Dr. Ahmed Farhan
Losing weight is supposed to be the fix. Drop the pounds, and your blood sugar, your liver, your whole metabolism should follow along and get better too.
That’s the standard advice, and it’s not wrong. But a new study asks a sharper question: if two people lose the exact same amount of weight, does it matter how they lost it?
The answer, it turns out, is yes. And the size of the difference surprised even the researchers who ran the study.
Why This Is Trending Now
In late August 2026, a team at Washington University School of Medicine in St. Louis published the results of a head-to-head diet trial in the journal Cell Metabolism.
They took 42 people who all had three things in common: obesity, prediabetes, and a fatty liver. Then they split them into three groups and had each group follow a different popular diet: ketogenic, Mediterranean, or a very low-fat, plant-heavy diet.
Here’s the part that makes this trial unusual. The researchers didn’t just tell people what to eat and hope they’d stick to it. They cooked and supplied every single meal, for months.
Participants met with a dietitian every week. More than 95% of people stuck to their assigned diet the whole way through. That kind of control is rare and expensive, which is part of why this study is getting so much attention.
What the Latest Research Says
All three groups lost weight — about 10% of their starting body weight, on average, roughly 25 pounds for a typical participant.
The researchers built the study this way on purpose, matching weight loss across all three diets, so that any differences they found afterward couldn’t just be explained by “one group lost more weight than the others.”
That’s what makes the next part interesting.
What Researchers Actually Found
Even with weight loss matched, the diets didn’t perform the same. The keto group’s liver fat dropped by about 67%.
The other two groups, Mediterranean and low-fat, saw liver fat drop by around 45%. About half the people in the keto group no longer met the medical criteria for prediabetes by the end of the study. In the Mediterranean and low-fat groups, far fewer people saw their prediabetes go away.
In plain terms: the keto group’s blood sugar control and liver health improved more, even though they’d lost the same amount of weight as everyone else.
What It Means for Obesity, Diabetes, and Metabolic Health
This matters because fatty liver disease and prediabetes are incredibly common and often show up together in people carrying extra weight. An estimated 40% of American adults have obesity, and a large share of them also have some degree of fat buildup in the liver, whether they know it or not.
Fatty liver disease used to be thought of as basically harmless unless it got severe. Doctors now understand it’s an early warning sign tied to future risk of type 2 diabetes, heart disease, and, in some cases, more serious liver damage down the road.
So a diet that clears liver fat faster and reverses prediabetes more often isn’t just a cosmetic win. It’s addressing something with real long-term stakes.
There’s a likely explanation for why cutting carbs specifically helped the liver so much. Your liver doesn’t just receive fat from the food you eat — it can also manufacture fat from carbohydrates, through a process scientists call de novo lipogenesis, which is really just a fancy way of saying “making new fat from scratch.”
When you eat a lot of carbs, especially sugar, your liver ramps up this fat-making machinery. Cutting carbs way down, the way a keto diet does, appears to slow that machinery, which may be a big reason the keto group’s liver fat dropped so much further than the others.
Practical Implications
None of this means the Mediterranean or low-fat diets failed. They didn’t. People in both groups lost real weight and saw real improvements in blood sugar and liver fat too, just to a smaller degree than the keto group.
If your main goal is simply losing weight and improving your health in general, any of these three approaches clearly works.
But if you specifically have prediabetes and a fatty liver, and weight loss alone hasn’t been enough, this study suggests the composition of your diet, not just the number on the scale, might be worth a closer look.
What We Know
- This was a randomized, controlled trial, the strongest type of nutrition research you can run in humans.
- All meals were provided and closely monitored, so adherence was unusually high and reliable.
- Weight loss was deliberately matched across groups, isolating the effect of diet composition itself.
- The keto group showed larger improvements in liver fat and prediabetes remission than the Mediterranean or low-fat groups.
- The findings are published in a respected, peer-reviewed journal, and build on a well-established mechanism — carbohydrate-driven liver fat production — that scientists have studied for years.
What We Don’t Know
- This was a small study — 42 people completed it, out of 55 enrolled. Small studies can turn up real effects, but they need to be confirmed in larger groups before anyone calls this settled science.
- The trial lasted four to five months. Nobody yet knows whether these liver and blood sugar benefits hold up, shrink, or grow over a year, five years, or longer.
- The keto diet used in this study was carefully designed and dietitian-supervised, with specific, pre-made meals. That’s very different from the average person’s version of “going keto” at home, which may look nothing like the study’s diet in food quality or balance.
- The study didn’t report long-term cholesterol or heart-disease outcomes, which is a separate and important question, covered below.
- It’s not yet clear whether these results would look the same in people without fatty liver disease or prediabetes to begin with.
Common Misconceptions
“This proves keto is the best diet for everyone.”
It doesn’t. This study looked at a specific group of people — those with obesity, prediabetes, and fatty liver — over a short period of time.
It doesn’t tell us keto is better for general health, athletic performance, heart health, or long-term weight maintenance in the broader population.
“If keto worked better here, low-carb must always beat Mediterranean.”
Also not quite right. The Mediterranean diet has decades of research behind it showing strong benefits for heart health specifically — benefits this shorter keto trial didn’t measure. Different diets can each be “winning” on different health measures.
“Weight loss doesn’t matter, only diet type matters.”
The opposite is closer to true. Weight loss was the biggest driver of improvement across all three groups. Diet composition looks like it added an extra boost on top of that, particularly for liver fat — not a replacement for losing weight in the first place.
Who Should Be Particularly Concerned
This research is most relevant to you if you have, or suspect you might have, a combination of extra weight, blood sugar that’s higher than normal (prediabetes), and a fatty liver.
Fatty liver disease often has no obvious symptoms, so plenty of people who fit this profile don’t know it yet. If you have obesity along with risk factors like a larger waistline, high triglycerides, or a family history of type 2 diabetes, it may be worth asking a doctor about a simple liver and blood sugar check.
People with existing kidney disease, a history of high LDL cholesterol, or other conditions that make a very low-carb diet riskier should be especially careful before trying this approach, and should only do so with medical guidance, not on their own.
Practical, Evidence-Based Recommendations
- If you’re dealing with prediabetes and a fatty liver, and general weight-loss advice hasn’t moved the needle, this is a reasonable topic to raise with a doctor or registered dietitian — specifically asking whether a lower-carbohydrate approach makes sense for you.
- Cutting back on added sugar and refined carbohydrates, even without going fully keto, still reduces the raw material your liver uses to manufacture new fat. You don’t necessarily need an extreme version of the diet to get some benefit from the same underlying mechanism.
- Because a ketogenic diet can raise LDL (“bad”) cholesterol in some people, anyone trying it — especially for months at a time — should have their cholesterol checked before starting and again a few months in.
- Diet quality still counts. A well-planned keto diet built around vegetables, healthy fats like olive oil, nuts, and fish, and reasonable protein looks very different — and is likely much healthier — than one built mainly around bacon, butter, and processed meat.
- Any major diet change is easier to sustain, and safer, with support from a dietitian, the way participants in this study had weekly check-ins built in.
FAQ
Does this mean I should switch to keto?
Not automatically. This study is strong evidence that low-carb eating may offer extra benefits for people with fatty liver and prediabetes specifically.
It’s not a blanket recommendation for everyone, and any major diet change is worth discussing with a healthcare provider first, especially if you have other health conditions.
Is the Mediterranean diet not as good, then?
Not necessarily. This trial only measured liver fat and blood sugar improvements over a few months. The Mediterranean diet has a much longer track record for heart health, which the keto diet may not match. “Better” depends on what health outcome you’re prioritizing.
How much weight did people lose in this study?
About 10% of their body weight — roughly 25 pounds for a typical participant — and this amount was intentionally similar across all three diet groups.
Is a fatty liver dangerous?
It can become a real problem over time. On its own, early-stage fatty liver often causes no symptoms, but it’s linked to a higher future risk of type 2 diabetes and can, in some people, progress to more serious liver damage. That’s why catching and addressing it earlier is worthwhile.
Can I get these benefits without going full keto?
Possibly, at least partially. Because the underlying mechanism involves your liver converting excess carbohydrate into fat, meaningfully cutting back on added sugar and refined carbs — even without eliminating carbs almost entirely — may offer some of the same benefit, though this specific study didn’t test a “moderate” version.
Does keto raise cholesterol?
It can, in some people, particularly LDL cholesterol. This is a well-documented effect of very low-carb, high-fat diets and is one reason medical guidance and monitoring matter if you try this approach for an extended period.
Conclusion
For years, the standard message has been simple: lose weight, however you do it, and your metabolic health will follow. This study doesn’t overturn that message — weight loss was still the biggest factor at work here — but it adds an important layer underneath it.
For people specifically dealing with a fatty liver and prediabetes, how you lose the weight may matter almost as much as whether you lose it.
That’s a genuinely useful piece of information, even if it comes with real limits: a small study, a short follow-up period, and open questions about long-term heart health that still need answers.
References
1. Petersen, M.C., Smith, G.I., Farabi, S.S., et al. Randomized controlled trial comparing ketogenic, Mediterranean, and low-fat diets in adults with obesity, prediabetes, and hepatic steatosis. Cell Metabolism, August 27, 2026.
2. EurekAlert! / Cell Press. “Keto diet has greater health benefits than other popular weight-loss diets, according to clinical trial.” August 27, 2026.
3. ScienceDaily. “Keto diet cut liver fat by 67% in a clinical trial.” August 28, 2026.
4. HCPLive. “Ketogenic Diet Offers Liver Benefits Beyond Weight Loss Alone.” August 2026.
5. Scientific American. “Is the keto diet better than the Mediterranean? New clinical trial reveals health benefits.” August 2026.
6. Harvard Health Publishing. “Should you try the keto diet?”
7. Softic, S., et al. “Dietary carbohydrates and fatty liver disease: de novo lipogenesis.” Current Opinion in Clinical Nutrition & Metabolic Care.
8. Case report and literature review. “Non-alcoholic Fatty Liver Disease and Carbohydrate Restricted Diets.” PMC.
9. Penn Medicine. “A cardiologist’s take on the keto diet.”
