranean and plant-forward diets failed. Both comparison diets reduced liver fat by 45%, and all three improved major metabolic measures.
It also does not prove that the result will persist. Weight regain, adherence and changing food quality can alter outcomes after a controlled intervention ends. Longer follow-up will be necessary to determine whether the liver advantage remains and whether any benefits are offset by other risks.
For now, the study is best understood as evidence that the composition of a weight-loss diet may matter for specific organs and metabolic pathways—even when the amount of weight lost is the same.
The Fitness Living Takeaway
A ketogenic diet cut liver fat more sharply than Mediterranean and very-low-fat diets in this small, highly controlled trial. The average reduction was 67%, compared with 45% in the other two groups, after all participants lost about 10% of body weight.
The more durable message is broader. Weight loss improved metabolic health across every diet, while the ketogenic approach appeared to add a short-term, liver-specific benefit. Whether that advantage is sustainable, safe and meaningful over years remains unknown.
Readers should treat this as a promising clinical finding—not a license to self-prescribe an extreme diet or a declaration that one eating pattern is best for every goal.
Primary research: Petersen MC, Smith GI, Farabi SS, et al. Effect of diet macronutrient content on the cardiometabolic response to weight loss: A randomized clinical trial. Cell Metabolism, published online August 27, 2026. View the PubMed record.
Washington University School of Medicine: Trial summary, diet protocols and reported liver-fat changes. Read the university report.
Independent expert reaction: Science Media Centre comments on the trial’s strengths, limitations and long-term uncertainties. Read the expert analysis.
Image credit: Louis Hansel / Unsplash.
Fitness Living Magazine™ reports on exercise and health research for general informational purposes. This article is not individualized medical advice.
