Sugar and fructose: the stage is your liver

You had blood drawn. Routine, nothing dramatic. A few days later there is a number on the GGT line that sits just above the limit. Maybe on ALT as well.

And then comes the question that almost always comes: do you drink a lot of alcohol?

You say no, and you mean it. A glass of wine at the weekend, that is all. The number is still there, and nobody quite knows what to do with it.

Many people know this moment. In my consulting room I hear it almost every week. And almost every time we then talk about something that rarely comes up in a conversation about liver values: sugar. More precisely, the half of it that is called fructose.

Two halves, two routes: what sits inside a spoon of sugar

A teaspoon of sugar looks like one thing. One white crystal, one taste, one calorie count. Biochemically it is two.

Table sugar is sucrose, a double molecule. Your small intestine splits it, and out come two very different sugars: half glucose, half fructose. From that point on their paths separate.

Glucose can be taken up and burned by almost every cell in your body. Muscle, brain, red blood cells, all of them have transporters for it. Insulin manages the distribution. That is the curve you see on a glucose sensor, and I write about it in detail elsewhere, for example in Avoiding blood sugar spikes.

Fructose does something else. It is taken up through its own transporter and processed first in the small intestine. Whatever is not handled there travels through the portal vein straight to the liver. Other organs barely take part.

This is exactly why fructose is almost invisible on a glucose sensor. A CGM shows you the glucose half of a meal, not the fructose half. If you measure for 14 days, as described in the guide 14 days with a glucose sensor, you get a useful map. Just one with a stretch of land missing.

And now you know why two people can eat the same amount of sugar and their measuring devices still tell different stories.

The bottleneck that fructose walks around

Why can your body throttle glucose and not fructose? This one question explains almost everything that follows in this article.

Picture a motorway. The regular slip road has a traffic light. When the carriageway is full, it turns red. That traffic light genuinely exists, it is called phosphofructokinase and it sits in the middle of glycolysis. When ATP and citrate rise, meaning the energy status of the cell rises, the enzyme throttles the flow.

Fructose takes a different slip road. Ketohexokinase turns it into fructose-1-phosphate very quickly, and aldolase B splits that into fragments that only join the road below the traffic light. So the feedback loop never engages. The liver lacks the lever it would need to say: slower please, I have had enough.

Fructose-1-phosphate is not merely an intermediate here. Mechanistic reviews describe it as a signalling molecule that switches on lipogenic programmes in the liver cell as well. Substrate and start command come from the same source.

A second lens, mechanistically plausible and still thin in humans: fructose that the small intestine does not absorb is fermented by the microbiota living there, among other things into ethanol and acetate. Both are building blocks for fat synthesis. A 2025 review also describes marked differences in that microbiota between population groups. That could be one trail towards why the same amount of sugar lands differently in two people.

And now you know why the liver cannot simply wave fructose away, the way it can with glucose.

ATP, uric acid and gout: the side road nobody talks about

You may know someone whose gout attack arrived after a barbecue evening. Beer, meat, the usual explanation. The lemonade on the same table rarely gets mentioned.

The link begins at the first step of fructolysis. Ketohexokinase attaches phosphate to fructose, so quickly that phosphate can briefly run short inside the liver cell. ATP drops, AMP rises, and AMP deaminase moves the surplus onward. At the end of that chain sits uric acid.

This is not a blackboard sketch. It is measurable in humans.

Supported by large cohorts: the Health Professionals Follow-up Study tracked 46,393 men for twelve years. At two or more servings of sweetened drinks a day, the relative gout risk sat at 1.85. In the Nurses Health Study with 78,906 women over 22 years it was 2.39, and 2.42 for orange juice. Diet drinks showed no association in either.

Two things belong to honesty here. First, the authors name the absolute numbers themselves: in women, 36 to 68 additional cases per 100,000 person years. Relevant, but not an earthquake. Second, the male cohort also showed a signal for fructose-rich whole fruit. I am leaving that in.

And then there is the counter test.

And now you know why a question about drinks belongs in a gout work-up, not only a question about meat.

Sugar becomes liver fat: de novo lipogenesis in numbers

Fat from sugar sounds at first like an arithmetic slip. Sugar is carbohydrate, fat is fat, surely two separate categories.

Inside the liver that boundary blurs. De novo lipogenesis is the name for building new fatty acids out of carbohydrate carbon. In everyday life this route runs on a low flame. It can be turned up, though, and fructose appears to be the stronger trigger.

That also explains why a lipid profile can shift under a lot of sugar without total cholesterol moving much. What rises are apoB, small dense LDL particles and remnants. Why those measures are more interesting than the single number on your report is covered in Cholesterol: the myth and the science.

Mechanistically intriguing, animal data: in a study in mice by Miguel Lanaspa, the liver built its own fructose from surplus glucose through the polyol pathway. Animals without fructokinase were protected from fatty liver and weight gain, even though they only received glucose solution. So the fructose route may not only be a question of what is on the plate. In humans this has not been shown that clearly.

And now you know why someone at a normal body weight can still have a fatty liver.

MASLD: why the liver stays quiet for so long

Liver tissue itself has no pain fibres. Only the capsule on the outside can pull when the organ swells. That is why silence is not a good sign here. It is the normal state.

A word on the name first. What was long called non-alcoholic fatty liver disease, NAFLD, has since 2023 and 2024 been called MASLD: metabolic dysfunction associated steatotic liver disease. NASH became MASH. The German-speaking specialist society adopted this in an amendment to its guideline. The new name says what the condition is about, instead of only what it is not.

In type 2 diabetes the number is considerably higher. A meta-analysis of 123 studies found a MASLD prevalence of 65.33 percent, and among those biopsied, 40.78 percent had relevant fibrosis. Liver fat and insulin resistance are closely linked, in both directions. The hormonal side is covered in Insulin resistance and losing weight.

For Germany there is no robust measured figure I could quote here. A modelling exercise for eight countries, Germany among them, expects a moderate rise in cases by 2030, but a steeper rise in the inflammatory form. Modelling means assumptions, not measurement.

One point of separation: this is about fat in the liver, not about detoxification. Those are two different building sites. What is documented on the detox topic and where marketing begins is something you can read in Liver detox: what counts and what is marketing.

And now you know why the phrase no symptoms says so little about this particular organ.

The dose question: isocaloric versus hypercaloric

Now comes the part where honest articles get uncomfortable. Because when you sort the evidence cleanly, it gets in its own way.

There are two ways to study fructose. Either you swap it in for other carbohydrates at the same calorie count, which is called isocaloric. Or you add it on top, which is hypercaloric. The two questions give different answers. That is exactly where the dispute sits.

Read only the first row and you conclude it is all a calorie question. Read only the last two and you conclude the opposite. Both sit in the same body of literature.

My reading, flagged openly as an interpretation: at group level and in short swap studies, energy balance dominates. On that point classical nutritional medicine is right, and that deserves to be acknowledged. At the same time, tracer studies show a fructose-specific synthesis route very consistently. The contradiction stands, and I see no value in smoothing it over.

One argument speaks particularly clearly for a route of its own. You can block the first enzyme of fructolysis and see what happens.

And the amount? In the mouse study mentioned above, the threshold at which the small intestine overflows sat at around one gram per kilogram of body weight. That is an animal finding, not a recommendation. For humans there is no cleanly mapped tipping point, and I am not going to invent one here. What remains as orientation is the target from the guidelines: free sugars below ten percent of total energy, so roughly 50 grams at around 2,000 kilocalories.

That energy balance and food quality are not opposites but two levels of the same question is covered in The calorie myth: quality over quantity and in Calorie deficit: why it holds up and still is not enough.

And now you know why two serious experts can say different things about the same molecule without either of them working sloppily.

Fruit is not a soft drink: matrix, speed and satiety

At this point people regularly write to me along these lines: so I am not allowed to eat bananas any more.

You are. Please eat fruit.

I am not saying that out of politeness, but because the data point that way. The same fructose behaves differently depending on what it sits inside.

Four differences explain that rather well.

The WHO draws its line at the same edge. Its recommendation refers to free sugars, meaning added sugars plus sugar from honey, syrup and fruit juice. Sugar inside whole fruit explicitly does not count.

In clinical practice I observe that many people with bloating after fruit do not have a disease but a quantity question. Fructose malabsorption is common: after 35 grams of fructose, 24 percent of healthy control participants showed a clear signal on the breath test, and 33 percent of those with functional complaints. The test is imprecise, and depending on the scoring method the classification changed in up to 49 percent. That is something different from the rare inherited fructose intolerance.

And now you know why an apple and a glass of apple juice are two different biochemical events, even though the label names the same sugar.

Sweetened drinks and sweeteners: two questions, two very different answers

Diet version, then. That is how the obvious way out sounds, and it is understandable.

Except these are two separate questions. For sugar-sweetened drinks the evidence is comparatively clear. For non-sugar sweeteners it is the opposite.

The clearer half: sugar-sweetened drinks

Supported by large observational data: in the Framingham cohorts, the odds of a fatty liver on CT rose to 1.61-fold at one serving a day or more, even after adjusting for BMI. Diet drinks showed no association there. In a meta-analysis of 17 cohorts with 38,253 cases of diabetes, incidence rose by 18 percent per daily serving, and still by 13 percent after adjustment for adiposity.

The murkier half: sweeteners and sugar alcohols

In May 2023 the WHO published a guideline on non-sugar sweeteners. It conditionally advises against using them for weight control or for lowering risk. That headline travelled everywhere. Four points from the small print almost never did.

What do the studies themselves say? A meta-analysis of seven randomised trials with 1,003 participants found no significant BMI effect. In 30 cohorts covering more than 400,000 people, unfavourable associations did appear, for instance with weight, waist circumference and type 2 diabetes. The most obvious explanation is reverse causation: people who are gaining weight reach for diet products, rather than the other way round.

And then there is the erythritol discussion. In cohorts drawn from cardiology work-ups, one of them at the Charité in Berlin, the risk of major cardiovascular events within three years was higher in the top erythritol quartile, with hazard ratios of 1.80 and 2.21. Alongside came raised platelet reactivity in the lab, stronger thrombus formation in the animal model, and high plasma levels for more than two days after a single drink. For xylitol the same researchers found a similar picture.

And now you know why a headline about the WHO guideline almost always falls short.

Liver values: what you can have measured and what the numbers miss

When an organ stays quiet this long, you need something that speaks for it. That is what the numbers from the opening of this article are about.

The three usual values measure different things. ALT, formerly GPT, is the most liver-specific. AST, formerly GOT, also comes from muscle, so exercise can move it. GGT reacts sensitively, but to a great many things: alcohol, medicines, bile flow. Add to that ultrasound, in studies MRI with fat quantification, and scores such as the Fatty Liver Index built from waist circumference, BMI, triglycerides and GGT.

The encouraging news sits in two intervention studies, both without calorie restriction.

A second lever alongside sugar is movement, apparently independent of weight as well. A meta-analysis of 17 studies with 373 people found a fall in liver fat of 3.31 percentage points after one to 24 weeks of training, and 2.16 percentage points even without relevant weight loss. What happens in fat metabolism during that is covered in Understanding zone 2 training.

How the detoxification phases of the liver work, the other building site of the same organ, is covered in Phase 1 and phase 2 of liver detoxification.

If you would like not only to read but to have your own values placed in context: below this article you will find the option to book an appointment.

And now you know why, with a slightly raised GGT, I do not ask about wine but about the drinks in between.

Common questions about fructose, sugar and the liver

These are the questions that reach me most often on this topic, in the consulting room and by email.

Sugar, the liver and the rest of your body

Liver fat does not stand alone. It hangs on hormones, on silent inflammation, on movement and on what a sensor shows or does not show. Several paths lead onward from here.