A ketogenic diet produced the largest reductions in liver fat and the greatest improvements in liver insulin sensitivity among three weight loss diets examined in a small clinical trial.

The research, published in the journal Cell Metabolism, involved 42 adults with obesity, prediabetes and excess liver fat. Participants were randomly assigned to a low carbohydrate ketogenic diet, a Mediterranean diet or a low fat, mostly plant based diet.

The diets differed significantly in their proportions of carbohydrates and fat. About 4% of calories in the keto diet came from carbohydrates, while 73% came from fat.

The Mediterranean diet provided about 50% of calories from carbohydrates and 35% from fat. Meanwhile, about 70% of calories in the low fat diet came from carbohydrates and 15% came from fat.

Here's What They're Not Telling You About Your Retirement

Participants lost weight in all three groups, and muscle insulin sensitivity improved by about 50% across the diets. That improvement helped muscles take up blood sugar more efficiently.

However, the keto group experienced larger changes in several other metabolic measures. Liver insulin sensitivity improved two to three times as much among participants following keto as it did in the other two groups.

Liver fat also fell by roughly 67% in the keto group, compared with about 45% among people following the Mediterranean and low fat diets. Researchers linked the keto group’s reduction to a decrease in the liver’s production of new fat.

Measures of glucose and insulin generally improved more with keto as well. By the end of the trial, about 50% of participants in the keto group no longer met the criteria for prediabetes.

This Could Be the Most Important Video Gun Owners Watch All Year

That result compared with 29% of participants assigned to the Mediterranean diet and 7% of those following the low fat diet. The findings suggested that diet composition may influence certain metabolic outcomes beyond the effects of weight loss alone.

However, the researchers cautioned that the short duration and small sample size limit how broadly the findings can be applied. Each diet group included about 14 people.

Registered dietitian nutritionist Ilana Muhlstein said the results highlighted differences that can emerge even when participants achieve similar weight loss. "I love more and more validation and recognition that calories are not equal," the California based expert told Fox News Digital.

Muhlstein said weight loss is often reduced to "just a calorie deficit," but the trial associated different proportions of macronutrients with different changes in health markers. The keto group experienced a larger reduction in the liver’s production of new fat, along with metabolic changes consistent with greater fat burning.

She explained that excess calories and metabolic dysfunction can lead to fat accumulating in the liver, including among people who do not drink alcohol. Muhlstein recommended that anyone with a fatty liver consider transitioning to a lower carbohydrate diet.

Reducing carbohydrate intake may encourage the body to use more stored fat for energy, according to Muhlstein. She also noted that participants following keto did not have higher cholesterol or blood fat measurements during the short study.

Still, Muhlstein urged caution for some people. "This is a really interesting study for someone who has prediabetes, someone who had blood sugar irregularity, but maybe still something to be cautious of if someone has heart disease in the family or is at high risk of heart disease," she added.

Overall, Muhlstein said keto may be worth discussing as a short term option for some people with obesity or prediabetes. She does not recommend it over the long term because of its high saturated fat content and tendency to translate into a diet lower in fiber, which she said is not good for heart health.

After significant weight loss, Muhlstein recommends transitioning toward a Mediterranean diet for long term overall health. Although keto produced the greatest improvements in liver fat and liver insulin sensitivity in this trial, the study’s size and duration remain important limitations.