Muscle responded to insulin about 50% better on keto, Mediterranean, and very-low-fat. The liver did not improve the same amount on all three.

Petersen, Klein, and colleagues at Washington University studied adults with obesity, prediabetes, and fatty liver. Cell Metabolism, 2026. Forty-two people finished, 14 per group. All meals were provided. People ate more than 95% of the study food. Each group lost about 10% of body weight over roughly 5 months.

Muscle took up sugar better after insulin. That improved about 50% in every group. The diets did not differ on that measure. Lean mass loss was similar.

The liver improved two to three times more on keto. Fat in the liver on MRI fell from 19.4% to 6.2% on keto, about 67%, versus about 45% on Mediterranean and very-low-fat. The liver also made less new fat from carbohydrate on keto. Average blood sugar marker HbA1c fell from 5.6% to 5.0% on keto. Fasting triglycerides fell 82 mg/dL on keto versus 41 and 24 on the other diets. Prediabetes went away in 50% on keto, 29% on Mediterranean, and 7% on very-low-fat. Those are small numbers.

LDL cholesterol, ApoB, and day-long triglycerides did not differ by diet, even though keto had 23% of calories from saturated fat. I still check those labs. In this trial they did not rise more on keto than on the other diets.

All food was packed and delivered. The study was small. The authors cannot say a milder carb cut without ketosis would do the same, or that these liver changes hold if someone only starts keto after the weight is already off.

Weight loss still helped muscle insulin response on every diet. After weight loss was matched, keto still did more for liver fat and liver insulin response.

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. 2026;38:1-12. https://doi.org/10.1016/j.cmet.2026.07.020